This International Student Edition is for use outside of the U.S.
tH
qe
Operations Management
in the Supply Chain
DECISIONS AND CASES
Hill
“
Roger Schroeder | Susan Meyer Goldstein
Digitized by the Internet Archive
in 2024
https://archive.org/details/operationsmanage0000schr_t600
Operations
Management in the
Supply Chain
Decisions and Cases
Eighth Edition
Roger G. Schroeder
Susan Meyer Goldstein
Carlson School of Management
University of Minnesota
°o
°
°
)
OPERATIONS MANAGEMENT
°
°
°
°
20
so
°
ore =
o
6°98
°
°
°
o
ag
°
°
°
re)
2
°
°
°
°
o
©
tee
°
)
=
IN THE SUPPLY CHAIN
Published by McGraw-Hill Education, 2 Penn Plaza, New York, NY 10121. Copyright © 2021 by McGraw-Hill
Education. All rights reserved. Printed in the United States of America. No part of this publication may be
reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the
prior written consent of McGraw-Hill Education, including, but not limited to, in any network or other electronic
storage or transmission, or broadcast for distance learning.
Some ancillaries, including electronic and print components, may not be available to customers outside the
United States.
This book is printed on acid-free paper.
123456789
LWI 24 23 22 21 20
ISBN 978-1-260-57143-1
MHID 1-260-57143-2
Cover Image: ©Shutterstock/Ekaphon maneechot
All credits appearing on page or at the end of the book are considered to be an extension of the copyright page.
The Internet addresses listed in the text were accurate at the time of publication. The inclusion of a website does
not indicate an endorsement by the authors or McGraw-Hill Education, and McGraw-Hill Education does not
guarantee the accuracy of the information presented at these sites.
mheducation.com/highered
°o
Mid
Aes
te)
°
°
.
The McGraw-Hill Education Series
Operations and Decision Sciences
SUPPLY CHAIN MANAGEMENT
BUSINESS RESEARCH METHODS
Benton
Purchasing and Supply Chain
Management
Third Edition
Schindler
Slater and Wittry
Business Research Methods
Thirteenth Edition
Practical Business Math Procedures
Thirteenth Edition
Bowersox, Closs, Cooper, and Bowersox
Supply Chain Logistics Management
Fifth Edition
Burt, Petcavage, and Pinkerton
BUSINESS FORECASTING
Keating and Wilson
Forecasting and Predictive Analytics
Seventh Edition
Supply Management
Eighth Edition
LINEAR STATISTICS AND REGRESSION
Johnson
Purchasing and Supply Management
Sixteenth Edition
Applied Linear Regression Models
Kutner, Nachtsheim, and Neter
Simchi-Levi, Kaminsky, and
Simchi-Levi
Designing and Managing the Supply
Chain: Concepts, Strategies, Case
Studies
Third Edition
Stock and Manrodt
Fundamentals of Supply Chain
Management
PROJECT MANAGEMENT
Brown and Hyer
Managing Projects:
Approach
A Team-Based
Larson
Project Management: The Managerial
Process
Eighth Edition
SERVICE OPERATIONS
MANAGEMENT
Bordoloi, Fitzsimmons, and Fitzsimmons
Service Management: Operations,
Strategy, Information Technology
Ninth Edition
MANAGEMENT
SCIENCE
Hillier and Hillier
Introduction to Management Science:
A Modeling and Case Studies Approach
with Spreadsheets
Sixth Edition
Fourth Edition
BUSINESS SYSTEMS DYNAMICS
Sterman
Business Dynamics: Systems Thinking
and Modeling for a Complex World
BUSINESS MATH
Slater and Wittry
Math for Business and Finance: An
Algebraic Approach
Second Edition
BUSINESS STATISTICS
Bowerman, Drougas, Duckworth,
Froelich, Hummel, Moninger,
and Schur
Business Statistics in Practice
Ninth Edition
Doane and Seward
Applied Statistics in Business and
Economics
Sixth Edition
OPERATIONS MANAGEMENT
Doane and Seward
Cachon and Terwiesch
Second Edition
Essential Statistics in Business and
Economics
Third Edition
Cachon and Terwiesch
Matching Supply with Demand: An
Introduction to Operations Management
Fourth Edition
Basic Statistics for Business and
Economics
Ninth Edition
Operations Management
Jacobs and Chase
Operations and Supply Chain Management
Sixteenth Edition
Jacobs and Chase
Operations and Supply Chain
Management: The Core
Fifth Edition
Schroeder and Goldstein
Operations Management in the Supply
Chain: Decisions and Cases
Eighth Edition
Stevenson
Operations Management
Fourteenth Edition
Swink, Melnyk, and Hartley
Managing Operations Across the Supply
Chain
Fourth Edition
Lind, Marchal, and Wathen
Lind, Marchal, and Wathen
Statistical Techniques in Business and
Economics
Eighteenth Edition
Jaggia and Kelly
Business Statistics: Communicating
with Numbers
Third Edition
Jaggia and Kelly
Essentials of Business Statistics:
Communicating with Numbers
Second Edition
McGuckian
Connect Master: Business Statistics
a
=
s
r
o
h
t
u
A
About the
Roger G. Schroeder
is the Frank A. Donaldson Chair Emeritus in Supply Chain and Operations Management
at the Curtis L. Carlson School of Management, University of Minnesota. He received
B.S. and MSIE degrees in Industrial Engineering with high distinction from the University of Minnesota, and a Ph.D. from Northwestern University. He held positions in the
Carlson School of Management as Director of the Ph.D. program, Chair of the Operations
and Management Science Department, and Co-Director of the Joseph M. Juran Center for
Leadership in Quality. Professor Schroeder has obtained research grants from the National
the Ford Foundation,
Science Foundation,
and the American
Production
and Inventory
Control Society. His research is in the areas of quality management, operations strategy,
and high-performance manufacturing, and he is among the most widely published and
cited researchers in the field of operations management. He has been selected as a member of the University of Minnesota Academy of Distinguished Teachers and is a recipient
of the Morse Award for outstanding teaching. Professor Schroeder received the lifetime
achievement award in operations management from the Academy of Management, and he
is a Fellow of the Decision Sciences Institute and a Fellow of the Production and Operations Management Society. Professor Schroeder has consulted widely with numerous organizations, including 3M, Honeywell, General Mills, Motorola, Golden Valley Foods, and
Prudential Life Insurance Company.
Susan Meyer Goldstein
is Associate Professor in the Supply Chain and Operations Department at the Curtis L. Carlson
School of Management, University of Minnesota. She earned a B.S. degree in Genetics and
Cell Biology and an M.B.A. at the University of Minnesota and worked in the health care
industry for several years. She later obtained a Ph.D. in operations management from Fisher
College of Business at The Ohio State University. She has served on the faculty at the University of Minnesota since 1998 and was a Visiting Professor at the Olin Business School at
Washington University in St. Louis for two years. Her current research and teaching interests
involve service process design and management, as well as operations strategy issues. Her
research has been published in Decision Sciences, Journal of Operations Management,
Production and Operations Management, and Manufacturing and Service Operations
Management, among others. She serves on the editorial boards of many operations and service journals. She is the recipient of several research awards and research grants, and has
received the Carlson School of Management Teaching Award and the Carlson School of
Management Service Award.
Dedication
To our families, whose encouragement and love we appreciate
—Roger G. Schroeder
—Susan Meyer Goldstein
oO
a”
°o
°
o
°
°o
FEATURES
Operations and supply chain management is an exciting and vital field in today’s complex business world. Therefore, students in both MBA and undergraduate courses have an
urgent need to understand operations—an essential function in every business.
This textbook on Operations Management in the Supply Chain emphasizes decision
making in operations with a supply chain orientation. The text provides materials of interest to general business students and operations and supply chain management majors. By
stressing cross-functional decision making, the text provides a unique and current business
perspective for all students. This is the first text to incorporate cross-functional decision
making in every chapter, which provides more relevance for non-majors.
The book is organized into five unique sections to help students understand the key types
of decisions made by operations and supply chain managers. See the illustration below.
Introduction
1. Process design
- How to get work done?
2. Quality
- How to satisfy customers?
3. Capacity and scheduling
- When and how much work to do?
4. Inventory
- How to manage parts and products?
5. Supply chain decisions
- How to manage across organizations?
The text provides a balanced treatment of both service and manufacturing firms. Many
books give only cursory treatment to service operations.
The most current knowledge is incorporated, including global operations, supply chain
management, service blueprinting, competency-based strategy, Six Sigma, lean systems,
3D printing, blockchain technology, artificial intelligence, analytics, sustainability, and supply chain risk. Complete coverage is also provided on traditional topics, including process
design, service systems, quality management, ERP, inventory control, and scheduling.
Decision-making framework for operations in the supply chain.
The Firm
Human
"
Supplier
Resources
;
fis
Supply
f
Chain
hi
Decisions
Information
Systems
ea
Fi
aie
ee
Distributor
Supply
a
j
Chain
7]
Decisions
UKE ec
Sn
int
°
th
°
°
°
°
oe
°o
°
While covering the concepts of operations and supply chain management in 18 chapters,
the book also provides 19 case studies. A key feature of this book is learning how operations
issues are tackled in real situations. The cases are intended to strengthen problem formulation skills and illustrate the concepts presented in the text. Long and short case studies are
included. The cases are not just large problems or examples; rather, they are substantial management case studies, including some from Amazon, 3M, Mayo Clinic, and Polaris Industries.
The softcover edition with fewer pages than most introductory books covers all the essentials students need to know about operations management in the supply chain, leaving out only
superfluous and tangential topics. By limiting the size of the book, we have condensed the
material to the basics. The book is also available in Connect and LearnSmart digital versions.
This book is ideal for regular operations and supply chain management courses and also
case courses and modular courses. It is particularly useful for those who desire a crossfunctional and decision-making perspective that reaches across the supply chain. Instructors
can easily supplement the text with their own cases, readings, or course materials as desired.
The Connect Library and Instructor Resources contain 22 Excel templates designed to
assist in solving analytic problems at the end of chapters and the case studies. These resources
also contain technical chapters on linear programming, simulation, transportation method, and
queuing, which can be assigned by the instructor, if desired. Using these resources covers
all the main analytics in operations and supply chain management. The resources also have
PowerPoint slides, a solutions manual, and the test bank. Access to these resources can be
obtained from your McGraw-Hill sales representative or directly in the Connect Library.
Walkthrough of Key Learning Features
® Over forty Operations Leader boxes are included in the chapters to illustrate current
practices implemented by leading firms
ia
| OPERA
S LEADER
Nike Does It
Nike
and
among
Nike-owned
other
athletic
offer customized
Converse,
shoe
shoes
Shoes
brands,
that can
with
a
customized
“fit”
are significantly more elusive. While
there are firms offering custom-fitted
be
ordered online for a reasonable price
weeks.
shoes, they sell at prices reflecting
the significant work to individually
Customers can select from numerous
size the shoe, most likely performed
and
delivered
within
a few
fabric or leather colors and patterns
on various pieces of the shoe, as well
as the colors of laces, stitching, and
in a job shop. These shoes are cusobsession.24k/Stockimo/Alamy Stock Photo
soles. These shoes, with their many customizable options,
are an example of successful mass customization.
tom, but not mass customized.
Mass customization gives customers many options, as well as the enjoyment of designing
and using a product with their own personal stamp on it.
® Every function in every organization touches Operations and the Supply
Chain in some manner. This is the first book to add materials in every
chapter to show how topics apply to majors in Marketing, Finance,
Accounting, Human Resources, and Information Systems. The handshake symbol indicates these cross-functional decisions.
=
°
wie
:
'
ai
ee
O°
So
eewe
re
°
:
at
ak
°
al
°
Bets
°°
°
2
°
$
°
Preface
| vii} 1
ag
® To help practice calculations, example boxes are included within chapters and solved
problems are added at the end of the chapter.
Example
Suppose demand at the receiving work center B is 2 parts per minute and a standard container holds 25 parts. It takes 100 minutes for a container to make a complete circuit from
work center A to work center B and back to A again, including all setup, run, move, and wait
times. The number of containers needed in this case is:
Ae 2100s
25
The maximum
inventory in the production system, a useful measure
of how lean the
system is, equal to the container size times the number of containers (200 units = 8 x 25),
since the most inventory we can have is all containers filled:
Maximum inventory = nC = DT
SOLVED PROBLEMS
Problem
1. Kanban and takt time. A work center uses kanban containers that hold 300 parts. To
produce enough parts to fill the container, 90 minutes of setup plus run time are needed.
Moving the container to the next workstation, waiting time, processing time at the next
workstation, and return of the empty container take 140 minutes. There is an overall
demand rate of nine units per minute.
a. Calculate the number of containers needed for the system.
b. What is the maximum inventory in the system?
c. A quality team has discovered how to reduce setup time by 65 minutes. If these
changes are made, can the number of containers be reduced?
d. What is the takt time for this process?
Solution
a. T is the time required for a container to complete an entire circuit, in this case
90 minutes for setup and run time plus 140 minutes to move the container through
the rest of the circuit.
n= DT
+C=(9 Xx (90 + 140)) + 300 = 6.9 (round up to 7)
b. Since production will stop when all the containers are full, the maximum inventory
is when all containers are full, that is, wC:
nC = 7(300) = 2100
c.
n= DT +C=(9 X (25 + 140)) = 300 = 4.95 (round up to 5), so yes, the number of
containers can be reduced from 7 to 5.
d. Takt time = 1/9 minute = 60/9 seconds = 6.67 seconds. Since the process produces
9 units per minute, the takt time is 1/9 minute or 6.67 seconds per unit.
® Students can both preview and review the key points and terms. These are found at the end
of each chapter.
7.7
KEY POINTS AND TERMS
Lean concepts, principles, and tenets can be deployed to reduce waste in manufacturing
and service firms. We have seen how the lean tenets create lean production systems with
non-value-added activities eliminated and waste minimized. Key points in the chapter
include the following:
*
Lean thinking is a way of thinking about processes that includes five tenets: specify
customer value, improve the value stream, flow the product or service, pull from the
customer, and strive for perfection.
*
The five lean tenets seek to eliminate waste by utilizing the full capability of workers
and partners in continuous improvement efforts. Lean tools, or methods, are described
for each of the five tenets.
¢
In manufacturing, smooth flow is ensured by a stable and level master schedule.
This requires consistent daily production within the master schedule and mixed
model assembly. Takt time matches the rate of output with the average demand rate
in the market.
le
°
°
ee)
=
uy
°°o
is]
°
°
°
a.
9
o
a
°
o.°98
°
°
o
°
oy
o
«@
9
°
?
si
ae
o
cs
°
°
e
°o
°
°
°
°
ae
yy | viii
© 4°
Co
°
Preface
¥
a
o
i
o
(0
ie
°
°
o
:
o°
=
6
°°
6
(TPS)
F
pe
o
°
119
°
°
°o
<
Push
123
Internal setup 128
Pull
123
External setup
121
schedule
127
Value stream 121
Value stream
mapping 121
Uniform load 127
Takt time 127
Reducing setup time
Gemba
Single setups 128
121
128
Cellular manufacturing 129
Preventative maintenance 129
Cross-training 130
Respect for people 130
Perfection 124
5 Whys 125
Son t25
Stabilizing the master
Just-in-Time (JIT)
manufacturing 119
Lean production 119
Lean thinking 120
Waste (muda)
:
°
Toyota Production System
Key Terms
°
°
Kanban
128
130
Reducing lead time 133
Supplier relationships 133
Co-location 133
Kaizen
135
® Twenty-two Excel spreadsheets are included for solving problems and analyzing case
studies using analytic methods.
i
in aa factory,
eXce [ 11. : As cereal boxes are filled
in
factory, they
are weighed for their contents by an automatic
scale. The target value is to put 10 ounces of cereal in
each box. Twenty samples of three boxes each have
been weighed for quality control purposes. The fill
weight for each box is shown below.
Observation
Sample
1
2
3
1
2
3
4
10.01
9.87
10.08
10.17
9.90
10.20
9.89
10.01
10.03
10.15
9.76
9.83
® Case studies provide students practice in formulating and solving unstructured problems.
Case Study
Lawn King, Inc.: Sales and Operations Planning
John Conner, marketing manager for Lawn King, looked
over the beautiful countryside as he drove to the corporate headquarters in Moline, Illinois. John had asked
his boss, Kathy Wayne, the general manager of Lawn
King, to call a meeting in order to review the latest fore-
leXcel]
The changeover cost of the production line depends
on which type of mower is being produced and the next
production model planned. For example, it is relatively
easy to change over from the 20-inch push mower
to the
20-inch self-propelled mower, since the mower frame is
® Throughout the text, company and industry examples illustrate real use of the ideas.
Harrah’s uses superior customer service to increase profits.
Grocery self-service is a provider-routed service.
Leonard Zhukovsky/123RF
Syda Productions/Shutterstock
°
°
‘
6
°°
°
°
cS)
°
°
bo
o
See
ae)
a
Oo
°o
o
°
°
»
°
+
.
ed
C
°
° 6
°
°
°
©
0
°°
°°
°
°
°
o
°
°
2
°
°
oo
°
Preface
| ix]
® Atthe end of every chapter, Learning Enrichment boxes provide videos and websites where
students can extend their knowledge on chapter topics using Internet content.
KEY CHANGES
LEARNING
What Is the New Product Development Process?
Video
ENRICHMENT
https://youtu.be/VQZjNIRpuFg
2:49
(for self-study or
Prototyping
Video
instructor assignments)
htps://youtu.be/SSWt-TSYD08
2:26
3D Printing Prototype Example
https://youtu.be/RpFTRT8FkPO
Video
3:02
Sustainability in New Product Development
https://youtu.be/-HS-slU-XTc
Video
3:56
IN THE EIGHTH EDITION
This book is known for its decision orientation and case studies. We have strengthened the
decision-making framework by adding new material on digital technology, lean systems,
sustainability, and global supply chains. We also include new and existing cases to address
these decisions. The Eighth Edition features a new 4-color design and the following
major changes:
1. Cross-functional. Most books for operations and supply chain core courses are merely
summaries for majors in operations and supply chain management. None address
the general business student who is interested in Marketing, Finance, Accounting,
or Information Systems. We make this book more applicable and interesting to the
approximately 80 percent of business students who don’t major in operations and
supply chain management. We add cross-functional materials in each chapter to show
how the topics apply to non-majors. The handshake symbols in the margin identify
the content.
2. Digital Technology. The Eighth Edition has substantial updates and additions on
four digital technologies. 3D printing is becoming useful for producing spare parts,
custom manufacturing, medical devices, dental implants, and architectural models.
Blockchain software is being developed and tested by many global logistics companies. Artificial intelligence is rapidly developing for service applications, automobiles, and manufacturing plants. Analytics are being applied to both large and small
databases. Analytics that are descriptive, predictive, or prescriptive in nature are
discussed. These digital technologies are described in detail in several chapters in
the book.
3. Supply Chain Sustainability. We introduce the idea of the triple bottom line regarding
environmental, social, and economic sustainability. Sustainability is preserving the earth
and resources for future generations. Environmental sustainability is related to global
warming,
clean water, clean air, and environmental
protection.
Social
sustainability
means hiring a diverse workforce, ethical practices, providing equal opportunity, and
safe working conditions, for example. Economic sustainability is making a sufficient
profit for the firm’s survival in the future. Operations and supply chain managers
are actively pursuing all three aspects of sustainability of operations and associated
supply chains.
°
°
°o
°
°
°
a
oo
re
°
oO
°
°
°
9
4
:
°
o
ie
“a
°
oOo
9
°°
°
°
°°
°
°
=
°
°
:
°
°
°
mix
| Preface
. Global Supply Chains. In this new edition we have increased our attention to
global supply chains by adding new sections on global services, global sourcing,
and global logistics. The text explains how to make global decisions that balance
the lower costs of overseas sourcing and logistics with the risks of quality failures,
loss of intellectual property, increased monitoring costs, and exposure to financial
and political risks.
. Lean Systems. Most books discuss up to 15 techniques of lean including reduced
setup time, small lot sizes, uniform load, and takt time. We have completely reorganized the lean chapter around the five tenets and principles of lean systems to
include all of these techniques. This clarifies lean systems in terms of creating
value for the customer, eliminating waste, ensuring flow, customer pull, and striv-
ing for perfection.
. Practical Examples. The text contains over 70 practical examples of concepts, ideas,
and analytics. Nineteen new Operations Leader boxes have been added for companies
including Southwest Airlines, Lego Group, Culver’s, Nike, LG Electronics, and Trader
Joe’s. In addition twenty-five existing Operations Leader boxes have been updated in
the various chapters.
. Learning Enrichment boxes. Every chapter ends with a Learning Enrichment box
for student self-study or instructor assignments. These boxes have YouTube video
links and websites that expand on the coverage in the chapter. They cover ideas from
the chapter in more detail or provide examples of the how the ideas are used. This
is one of the first books to make extensive use of the Internet to enrich the material
covered in the text.
When reviewers of this book were asked how they would describe the text to a colleague, they said:
“T would highly recommend the book to them. While other textbooks either focus on the
techniques or concepts, this book does a good job in addressing both equally well.”
“A solid textbook that is well-written. Good coverage of basic operations management
material.”
“Tt is a guide to operations that takes a practical approach with a strong emphasis on
case materials to put concepts into practice.”
CHAPTER REVISIONS AND CASES FOR THE EIGHTH EDITION
1. Introduction to Operations. The first part of the chapter is rewritten to clearly define
operations and supply chain management. A new section explains the role of operations
in the firm and the economy including productivity calculations. The triple bottom line
is defined for environmental, social, and economic sustainability. Internet links are provided in the Learning Enrichment box on sustainability and globalization.
. Operations and Supply Chain Strategy. A new Operations Leader box on Southwest
Airlines is added. Sustainability, as an objective, is added to cost, quality, delivery and
flexibility. Emphasis is placed on decision making in operations that is contingent on
Strategy.
. Product Design. New content is added on the use of 3D printing for creating prototypes. Concurrent engineering is illustrated using a new example from NASA. New
Operations Leader boxes describe how The LEGO Group tackles sustainability challenges and how TPI Composites is developing and manufacturing blades for wind turbine energy systems.
°
°
°
°
Py
°
S
°
°
o
or
°
°
Preface
xi
. Process Selection. There is a new example of focused operations in a service firm,
Midwest Orthopedic Specialty Hospital. Two new Operations Leader boxes describe
the food production system at Culver’s and mass customization at Nike. We also
expand on the role of 3D printing in modern manufacturing, particularly in the medical sector.
. Service Process Design. The relevance of service operations to non-majors is discussed. A new Operations Leader box on the City of Fort Collins is added. This
edition is the first to offer sections on Technology for Services and Globalization
of Services.
. Process-Flow Analysis. Cross-functional material is added to show its importance to
systems thinking. A Learning Enrichment box is included with YouTube videos and
Internet links on process mapping, Little’s Law, and queueing at Disney.
. Lean Thinking and Lean Systems. This chapter is completely reorganized around
the five lean tenets to clarify the principles and concepts underlying lean thinking.
New material is added on cellular manufacturing and the pull system. The chapter is
the first to take a principle and conceptual approach to lean systems, rather than a listing of techniques and methods used.
. Managing Quality. YouTube videos are added to the Learning Enrichment box to
expand on ISO9000 certification, mistake proofing, and the Baldrige Award for health
care. Quality is expanded to include the entire supply chain, not just the focal firm.
The highly cross-functional nature of quality is emphasized.
. Quality Control and Improvement. We explain why all business students should
learn about quality control. The difference between special causes and common causes
is emphasized. We clarify the differences between statistical process control and process capability. The section on Six Sigma was rewritten to expand the content.
10. Forecasting. We shift our description of forecasting to “analytics” so that students
can understand how the popular focus on analytics is utilized in operations and supply
chain. There is a new section on big data and its use in forecasting, along with a new
Operations Leader box on how Amazon uses big data in its own forecasting. When
and how to use MAD, 1s clarified, in addition to many minor clarifications in using
formulas throughout the chapter.
. Capacity Planning. We expand discussion about how all functions are involved in
and impacted by capacity planning. Improvements in the descriptions of Sales and
Operations Planning (S&OP) as well as aggregate planning help to clarify the process
involved and the challenges faced. Calculations for level and chase strategies are clarified. New Operations Leader boxes on Delta Airlines and Hostess Brands make these
concepts tangible for students.
. Scheduling Operations. We add new material on the theory of constraints about how
to identify the bottleneck constraint and eliminate it while subordinating everything
else. In the Learning Enrichment box interesting YouTube videos are provided on job
shop scheduling at Washburn Guitar, round-robin CPU scheduling, and the theory
of constraints.
. Project Planning and Scheduling. The chapter is updated to illustrate the many
industry settings in which projects require skilled management—from manufacturing
to service firms, nonprofits, and government. A new Operations Leader box on the
Carlsbad Desalination Plant in San Diego provides a nice example of a major multigovernment project. Other updates include the Project Management Institute’s Body
of Knowledge in Table 13.3.
°
°
”
i+]
o
-
°
°
°
° °
°
°
°
°
°
°
ere
°
°
°
:
co
e
x
°°
° 6
6
=
9
9
°
o
°
°
1-2
°
:
°
°
|
xi
Preface
14. Independent Demand Inventory. The chapter includes additional content on vendor managed inventory (VMI), along with a new Operations Leader box illustrating
VMI at Procter and Gamble Co. An additional new Operations Leader box on IKEA
describes the use of a min/max inventory replenishment system. The Learning Enrichment box at the end of the chapter provides video and web sources for additional
information.
15. Materials Requirements Planning and ERP. We clarify for students exactly which
elements constitute the MRP system. We also describe the use of Oracle’s ERP software at Cleveland Clinic to help students understand the breadth of these system’s use
in industry. A new Operations Leader box on LG Electronics provides a useful illustration of how a global firm benefits from these systems.
16. Supply Chain Management. This is one of the first books to have a separate section
on blockchain technology to explain its uses and methods. More details are also provided on the SCOR model. We rewrote the section on measures of throughput time,
cash-to-cash cycle time and total delivered cost for analyzing an entire supply chain.
We added a new section on the Amazon effect and omni-channel marketing, also a
first in textbooks.
17. Sourcing. The chapter includes an entirely new section on Global Sourcing, including
discussion of risks and benefits. Presentation of Total Cost Analysis is expanded. A
new Operations Leader box on Trader Joe’s sourcing strategy will appeal to students.
18. Global Logistics. A new section on Global Logistics includes a figure to illustrate the
multimodal activities in global supply chains. The concepts of intermodal and shipping zones have been added and described. A new Operations Leader box on Home
Depot provides insight on how online sales are served from stores and warehouses,
while an expanded box on Ryder gives students a glimpse of the people and assets
needed for this major 3PL provider.
\
Case Study Revisions
A few of the 19 case studies are described below:
Amazon Revolutionizes Supply Chain Management. This new case, written exclusively
for this book, describes the evolution of Amazon’s supply chain and its purchase of
Whole Foods. It challenges students to think about how Amazon can change the future
of Whole Foods to increase its revenues and earnings. The case also contrasts what Walmart
is doing to use e-commerce to compete with Amazon. The case asks students to define the
effect of the emergent business strategies of Amazon and Walmart on the supply chains of
these companies in terms of locations, sourcing, capacity, and inventory.
Operations Strategy at BYD of China, Electrifying the World’s Automotive Market.
BYD, the leading electric vehicle company in the world, must develop a strategy for adapting its supply chain in the future. We updated this case from its last update in 2015 and
revised the teaching note for this rapidly changing industry.
Early Supplier Integration
for John Deere Skid-Steer Loader. Deere and Company must
decide how to involve suppliers in the design of its new Skid-Steer Loader. This case is
updated and the teaching note revised.
The Evolution of Lean Six Sigma at 3M Inc. Students are asked to evaluate 3M’s use of
Six Sigma and Jean thinking. We added significant new information since the last update
in 2012.
Consolidated Electric: Inventory Control. Management is designing a new inventory control system. The student questions are tailored to increase student learning about
this system.
°
°
°
:
a]
°
°
o
°
,
°
a)
°
°
°
°
°
a
°
rey
©
°
°
° °
°o
6
©
2
°
°
°
°
°
~
°
°
°
°
°
°
°
°
°
i]
o
ae
°
4
°
°
a
°
°
°
°
Preface
| xiii||
Altimus Brands: Managing Procurement Risk. Altimus is deciding which of four offshore suppliers offers the lowest cost and risk for future purchase contracts. We updated
the case and wrote a new teaching note.
ShelterBox: A Decade of Disaster Relief. After the Haiti earthquake of 2010, ShelterBox
provided immediate relief and is considering what decisions should be made in advance of
future disasters. A new teaching note was written for this case.
INSTRUCTOR
eam comect
RESOURCES
McGraw-Hill Connect®
McGraw-Hill Connect® is an online assignment and assessment solution that connects
students with the tools and resources they’Il need to achieve success through faster learning, higher retention, and more efficient studying. It provides instructors with tools to
quickly pick content and assignments according to the topics they want to emphasize.
Instructor Library. The Connect Operations Management Instructor Library is your
repository for additional resources to improve student engagement in and out of class. You can
select and use any asset that enhances your lecture. The Connect Instructor Library includes:
¢
Solutions Manual. Prepared by the authors, this manual contains solutions to all the
end-of-chapter problems and cases.
¢
Test Bank. The Test Bank includes true/false, multiple-choice, and discussion ques-
¢
¢
¢
¢
tions/problems at varying levels of difficulty. All test bank questions are also available
in a flexible electronic test generator. The answers to all questions are given, along with
a rating of the level of difficulty, chapter learning objective met, Bloom’s taxonomy
question type, and the AACSB knowledge category.
PowerPoint Slides. The PowerPoint slides draw on the highlights of each chapter and provide
an opportunity for the instructor to emphasize the key concepts in class discussions.
Digital Image Library. All the figures in the book are included for insertion in PowerPoint slides or for class discussion.
Excel Spreadsheets. Twenty Excel Spreadsheets are provided for students to solve designated problems at the end of chapters.
Technical Chapters. Additional Operations analytics are available in four inline Technical Chapters. These are available in the Instructor Resource Library through Connect,
or by visiting the URL at www.mhhe.com/schroeder8e
The instructor and student resources can also be accessed directly at www.mhhe.com/
schroeder8e.
eam @eority
Tegrity Campus: Lectures 24/7
Tegrity Campus is a service that makes class time available 24/7 by automatically capturing every lecture in a searchable format for students to review when they study and
complete assignments. With a simple one-click start-and-stop process, you capture all
computer screens and corresponding audio. Students can replay any part of any class with
easy-to-use browser-based viewing on a PC or Mac.
Educators know that the more students can see, hear, and experience class resources,
the better they learn. In fact, studies prove it. With Tegrity Campus, students quickly recall
key moments by using Tegrity Campus’s unique search feature. This search helps students efficiently find what they need, when they need it, across an entire semester of class
recordings. Help turn all your students’ study time into learning moments immediately
supported by your lecture. To learn more about Tegrity, watch a two-minute Flash demo at
http://tegritycampus.mhhe.com.
°
°
S
9
9
°
4
°
°
°
°
°°
°
°
°
:
°
°
o
°
°
°
°
°
a
o
oOo
°
_
°
-
°
°o
°
o
°
ae
°
°
°
°
by=
ss
°
°
& P|xiv
Preface
°
° id
°
°
°
°
sa°
Pm
°
°
°
°
o
o
°
°
°o
6
MCGRAW-HILL CUSTOMER CARE CONTACT INFORMATION
At McGraw-Hill, we understand that getting the most from new technology can be challenging. That’s why our services don’t stop after you purchase our products. You can e-mail
our Product Specialists 24 hours a day to get product-training online. Or you can search
our knowledge bank of Frequently Asked Questions on our support website. For Customer
Support, call 800-331-5094 or visit www.mhhe.com/support. One of our Technical Support Analysts will be able to assist you in a timely fashion.
Preface
| ox] |
ACKNOWLEDGMENTS
The authors would like to acknowledge the many individuals who have assisted with this
book. Special thanks go to the reviewers for this edition:
Abirami Radhakrishnan
Mark Goudreau
Morgan State University
Johnson & Wales University
Anita Lee-Post
Mark Hanna
University of Kentucky
Georgia Southern University
Canchu Lin
Mark Jacobs
Carroll University
University of Dayton
Enar Tunc
Richard Hopfensperger
California Polytechnic State University San
Luis Obispo
Marian University of Wisconsin
John Wu
Bradley University
California State University San Bernardino
Steven Dickstein
Jooh Lee
Fisher College- Ohio State University
Ross Fink
Rowan University
Therese Gedemer
Jose Ablanedo-Rosas
Marian University of Fond du Lac
University of Texas El Paso
Todd Henning
Kathy Schaefer
Indiana University
Southwest Minnesota State University
Veena Adlakha
Kimball Bullington
University of Baltimore
Middle Tennessee State University
Weiyong Zhang
Kwasi Amoako-Gyampah
Old Dominion University
University of North Carolina Greensboro
William Ramshaw
Eastern Washington University
The authors would also like to thank the staff at McGraw-Hill Education who had a
direct hand in the editing and production of the text, including Ryan McAndrews, product
developer; Noelle Bathurst, portfolio manager; Harper Christopher, executive marketing
manager; and Fran Simon and Angela Norris, project managers.
We would like to thank our colleagues at the University of Minnesota who listened to
our ideas and provided suggestions for book improvement. Additional thanks go to Doug
and Letty Chard, who diligently and carefully prepared the index. We would also like to
thank Tom Buchner of the University of Minnesota who carefully prepared the test bank
questions. Our thanks to Ed Pappanastos of Troy University for constructing the Connect
solutions to problems. Finally, we thank our families for their patience and perseverance
during the many months of writing and editing. Without their support and encouragement
this textbook would not have been possible.
Roger G. Schroeder
Susan Meyer Goldstein
s
°
°
°
o
°
ff
°
“s
°o
a0.
o
o
°
o
°
So.
°
°
°
©
°
°°
6
°
a
°
o
oe
°
°
°
9
FOR INSTRUCTORS
You’re in the driver’s seat.
Want to build your own course? No problem. Prefer to use our turnkey,
prebuilt course? Easy. Want to make changes throughout the semester?
Sure. And you'll save time with Connect’s auto-grading too.
6
5
re
Less Time
Grading
They’ll thank you for it.
H
Bocovve otesex) DES: lleBiwma) Fa beaCe aoc phaeES,
fomSas Yr saa Vowstacy
etisalat,
teFase
‘Tan daeSpey ofashenpei Ha;foUbeBrae.$0,Wo wera evan Lhnone,Ukethegt anomaReGr.oeGeert it)
Ieler te
(QtWN dae rene (ltNb cn Nc Ta gt keine ryvyUa,
Adaptive study resources like SmartBook® 2.0 help
your students be better prepared in less time. You
can transform your class time from dull definitions to
dynamic debates. Find out more about the powerful
personalized learning experience available in
SmartBook 2.0 at www.mheducation.com/highered/
connect/smartbook
Make it simple,
make it affordable.
Solutions for your
challenges.
Connect makes it easy with seamless
integration using any of the major
A product isn’t a solution. Real
Learning Management Systems—
Blackboard®, Canvas, and D2L, among
others—to let you organize your course
in one convenient location. Give your
students access to digital materials at
a discount with our inclusive access
program. Ask your McGraw-Hill
representative for more information.
and come with training and
Ongoing support when you need it
and how you want it. Our Customer
Experience Group can also help
you troubleshoot tech problems—
although Connect’s 99% uptime
means you might not need to call
them. See for yourself at status.
mheducation.com
solutions are affordable, reliable,
Effective, efficient studying.
Connect helps you be more productive with your study time and get better grades using tools like
SmartBook 2.0, which highlights key concepts and creates a personalized study plan. Connect sets you
up for success, So you walk into class with confidence and walk out with better grades.
Study anytime, anywhere.
Download the free ReadAnywhere app and access your
online eBook or SmartBook 2.0 assignments when it’s
CIE GISEERS SES
app—it made it easy
to study when
you
;
convenient, even if you’re offline. And since the app
don't have your text-
automatically syncs with your eBook and SmartBook
book in front of you.’
2.0 assignments in Connect, all of your work is available
every time you open it. Find out more at
www.mheducation.com/readanywhere
- Jordan Cunningham,
9
Eastern Washington University
No surprises.
The Connect Calendar and Reports tools keep you on track with the
work you need to get done and your assignment scores. Life gets busy;
Connect tools help you keep learning through it all.
Learning for everyone.
McGraw-Hill works directly with Accessibility Services
Departments and faculty to meet the learning needs
of all students. Please contact your Accessibility
Services office and ask them to email
accessibility@mheducation.com, or visit
www.mheducation.com/about/accessibility
for more information.
Briet Table of Contents
ee
el
1
@
About the Authors — iv
12
Scheduling Operations
Preface
13
Project Planning and Scheduling
v
Inventory
1
1
Introduction to Operations
2
Operations and Supply Chain
Strategy
20
3
Product Design
2
14
15
37
PART TWO
Process Design
267
PART FIVE
PART ONE
Introduction
249
291
Independent Demand Inventory
292
Supplement: Advanced Models
320
Materials Requirements Planning
and RP
S75
PART SIX
Supply Chain Decisions
53
4
Process Selection
54
5
Service Process Design
6
Process-Flow Analysis
7
Lean Thinking and Lean
Systems
118
347
16
Supply Chain Management
76
17
Sourcing
97
18
Global Logistics
348
376
397
PART SEVEN
Case Studies
423
PART THREE
Quality
APPENDIXES
141
8
Managing Quality
9
Quality Control and
Improvement
163
142
11
ay
i
see
Forecasting
509
ACRONYMS
PART FOUR
Capacity and Scheduling
10
INDEX
189
507
519
Technical Chapters available in the Instructor’s
Resource Library in Connect
Waiting Lines
190
Simulation
Supplement: Advanced
Methods
215
Transportation Method
Capacity Planning
Linear Programming
220
°
*
°
Contents.
iit
Preface
v
PART
ONE
°
1.3
1.4
15
1.6
1.7
1.8
0°
°
ee
8
ew
2.7.
1
2
Definition of Operations and Supply Chain
Management
3
The Role of Operations and Supply Chain
Management
4
Why Study Operations and Supply Chain
Management?
6
Decisions at PizzaU.S.A.
9
Operations Decisions in the Supply
Chain—A Framework
10
Cross-Functional Decision Making
12
Operations as a Process
13
Trends in Operations and Supply Chain
Mangement
15
Services
15
Global Operations and Supply Chains
3.1
3.2
Learning Enrichment
35
Discussion Questions
36
37
Strategies for New Product
Introduction
38
New Product Development Process
Concept Development
41
Cross-Functional Product Design
42
Supply Chain Collaboration
43
Quality Function Deployment
45
Customer Attributes
46
Engineering Characteristics
3.6
3.7.
Modular Design
48
Key Points and Terms
49
Learning Enrichment
50
Di
:
39
ESS OMIOEOS
16
39
40
Product Design
40
Pilot Production/Testing
3.3.
3.4
3.5
30
Supply Chain Strategy
31
— Environment and Sustainable
Operations
33
~—‘Key Points and Terms
34
Chapter 3
x
Product Design
15
Digital Technologies
6
°
2.5
2.6
Sustainability
-
Q
2.4
Chapter 1
:
:
Introduction to Operations
1.2
o
iv
INTRODUCTION
14
—o
0
ro)
Yah
About the Authors
bs
46
Integration ofDecisions Internally and
Externally
16
Globalization of Operations and the Supply
Chain
1.9
2.2
2.3.
53
17
Key Points and Terms
17
Chapter 4
Learning Enrichment
18
Process Selection
Discussion Questions
18
Chapter 2
Operations and Supply Chain
Strategy
20
ZA
PART TWO
PROCESS DESIGN
Operations Strategy Model
22
Corporate and Business Strategy 23
Operations Mission 24
Operations Objectives 24
Strategic Decisions 25
Distinctive Competence 26
Competing with Operations Objectives
27
Cross-Functional Strategic Decisions
28
54
4.1
‘i
Product-Flow Characteristics
4.2
Approaches to Order Fulfillment
4.3
4.4
4.5
4.6
4.7
63
Process Selection Decisions
Product-Process Strategy
64
Focused Operations
66
Mass Customization
67
3D Printing and Additive
Manufacturing
69
Environmental Concerns
70
Cross-Functional Decision Making
Key Points and Terms
72
Learning Enrichment
74
Discussion Questions
74
4.8
4.9
4.10
55
60
71
Eg
Contents
rye:
°°
5.1
5.2
5.3.
5.4
5.5
5.6
Defining Service
78
Service-Product Bundle
79
Service Delivery System Matrix
80
Customer Contact
83
Service Recovery and Guarantees
86
Technology for Services
87
6.1
6.2
6.3.
6.4
6.5
6.6
6.7
6.8
Learning Enrichment
95
Discussion Questions
96
Solved Problems
Problems
92
QUALITY
Value Stream
121
Strive for Perfection
Ensure Flow
Quality as Customer Requirements
Product Quality
144
Service Quality
146
8.4
Quality Planning, Control, and
8.7
8.8
W. Edwards Deming
8.9
8.10
115
143
Improvement
146
Mistake-Proofing
149
— Ensuring Quality in the Supply
Chain
150
Quality, Cost of Quality, and Financial
Performance
151
Quality Pioneers
154
Joseph Juran
154
154
ISO 9000 Standards
156
Malcolm Baldrige Award
8.11.
158
Why Some Quality Improvement Efforts
8.12
Fail 160
Key Points and Terms
161
Learning Enrichment
162
Discussion Questions
162
118
119
Chapter 9
Quality Control and Improvement
9.1
Design of Quality Control
Systems
164
9.2
Process Quality Control
167
9.3.
Attribute Control Chart
169
9.4
Variables Control Chart
170
9.5
Using Control Charts
171
9.6 _ Process Capability
172
9.7
Continuous Improvement
174
9.3"
“Six Sigma
#178
124
126
Stabilize Master Schedule
127
Reducing Setup Time and Lot Sizes
127
Changing Layout and Maintenance
129
Cross-Training and Engaging
Workers
142
8.1
8.2
8.3.
8.5
8.6
Ensure Flow
122
Customer Pull 123
7.3
141
Chapter 8
Managing Quality
Chapter 7
Lean Thinking and Lean Systems
120
139
140
PART THREE
115
Create Value
138
138
Discussion Questions
113
Evolution of Lean
Lean Tenets
120
i
Customer Pull
130
Changing Relationships with
Suppliers
133
Implementation of Lean
135
Key Points and Terms
137
113
Discussion Questions
Problems
°
Learning Enrichment
Process Thinking
98
The Process View of Business
99
Process Flowcharting
100
_—_Process-Flow Analysis as Asking
Questions
104
Process Analytics
106
Analyzing Process Flows at Pizza
U.S.A»
108
_ Process Redesign
110
Key Pointsand Terms
112
Learning Enrichment
7.1
7.2
7.6
7.7
97
Solved Problems
Vis
°
8&8
— Globalization of Services
90
Service Profitability and Employees
Key Points and Terms
94
Chapter 6
Process-Flow Analysis
*
7.4
7.5
76
Artificial Intelligence
5.7.
5.8
5.9
w°
* 6
Chapter 5
Service Process Design
°
130
fo
°
°8
°
wre
_
6
5°
Be,
° .
°
°
163
o
°
oO.
°
’
9.9
9.10
°
S
.6
ou
:
.
<=
°
Ss
‘
°
i
°
P-9
Solved Problems
°
_
11.9
in
7
°
Contents
<:
Key Pointsand Terms
Learning Enrichment
183
Solved Problems
183
239
240
240
Discussion Questions
Discussion Questions
Problems
en
°
Leanand Six Sigma _ 180
Key Points and Terms
182
Learning Enrichment
Ps
o ?
aera .o
186
Problem
245
246
186
Chapter 12
PART FOUR
CAPACITY
Scheduling Operations
AND SCHEDULING
= 189
Chapter 10
10.1
10.2
10.3.
10.4
190
Forecasting for Decision Making
Qualitative Forecasting Methods
Time Series Analytics
195
Moving Average
Ye
192
193
Batch Scheduling
196
10.5
Exponential Smoothing
Forecast Accuracy
10.7
10.8
Advanced Time-Series Forecasting
Causal Forecasting Analytics 204
10.9
Selecting a Forecasting Method
Big Data 207
10.10
Collaborative
12.2.
Gantt Charts
12.4
Theory of Constraints
by
ae
:
asia peer pale “et
i"Bn abe ee aya Sqaay
Sy Ee os
i
les
Planning,
Learning Enrichment
198
eH
254
256
263
203
Problems
and
13.1
13.2
13.3
13.4
13.5
13.6
210
210
212
213
215
Objectives and Trade-offs
268
Planning and Control in Projects
Scheduling Methods
272
Constant-Time Networks
273
CPM Method
279
Use of Project Management
Concepts
281
Key Points and Terms
282
13.7.
269
Learning Enrichment
283
Solved Problems
284
220
Discussion Questions
Capacity Defined
221
Facilities Decisions
223
Problems
Amount of Capacity
224
225
Size ofFacilities
Timing of Facility Decisions
Facility Location
226
Types of Facilities
227
PART
Chapter
Sales and Operations Planning
228
287
FIVE
INVENTORY
226
287
291
14
1A
CiOR FHRSOHUUNECS cESEORN 236
Independent
e
pea
ndependent Demz
Demand Inventory
11.5
11.6
Planning Options
231
Basic Aggregate Planning
Strategies
233
Aggregate Planning Costs 234
Aggregate Planning Example
235
14.1
14.2
14.3.
14.4
11.7.
11.8
267
ets
Chapter 11
Capacity Planning
DAR
Chapter 13
Project Planning and Scheduling
ting,
Supplement: Advanced Methods
RESTiayIS
splik
ibicapeg?2!
206
F
Discussion Questions
OE
DTS
TRononishisatko O08 o soa
Key Points and Terms
209
Solved Problems
11.3.
251
Finite Capacity Scheduling
201
Learning Enrichment
11.1
11.2
250
&
10.6
40.11.
12.1.
12.3
Forecasting
249
P
°
o
=
*.
@
ea
, °
gt
° z
°
°
*o
. . °
°
os
‘3 °°
°
9
Definition of Inventory
293
Purpose of Inventories
295
Costs of Inventory
296
Independent versus Dependent
Demand
297
°
eae,
aay
° Ba
°
es 4
°
°
29Nb
|xxi||
.
i |xxii | Contents
Py]
14.5
14.6
14.7
14.8
14.9
14.10
14.11.
::
sto
we
a0
16.3
Economic Order Quantity 298
Continuous Review System
302
Periodic Review System
306
Using P and Q Systems in Practice
309
Vendor Managed Inventory
311
ABC Classification of Inventory
312
313
Key Points and Terms
Learning Enrichment 314
Solved Problems 315
Discussion Questions
Problems
6
See
fectae
5°
* 1a
16.4
16.5
16.6
16.7.
4
Parae
we , —
Supply Chain Dynamics—The Bullwhip
Effect Baas
Improving Supply Chain
Performance
358
Supply Chain Structural
Improvements
358
361
Supply Chain System Improvements
Technology and Supply Chain
Management
362
317
E-commerce and Omni-channel
317
Marketing
Supplement: Advanced Models
320
364
Blockchain Technology
364
16.8
Supply Chain Risk and Resilience
366
Analysis of Supply Chain Risk 367
Materials Requirements Planning and
16.9
Sustainability of the Supply Chain
ERP
323
16.10
Key Points and Terms
372
Learning Enrichment
374
15.1
15.2
The MRP System
324
MRP versus Order-Point Systems
15.3.
Parts Explosion: How an MRP System
Chapter
15
Works
15.4
pe
Inventory Records
333
Capacity Planning
Purchasing
334
Shop-Floor Control
334
334
15.6
The Successful MRP System
15.7
Enterprise Resource Planning
Systems
337
Key Points and Terms
340
SIX
Sourcing Goals
17.3.
Insource or Outsource?
377
378
378
Advantages of Outsourcing
379
Disadvantages of Outsourcing
380
Total Cost Analysis
381
335
17.4
Offshoring
336
382
Costs of Offshoring
17.5
17.6
382
Reshoring 383
Global Sourcing
384
Supply Base Optimization
341
Spend Analysis
343
Single or Multiple Suppliers
341
Discussion Questions
Problems
17.2
\
Operating an MRP System
Learning Enrichment
pourcin
eae
17.1
Importance of Sourcing
333
15.5
PART
Chapter 17
MRP System Elements
332
Master Scheduling 332
Solved Problem
344
17.7
The Purchasing Cycle
347
347
385
385
Total Number of Suppliers
SUPPLY CHAIN DECISIONS
385
386
387
Internal User-Buyer Interface
387
Sourcing Make-Buy Decision
Find Suppliers
388
388
Supplier Selection
388
Supplier Relationship Management
Chapter
16
Supply Chain Management
17.8
17.9
348
Challenges Facing Purchasing
Key Points and Terms
391
Learning Enrichment
16.1
16.2
374
326
327
Bill of Materials (BOM)
15.8
Discussion Questions and Problems
369
Supply Chain and Supply Chain
Management
349
Measuring Supply Chain Performance
351
°
°
;
2
qe
bs
°
°
© 6
ee
o
° %
a
°°
rm bs o.
$
°
392
Solved Problems 393
Discussion Questions 394
Problems 394
So
o
;
eo ©
bs
°
ig ;
°
788
°
389
389
Chapter 18
Global Logistics
18.1
18.2
oO
ed
°
=
°
°
&
ss
5
Capacity and Scheduling
397
Best Homes, Inc.: Forecasting
463
Polaris Industries Inc.: Global Plant
Location
465
Role of Logistics in Supply Chain
Management
398
Transportation
400
Transportation Economics
400
Lawn King, Inc.: Sales and Operations
Planning 470
Modes of Transportation
401
Selecting the Transportation Mode
18.3
18.4
Inventory
404
Distribution Centers and Warehousing
Logistics Networks
408
Location
Consolidated Electric: Inventory
Control
474
405
Southern Toro Distributor, Inc.
408
ToysPlus, Inc.: MRP
Number of Warehouses (Distribution
Centers)
18.5
18.6
18.7
18.8
411
Global Logistics
412
Third-Party Logistics Providers
Logistics Strategy
416
Key Pointsand Terms
418
Discussion Questions
414
Amazon Revolutionizes Supply Chain
Management
489
Altimus Brands: Managing Procurement
Risk 496
Murphy Warehouse Company: Sustainable
Logistics 499
ShelterBox: A Decade of Disaster
Relief 503
421
421
PART SEVEN
CASE STUDIES
Introduction
423
Operations Strategy at BYD of China,
Electrifying the World’s Automotive
Market
424
Early Supplier Integration for John Deere
Skid-Steer Loader
430
APPENDIXES
A _ Areas Under the Standard Normal
Probability Distribution
507
B
Random Number Table 508
INDEX
509
ACRONYMS
519
Process Design
Eastern Gear, Inc.: Job Shop 432
Sage Hill Inn Above Onion Creek: Focusing
on Service Process and Quality 435
U.S. Stroller: Lean
439
The Westerville Physician Practice: Value
Stream Mapping
445
Quality
Mayo Clinic and the Path to Quality 449
Toledo Custom Manufacturing: Quality
Control
455
The Evolution of Lean Six Sigma at
3M, Inc.
457
479
485
Supply Chain
Learning Enrichment
419
Solved Problems
419
Problems
Contents
°
Online Technical Chapters
Technical Chapters available in the
Instructor’s Resource Library
in Connect
Waiting Lines
Simulation
Transportation Method
Linear Programming
soxiii ||
WE
awe
=.
4.5
Economic
0k merlatanty= bras yiizaqs
ie
eeptin
: SAM S8inof yd
14.
14.8
Pehaa renin Nem iag a wlo4
Gang Pand' Q Systrags "hOBRIOF
a4
nee en BRSO BHRISHLE™ SiS iFtet
14.10°
14M
ABC Classificahonpeg ing:
Key Poivits and Fern
313
Leaming Exriciunear
3I2
:
Sid
yioinevnl
CothS Hit Sisal barebilozmD_
Discussion Questing», 2 ‘Tourn
ery’ SHYSomdAueil oT erraddinoe
Supe om SLAF Aaa
Chapter 15
nied vigque
Mater ala Syqiie menial GAs Rese!
ERP
323
O3h
JsomivesnkM
usfran rbpmtl cumitlA
MRP versus Order Punt
tyadalt 124
ging? PNAC
WEE
hey
Chenier
Works 327 -S@P zoimigo.!
15.4 anaes 30ohiah mg@railon2
SS
Maiwt
Scheduling
60¢
Sonrrane
7.9
VW
oilodt
Bill af Miskericils | ROAM)
i33
Javentor( Revords
CSUZIGKATL
335
Tame! Hrebaet? ddtiobaU enna
5
5.6
%S.7.
15.8
. 72
A
=a
The Successful MRP Syste
:
. Enterprise Resource Piannin ‘eoe" can
ae
3 im wUNce i SORA
Sourcing Goals 376
Inarniren ursigsounce?
Ternw mae Aah {YOH
vr
ihonnrge, of Optom
VM
”
7.6
nison,
eae
ape
ws
ed
337
Ary Points and
>
Vesdeuatia gesimi
«of Orem
Por SOR, aditudinteid yiilidadort
Sey 80@" GdeP edi
mobnust
Operalingan MRP Sysem
$55
Systems
i oe
Suna
os
lesinslast onilaQ
Prag
Peiciuast
a
ot att
oidnliave mated’) lsoinstosT
vind. 1 aviwoe st2ahritent
oan) ai
PART
el
Sik
7
347
vontt anivisw
Iv DEC ISTO Weelunte
bolisé sopahogensy]
hap!
youninesgord tesnit
sup
€ hain
Vaoeeerient
4h
6.1
supply
Chain
Supply
Chins
aed
lupptier ,
Mr po he
clitpartip Masaremen
a
PART
Introduction
1.
Introduction to Operations
2.
Operations and Supply Chain Strategy
3.
Product Design
The introductory part ofthis text provides an overview of operations management
in the supply chain. In Chapter 1 students gain an appreciation for the importance
to the firm of decisions made in the operations function and its associated supply
chain. In Chapter 2 the need for strategy to guide all decision making is emphasized.
In Chapter 3 new product design is treated as a cross-functional decision responsibility that precedes the production and delivery of goods or services. @
Co lal (NAP
a je IR
Introduction to
Operations
2
LEARNING
OBJ ECTIVES
ee
ee
a
101.3
ee
WNieu reading this chapter you should be able to:
ee
ae
LO1.2
ee
Review the role of operations in the firm and the economy.
Describe the five main decisions made py operant and si ipply cr managers.
LO1.4
Explain the nature of cross-functional decision making with operations.
101.5
Describe typical inputs and Dabiic
LO1.6
Analyze trends in operations and supply chain management.
yaaa: trans
i
The operations and supply chain management field deals with the production of goods
and services and management of the associated supply chain. Every day we come in
contact with an abundant array of goods and services, all of which are produced by
the operations function within the firm. Without management of operations and its
associated supply chain, a modern industrialized society cannot exist. The operations
function is the engine that creates goods and services for the firm and in aggregate the
global economy.
Supply chains are critical in supporting operations within the firm. Supply chains consist of a network of organizations outside the firm that supply the materials and services
to the firm and distribute the product or service to the ultimate customer. A global supply
chain engages in all the activities and processes needed to plan, source, make, deliver, and
return or dispose of a set of products or services. Managing the supply chain is essential in
addition to managing internal operations of the firm.
a
i
HI
o
°
“5
°
Oo
9
°
°
°
°
°
°
~
9
Chapter 1
P
SS
=
supply chain. This means we take a supply chain perspective to traditional operations. We take ideas from
Operations management within the firm and combine
them with a view of the entire supply chain.
At first glance it may appear that service operations
have little in. common with manufacturing operations.
However, the unifying feature of these operations is that
both can be viewed as transformation processes inside
organizations that are themselves embedded in supply
chains. In manufacturing, inputs of materials, energy,
;
;
eel
labor, and capital are transformed into finished goods
_
for customers. In service operations, the same types of
|
Sr
| 3ie
This book deals with operations management in the
——~
—
Introduction to Operations
|
inputs are transformed into services, for example, sur-
_—s
egeries in hospitals. Managing transformation processes
in an efficient and effective manner
is the task of the
operations manager in any organization.
Most Western economies have shifted dramatically from the production of goods to the production
of services. It may come as a surprise that more than
80 percent of the U.S. workforce is employed in service
industries.' Even though the preponderance of employment is in the service sector, manufacturing remains
important to provide goods needed for export and internal consumption. Because of the importance of both
service and manufacturing operations, they are treated
on an equal basis in this text.
In the past when the field was primarily related to
manufacturing,
operations management was called proApple manages a complex supply chain and operations
‘
across the globe.
duction management. Later, the name was expanded to
Jill Braaten/McGraw-Hill Education
operations management to include both manufacturing
and service industries. Now it has been expanded again
to operations and supply chain management to include not only operations, but also its
associated supply chain.
1.1.
DEFINITION OF OPERATIONS AND SUPPLY
CHAIN MANAGEMENT
LO1.1 Define
operations and
supply chain
management.
Operations management is defined as managing the production of goods and services.
The focus is on production within an organization, for example, production within the
four walls of a factory or within multiple factories in a company. Similarly, production of
services is treated as within individual service locations or across multiple locations within
a single company.
Operations management focuses on decisions
for the internal production of the firm’s
products or services.
' U.S. Census Bureau, Statistical
Abstract of the U.S., Washington, D.C. 2019 ed.
ni4
Part One
Jntroduction
FIGURE 1.1
A typical supply chain.
Suppliers
Suppliers
Factories
Warehouses
Retail
Customers
Supply chain management deals with managing the flow of materials, information,
and money across multiple organizations from the suppliers to operations to distribution
to the final customer, along with reverse flows. The entire supply chain is included from
the raw materials through suppliers, factories, warehouses,
and retailers to the ultimate
customer as shown in Figure 1.1. Reverse flows also occur in the supply chain for returned
products, recycled products, and information.
Operations and supply chain management deals with the sourcing, production and distribution of the product or service along with managing the relationships with supply chain partners.
This definition expands the notion of operations and reflects the role of operations in
the supply chain. It adds sourcing and distribution to the traditional definition of operations,
and it adds managing the relationships/interactions with supply chain partners. Sourcing,
also called purchasing, works with manufacturing and service suppliers of the firm and is
part of the larger definition of operations and supply chain management. Distribution, also
called logistics, is concerned with transporting materials into operations and taking the
output of operations to the ultimate customer. Operations and supply chain management
must be concerned with managing not only the internal operations of the firm, but also the
relationships and processes shared with supply chain partners along the supply chain.
This text brings together two previously separate fields of operations management and
supply chain management into what is called operations and supply chain management.
While operations management is concerned with the internal operations of the firm, supply chain management adds external relationships with other firms.
Moreover, every organization along the supply chain needs to manage its own operations
together with the relationship with the rest of the supply chain partners. Thus operations
management appears in the suppliers, the factories, the wholesalers, and the retail companies,
since each of them is producing a product or service that is passed along the supply chain.
1.2
LO1.2
THE ROLE OF OPERATIONS AND SUPPLY CHAIN MANAGEMENT
Review the
role of operations
in the firm and the
economy
All countries are dependent on economic growth as measured by their GDP (Gross
Domestic Product). GDP is the monetary value of all the goods and services produced
within a country’s borders. GDP growth leads not only to prosperity of the country, but
income growth for the population, since income is closely related to GDP per capita.
Chapter 1
Introduction to Operations
a)
Productivity is the amount of output from a given amount of inputs or vice versa. It is
one factor that leads to economic growth and profitability of the firm. More output over
time from the same resources increases GDP and allows people to get more of what they
want beyond mere survival. Operations and supply chain management plays a central role
in achieving GDP growth and productivity not only for individual companies but, in aggregate, for the entire economy of a country.
Productivity is output divided by inputs, all in constant dollars. Since price changes
should not affect productivity, constant dollars are used.
ae
output
Productivity = fs
eR
capital + labor
If the same output can be achieved by less capital and labor, productivity will improve. Vice
versa, more Output achieved from the same levels of capital and labor will also improve
productivity. Productivity ratios can be calculated at the firm, industry, and national levels.
Henry Ford believed in producing a reliable automobile at the lowest possible cost to
be affordable for all Americans. He represented the epitome of productivity improvement
by using better production line design and labor training to produce more output at lower
costs, illustrating the power of production in improving productivity and profits, and providing higher wages to employees. This story has been repeated thousands of time in many
industries, even to the present time. See the Operations Leader box for how Dell drives
productivity and value for its customers.
How are productivity growth and firm profitability achieved? It is only through
the creativity, imagination, and innovation of operations and supply chain employees,
Dell Delivers Productivity and Value
In 1984 Michael Dell founded Dell Computer Corporation
Orders for products, once taken, are assembled in one of
with $1000
Dell’s factories and often shipped to customers or retail
stores within days, with the factories carrying very little
finished goods inventory.
in start-up capital and a business
model
to
sell custom-configured personal computers directly to customers while passing along cost
savings
to customers
out the middlemen.
by cutting
The company
offers a range of products beyond
~
_ personal desktop and mobile com-
~~
Pe3k/Shutterstock
ing and
imaging
puting products; servers, storage,
and
networking
products;
products;
electronics
and accessories;
print-
enhanced
business and consumer services; and business solutions. Nearly half of Dell’s revenue comes from
In addition to the importance of the operations funcat Dell, sourcing and logistics activities are critical. Sourcing managers source the many components
required to manufacture Dell products, and logistics
managers handle the global movement of components
and finished goods to satisfy customer demand. Managing Dell’s fast and rapidly changing supply chain is a challenging task that they perform well.
tion
Dell today is pursuing environmentally friendly best
practices: Its global headquarters campus is now pow-
outside of the U.S.
A key to Dell's strategy is its customer-driven
approach to innovation. This approach signals a commit-
ered
ment to delivering
emissions;
new
products and services that are
by 100 percent green
energy;
its desk computer
systems have been designed to reduce carbon dioxide
Dell was
the first computer
manufacturer
offer free computer
while
improving
how
Dell pioneered
and its “Plant a Tree for Me” and “Plant a Forest for Me”
programs have planted over 600,000 trees.
customer
productivity.
This
approach
the direct-selling
system
explains
to allow
recycling to customers
to
valued by customers and that address customer needs
orders to be placed over the Internet or over
the phone and, since 2007, through select retail outlets.
Source: Adapted from www.dell.com, 2019.
worldwide;
is
| Part One
Introduction
managers, and executives. This can be done through innovative product or service design,
but also through continuous improvement of the production process and supply chain.
A large component of this is achieved through automation, but also the imagination of
designers and operations managers for both product and process improvements. It is the
responsibility of operations and supply chain managers to improve profitability and pro-
ductivity for the firm.
Below is a simple example of a computation of labor productivity in a firm.”
Example
For a firm with the following labor and output numbers, calculate the rate of labor productivity change assuming annual labor inflation of 3 percent and annual output (sales)
inflation of 2 percent.
Year 1
Year 2
Annual Inflation
Output (sales) $million
$56.7
$64.8
2%
Labor (payroll) $million
$23.4
$26.6
3%
Labor productivity year 1 =
Labor productivity year 2 =
Output year 1
56.7
Labor year 1 F284
= 2.42
Deflated output year2 _ 64.8(-98)
Deflated labor year 2 r 26.6(.97)
= 2.46
2.46
Change in productivity = aD = 1.016 which is a 1.6% increase
Notice, both the output and input in year 2 have been adjusted for inflation between year 1
and year 2. The productivity increase achieved by operations and supply chain managers
in one year is 1.6 percent.
Productivity improvement is not the only way to increase GDP and prosperity in a country. Just being more efficient is not enough, if there is no longer demand for the product.
Therefore, GDP growth requires innovation and new product development to meet the
evolving needs of markets. As a result, productivity improvement and innovation are two
of the most important factors that affect GDP growth, firm output, and firm profitability.
Operations and supply chain managers have a key role in introducing new products and
achieving scale-up of production. They also participate in product design teams consisting of design, marketing, and finance managers. In this way they contribute to both new
product innovation and productivity improvement for existing products and ultimately the
profitability of the firm and GDP growth.
1.3
WHY STUDY OPERATIONS AND SUPPLY CHAIN MANAGEMENT?
For students majoring in operations and supply chain management this will be an introductory course to the subject. This major leads to challenging and interesting jobs both
in domestic and international industries ranging from entry level to middle management
For simplicity we consider only labor inputs for the partial labor productivity ratio. A total productivity
ratio would include both capital and labor, and perhaps energy and materials inputs, as well.
Chapter 1
eae
ee
Introduction to Operations
7
RUN
=| OPERATIONS LEADER
|
Careers in Operations and Supply Chain from Monster.com
SUPPLY
CHAIN
ANALYST
all operations functions associated with branches and cen-
PayPal, owned by online shopping site eBay, is hiring a
supply chain management professional responsible for
end-to-end support for PayPal’s new Here product. The
job requires international travel
to manufacturing and distribution
sites.
Responsibilities
include
tral operations. This individual will participate in the development of strategic implementation plans and related
objectives. Candidates must have strong communication
skills and
acknowledge
customer members
CONTINUOUS
ment,
PLANT
reviewing
time
and
on
budget;
relationship
with
and mission.
product and distribution manage-
on
the important
in supporting the credit union’s vision
IMPROVEMENT
LEAD
inventory reports with
supply partners; arranging freight
shipments globally; and coordi-
ConAgra Foods seeks a partner to roll out a system
@Stablishing a zero-loss manufacturing culture. Coordi-
nating and collaborating with internal groups within Pay-
Mating with the Plant Manager, this Plant Lead executes
Pal and eBay. The job description also requests “maniacal
attention to detail.”
Plans for sustainability, develops and maintains training
and tracking standards, and coaches sites on improve-
NetPics/Alamy Stock Photo
ment
BUSINESS
METRICS/ANALYTICS
SUPPLY CHAIN ANALYST
Cardinal Health is seeking an analyst to develop, quantify, and
evaluate
the
transformation
of internal
and
external information into business intelligence. Qualified
candidates will demonstrate knowledge of concepts and
principles of business metrics and analytical techniques/
tools. The position requires listening to internal/external
customers’ needs and proactively providing them a quality experience through effective communication.
VICE
PRESIDENT
Envista
Credit
Union
OF
OPERATIONS
is seeking
an
executive
methodologies.
This
position
serves
as
a
key
development role for a future Plant Manager.
MATERIALS SOURCING
MANAGER
Herbalife,
a direct-sales nutrition company, is hiring a
sourcing manager for global spending of
$200 million on raw materials. Responsibilities include
reducing raw materials costs yearly, analyzing market
intelligence for trends in commodity markets, and making strategic recommendations to senior management
for each category of raw materials. This job also requires
senior-level
maintaining appropriate inventory levels and developing
strategic supplier relationships.
whose
responsibilities include organizing, planning, and directing
Source: Abstracted from www.monster.com.
and top management positions. Some of these positions in both service and manufacturing
industries are illustrated in the Operations Leader box on careers in operations and supply
chain management.
For non-majors this course is important to gain an understanding of what operations
does and its interactions with other functions within the firm and its supply chain. Every
decision is cross-functional in nature.* You will be working with operations and need to
understand it no matter what career path you choose. The organization in which someone
works only with people from his or her own function does not exist. That is why we take a
cross-functional perspective in this text, so the content is useful to the majority of students
who are not majors.
3 The “handshake” symbol in the margin identifies a point of cross-functional emphasis and is designed
to illustrate that the various functions must work together for an organization to be successful and thrive.
nis
|
Part One
Introduction
As you study operations, you will find that many of the ideas, techniques, and principles can be applied across the business, not just in operations. For example, all work is
accomplished through a process (or sequence of steps). The principles of process thinking found in this text can be applied to all parts of business. Toyota, for example, uses
lean thinking to improve processes in human resources, accounting, finance, information
systems, and even the legal department. Many students find that the ideas learned in this
course can be applied to their own department, career, or functional responsibilities and
are useful to non-majors.
Operations and supply chain management is an exciting and challenging field of study.
The ideas that you learn are both qualitative and quantitative, and both are essential to
good management practice. You are embarking on a journey that is interesting and useful
no matter what career you choose!
There are three aspects of operations and supply chain management
that require
elaboration:
1. Decisions. Since managers make decisions, it is natural to focus on decision making as a central theme in operations. Within the broader context of supply chain,
this decision focus provides a basis for identifying major decision types. In this text,
we specify the five major decision responsibilities of operations and supply chain
management as process, quality, capacity, inventory, and supply chain. These decisions provide the framework for organizing the text and describing what operations
and supply chain managers do. We will discuss these decisions in greater detail in
subsequent chapters.
2. Function. Operations is a major function in any organization, along with marketing and finance. In a manufacturing company, the operations function typically is
called the manufacturing or production department. In service organizations, the
operations function may be called the operations department or some name peculiar
to the particular industry (e.g., the policy service department in insurance companies). In general, the generic term “operations” refers to the function that produces
and delivers goods or services. While separating operations out in this manner is
useful for analyzing decision making and assigning responsibilities, we must also
integrate the business by considering the cross-functional nature of decision making
in the firm.
3. Process. Operations managers plan and control the transformation process and its
interfaces in organizations as well as across the supply chain. This process view is a
powerful basis for the design and analysis of operations in an organization and across
the supply chain. Using the process (or systems) view, we consider operations and supply chain managers as designers of the conversion process in the firm. But the process
view also provides important insights for the management of productive processes in
functional areas outside the operations function. For example, a sales office may be
viewed as a production process with inputs, transformation, and outputs. The same is
true for an accounts payable office and for a loan office in a bank. In terms of the process view, operations management concepts have applicability beyond the functional
area of operations.
Since the field of operations and supply chain management can be defined by decisions, function, and processes, we will expand on these three elements in detail in
this chapter. But first we provide an example of the decisions that would be made
by operations and supply chain management in a typical company that makes and
markets pizzas.
Chapter
1.4
1
Introduction to Operations
| ole
DECISIONS AT PIZZA U.S.A.
LO1.3 Describe the
five main decisions
Pizza U.S.A., Inc., produces and markets pizzas on a national basis. The firm consists of
285 company-owned and franchised outlets (each called a store) in the U.S. The opera-
made by operations
_ tions function in this company exists at two levels: the corporate level and the level of the
and supply chain
managers.
individual store.
The major operations and supply chain decisions made by Pizza U.S.A. can be described
as follows:
Process
Corporate staff makes some of the process decisions, since uniformity across different
stores is desirable. They have developed a standard facility design that is sized to fit a particular location. Each store incorporates a limited menu with equipment that is designed
to produce pizza to customer orders. As pizzas are made, customers can watch the process
through a glass window; this provides entertainment for both children and adults as they
wait for their orders to be filled. Because this is a service facility, special care is taken to
make the layout attractive and convenient for the customers.
Within the design parameters established by the corporate operations staff, the store
managers seek to improve the process continually over time. This is done both by additional investment in the process and by the use of better methods and procedures, which
often are developed by the employees themselves. For example, a store might re-arrange its
layout to speed up the process of producing pizzas.
Quality
Certain standards for quality that all stores must follow have been set by the corporate
staff. The standards include procedures to maintain service quality and ensure the quality
and food safety of the pizzas served. While perceptions of service quality may differ by
customer, the quality of the pizzas can be specified more exactly by using criteria such as
temperature at serving time and the amount of raw materials used in relation to standards,
among others. Service-quality measures include courtesy, cleanliness, speed of service,
and a friendly atmosphere. Service quality is monitored by store manager observation,
comment cards, and occasional random surveys. Each Pizza U.S.A. store manager must
carefully monitor quality internally and with suppliers to make sure that it meets company standards. All
employees are responsible for the quality of their work
to ensure that service quality and food quality are meeting the standards of the company.
Capacity
Pizza U.S.A. satisfies its customers by carefully managing the
four key decision areas in operations.
Steve Mason/Getty Images
Decisions about capacity determine the maximum level
of output of pizzas. The capacity available at any point
in time is determined by the availability of equipment
and labor inputs for the pizza-making process at that
time. First, when the initial location and process decisions are made, the corporate staff determines the physical capacity of each facility. Individual store managers
then plan for annual, monthly, and daily fluctuations
in capacity within the available physical facility. During peak periods, they may employ part-time help, and
H] |10 | Part One
Introduction
advertising is used in an attempt to raise demand during slack periods. In the short run,
individual personnel are scheduled in shifts to meet demand during store hours.
Inventory
Each store manager buys the ingredients required to make the recipes provided by corporate staff. The store managers decide how much flour, tomato paste, sausage, and other
ingredients to order and when to place orders. Store operators must carefully integrate
sourcing and inventory decisions to control the flow of materials in relation to capacity.
For example, they do not want to purchase ingredients for more pizzas than they have the
capacity to bake. They also do not want to run out of food during peak periods or waste
food when demand is low.
Supply Chain
The supply chain decisions consist of sourcing and logistics. Sourcing is done by the corporate office. They select the specific suppliers for all inputs, negotiate prices, write contracts, and issue blanket purchase orders that stores use to order individual ingredients
and items as they need them. The orders are then fulfilled by the suppliers, and a logistics
provider ensures the orders are delivered on time. Logistics is handled by a third-party
provider who secures transportation and uses its distribution centers to make deliveries to
Pizza U.S.A. stores.
1.5
OPERATIONS DECISIONS IN THE SUPPLY CHAIN—A FRAMEWORK
The five decision groupings showcased in the Pizza U.S.A. example provide a
framework for understanding the various decisions made by operations and supply
chain managers. Although many different frameworks are possible, the primary one
used here is a conceptual scheme for grouping decisions according to decision responsibilities. The five key decision areas—process, quality, capacity, inventory, and supply chain—encompass what operations and supply chain managers do. This novel and
useful decision framework is shown in Figure 1.2. Notice how the five decision areas
in the figure apply not only to the firm, but to the supplier and the distributor, since
all three must make the same types of decisions in managing their own operations and
coordinating those decisions across the supply chain. In Table 1.1, examples are given
of key decisions in each area.
FIGURE
1.2
Decision-making framework for operations in the supply chain.
The Firm
Human
Resources
oak
Finance
Supply
Chain
Decisions
Distributor
Supply
Chain
Decisions
Marketing
Information
Systems
Accounting
Chapter 1
TABLE 1.1
;
Operations and
Supply Chain
Devcon
Gl
ged ne tne
Ride
ipLond
PLuae ATycib
elie
/ntroduction to Operations
Li eeait
TRIS Duets
Geen
Decisions
Examples of Decisions
1. Process
+
*
+
+
*
2. Quality
* What should the quality standards be?
* How can quality be controlled and improved?
* What statistical approaches should be used (e.g., control charts and
Six Sigma)?
* How should the suppliers and customers be involved in quality? |
3. Capacity
*
*
*
*
What is the facility strategy for size, location, and timing?
How should Sales and Operations Planning be implemented?
How should variable demand be handled with capacity adjustments?
What priority rule should be used for scheduling?
4. Inventory
*
+
*
*
How much inventory should be held?
What should the order size and reorder frequency be?
Who should hold the inventory?
How can the inventories of suppliers and customers be coordinated?
5. Supply Chain
*
*
*
*
What suppliers should be used for products and services?
How should sourcing be conducted and evaluated?
What form of transportation should be used?
How should warehouses be used to allow economic flow of materials?
Framework
14)
What type of eee ones be selected?
How should the service delivery system be designed?
How should material and customer flows be managed?
What principles of lean systems should be deployed?
How should environmental and global goals be met?
Careful attention to the five decision areas in the framework is the key to the successful
management of operations and the associated supply chain. Indeed, well-managed operations and its supply chain can be defined in terms of this decision framework. If decisions
in each of the five groupings support the strategy of the firm, provide value, and are well
integrated with the other functions of the organization, the operations function and its
associated supply chain can be considered well managed.
Each major section of this text is devoted to one of the five decision categories.* The
framework thus provides an integrating mechanism for the text that covers both the decisions faced by operations and supply chain managers as well as the cross-functional issues
that must be considered.
The five decisions areas of operations and supply chain can be compared to the 4 Ps
of marketing: product, price, place, and promotion. These four Ps in the marketing mix
are the tools or decision types that marketing managers can use to influence demand. In a
similar way the five decision types of operations and supply chain can be used to influence
supply and inust be compatible and consistent with the 4 Ps of marketing. They are just two
sides of the same coin, one influencing demand and the other supply.
Analytics is the analysis of data to make better decisions. Analytics uses many techniques for the analysis including those from operations research, statistics, data sciences,
and computer science. The analysis can use either big data from massive databases or
small data depending on the application. Analytics can be descriptive, predictive, or prescriptive in nature. A descriptive analysis typically summarizes the present situation from
data. The data can be used to go one step further and predict what will happen in the future.
Prescriptive analytics typically uses mathematical models to find an optimal or best decision. Analytics are used in operations and supply chains for a variety of decisions, including quality control, forecasting, capacity, scheduling, inventory, logistics, and sourcing.
4 Students have called these five categories QPICS, Quality, Process, Inventory, Capacity, and Supply
chain, pronounced “Q-PICS.”
mM] 42 | Part One
Introduction
Throughout the text, best practices are presented. Additionally, discussion and examples
of firms in which the best practice is not the best for their particular situation are included.
These contingencies, situations, or conditions that require different solutions offer a more
nuanced view of operations decision making. For example, successful implementation of
a new method such as lean or Six Sigma is contingent on top management support. Similarly, the “best” forecasting tools and concepts depend on the availability of data. If there
was a single best practice that works for all firms, then operations would not be the challenging function to manage that it is. Therefore, by offering insight into specific conditions
in which best practices may not be best, the text addresses the various contingencies or
prerequisites or situations that need to be considered.
1.6
CROSS-FUNCTIONAL
DECISION MAKING
The operations function is a critical element in every business. No business can survive
without good decisions being made by operations managers. The operations function is
cross-functional
one of the three primary functions in an organization, along with marketing and finance.
decision making with
In addition, an organization has supporting functions that include human resources, inforoperations.
mation systems, and accounting. Some organizations also have
separate sourcing and logistics functions that support operations. In
others, the operations, sourcing, and logistics functions are joined
together to become the supply chain function.
Functional areas are concerned with a particular focus of
responsibility or decision making in an organization. The marketing function is typically responsible for creating demand and generating sales revenue; the operations function is responsible for
the production and distribution of goods or services (generating
supply); and finance is responsible for the acquisition and allocation of capital. Within for-profit businesses, functional areas tend
to be closely associated with organizational departments because
Managerial decision making is cross-functional
businesses
typically are organized on a functional basis. Supportin nature.
ing
functions
are essential to provide staff support to the three
ammentorp/123RF
primary functions.
Every function must be concerned not only with its own decision responsibilities but also with integrating decisions with other functions. The five
areas of operations and supply chain decisions, for example, cannot be made separately; they must be carefully integrated with one another and, equally important, with
decisions made in marketing, finance, and other parts of the organization. In the Pizza
U.S.A. example, if marketing decides to change the price of pizza, this is likely to
affect sales and change the capacity needs of operations as well as the amount of ingredients (materials) used. Also, if finance cannot raise the necessary capital, operations
may have to redesign the process to require less capital or manage pizza-related inventories more efficiently. This in turn may affect the response time to serve customers,
costs, and so on.
Decision making is therefore highly interactive and systemic in nature. Unfortunately,
functional silos have developed in many organizations and impede cross-functional decision making. As a result, the overall organization suffers due to an emphasis on functional
prerogatives. In a similar way, the five decisions need to be coordinated externally with
supply chain partners. If capacity of the firm is increased, the capacity of suppliers and
distributors must also be increased. The same can be said about coordination of decisions
about inventory, quality, sourcing, and logistics.
LO1.4
Explain
the nature of
Chapter 1
Introduction to Operations
13)
But some companies are different. Texas Instruments, for example, has been a leader in
fostering cross-functional integration. Some companies have done a very good job of crossfunctional management. They do this by forming cross-functional teams for new product
introductions and for day-to-day improvement. Each member of the team is trained in common methodologies, and the team is given responsibility for achieving its own goals. Some
of the key cross-functional decision-making relationships are shown in Table 1.2.
1.7
LO1.5
OPERATIONS AS A PROCESS
Describe
typical inputs
and outputs of
an operations
transformation
system.
Operations can be defined as a transformation system (or process) that converts inputs
into outputs. Inputs to the system include energy, materials, labor, capital, and information
(see Figure 1.3). Process technology is then used to convert inputs into outputs. The process technology is the methods, procedures, and equipment used to transform materials or
inputs into products or services.
Viewing operations as a process is very useful in unifying seemingly different operations from different industries. For example, the transformation process in manufacturing
is one of material conversion from raw materials into finished products. When an automobile is produced, steel, plastics, aluminum, cloth, and many other materials are transformed
into parts that are then assembled into the finished automobile. Labor is required to operate
Key Decision Area
Marketing
Market segment and needs
Market size (volume)
Quality design and quality management
Type of process selected (assembly line, batch, or project) and capacity required
Distribution channels
Pricing
New product introduction
Quality, capacity, and inventory
Cross-functional teams
Finance and Accounting
Availability of capital
Efficiency of conversion process
Inventory levels, degree of automation, process type selected, and capacity
Process type selection, process flows, value-added determination and sourcing
Net present value and cash flow
Process costing or job costing
Measurement of operations
Automation, inventory, and capacity
Type of process selected
Costing systems used
Human Resources
Skill level of employees
Number of employees and part-time or
full-time employment
Process type selected and automation
Capacity and scheduling decisions
Training of employees
Job design
Teamwork
Information Systems
Determination of user needs
Design of information systems
Software development
Hardware acquisition
Inventory levels and logistics
Quality improvement and skills
Process and technology choice
Cross-functional decisions in operations
Systems should support all users in operations
Systems should help streamline operations and support all analytics and
decisions in operations
Software is needed for capacity, quality, inventory, scheduling, and supply
chain decisions
Hardware is needed to support automation decisions in operations and to
run software
|
14 | Part One = Introduction
EXTERNAL BUSINESS
FIGURE 1.3
An operation as a
productive system.
ENVIRONMENT
OPERATIONS MANAGEMENT
OUTPUTS
INPUTS
Energy
Materials
Transformation
(conversion)
process
Goods or services
Information
BEBBBSSEGREREBRRRREeee
ie
oe
tee
ee
Ce
ABC
RRECAAEEAEWEEUBHE®
Feedback information for
control of process inputs
and process technology
NATURAL ENVIRONMENT
and maintain the equipment, and energy and information are also required to produce the
finished automobile.
In service industries a transformation process is also used to transform inputs into service outputs. For example, airlines use capital inputs of aircraft and equipment and human
inputs of pilots, flight attendants, and support personnel to produce safe, reliable, fast, and
efficient transportation. Transformations of many different types occur in all industries, as
indicated in Table 1.3. By studying these different types of transformation processes, you
can learn a great deal about how to analyze and manage any operation.
Operations as a process provides a basis for seeing an entire business as a system of
interconnected processes. This makes it possible to analyze an organization and improve
it from a process point of view. All work, whether in finance, marketing, accounting, or
other functions, is accomplished by processes. For example, financial analysis of a stock,
closing the books at the end of the year, or conducting market research are each conducted
by carrying out an appropriate process. Thus, process principles and tools can be applied
in every function in a business.
All of these processes and systems interact with their internal and external
environments. We have indicated the nature of internal interaction through crossfunctional decision making. Interaction with the external environment occurs through the
economic, physical, social, and political environment of operations. Examples include
TABLE 1.3
Examples of
Productive Systems
Operation
Inputs
Outputs
Bank
Tellers, staff, computer equipment,
facilities, and energy
Cooks, waiters, food, equipment,
facilities, and energy
Doctors, nurses, staff, equipment,
facilities, and energy
Faculty, staff, equipment, facilities,
energy, and knowledge
Equipment, facilities, labor, energy, and
raw materials
Financial services (loans, deposits,
safekeeping, etc.)
Meals, entertainment, and satisfied
customers
Health services and healthy
patients
Educated students, research, and
public service
Finished goods
Planes, facilities, pilots, flight atten-
Transportation from one location
to another
Restaurant
Hospital
University
Manufacturing plant
Airline
dants, maintenance people, labor,
and energy
Chapter 1
Introduction to Operations
415)
economic changes such as rising labor costs, social changes such as customer preference
for “green” products, and political changes such as regulations. Each of these can mean
that the operations function and associated supply chain will have to change the way it was
producing products and services.
Operations is surrounded by both internal and external environments and constantly
interacts with them. The interactive nature of these relationships makes it necessary to
constantly monitor the environment and make decisions related to corresponding changes
in operations and the supply chain when needed. In the fast-changing world of today’s
global business, constant change has become essential as a means of survival. Viewing
operations as a process or a constantly updating transformation system helps us understand
how operations and the supply chain cannot be insulated from changes in the environment
but rather must adapt to them.
1.8
TRENDS IN OPERATIONS AND SUPPLY CHAIN MANGEMENT
LO1.6 Analyze
trends in operations
Operations and supply chain managers face serveral opportunities and trends that will be
—_addressed repeatedly throughout this text. These trends make operations and supply chain
and supply chain
management.
management an exciting and interesting career for future leaders.
Sustainability
The focus on sustainability of the natural environment has been heightened in recent
years with concerns over global warming, water contamination, air pollution, and
so on. Organizations are increasingly being asked to produce and deliver products
or services while minimizing the negative impact on the global ecosystem and not
endangering the ability to meet the needs of future generations. See the Operations
Leader box titled “Sustainability in Interface Inc.’s Operations Transformation Process” for an example of one firm’s success in facing these issues. Operations and
supply chain partners have made tremendous strides in reducing pollution of the environment from air to ground to water, but there is still a long way to go. Operations
and its supply chain are going beyond environmental sustainability to include social
and economic sustainability: the so-called triple bottom line. Social sustainability
means hiring a diverse workforce, ethical practices, providing equal opportunity, and
safe working conditions, for example. Economic sustainability is making a sufficient
profit for firm survival into the future. Operations and their supply chains are finding they can reduce pollution, conserve resources, recycle products, and be socially
responsible to provide a sustainable world for future generations. Sustainability is an
opportunity that progressive operations and supply chain organizations are pursuing.
For more details, see the sustainability link in the Learning Enrichment box at the end
of this chapter.
Services
Operations concepts and ideas have been applied in service operations for years. Yet,
service operations lag behind manufacturing in applying the latest ideas in supply chain
management, Jean operations, and quality improvement. This represents a tremendous
opportunity to apply what is learned in this course. Also, service-specific ideas such as
service recovery, web-enabled service, and globalization of service still represent implementation challenges. Nevertheless, some leading service businesses do excel in operations including Walmart, Nordstrom, Starbucks, Amazon.com, FedEx, and Delta Airlines,
to name only a few. They excel by applying many of the operations concepts that are presented in this text.
i jis | Part One
Jntroduction
neees
|maaeee8885ccne
OPERATIONS LEADER
Sustainability in Interface Inc.’s Operations Transformation Process
With production on four con-
The
philosophy
of
sustainability is “meeting the needs
of the present without compromising the ability of future
tinents and offices in more than
100 countries, Interface Inc. is
the global leader in the design,
production,
generations to meet their own
needs.” Over the past 15 years,
carpet manufacturer Interface
Inc. has shifted its operations
toward this philosophy and triple bottom-line impacts: social,
People,
and Profit.
Following
production,
tomers
use
and
then
diverted from landfills to serve
as raw materials for new carpet
squares. They have achieved
a series of major milestones
cus-
the
dispose
It is not a perfect
system,
it still requires
as
some
European
at
manufacturing
facility in The Netherlands. As of
2015, the plant is operating with
of carpet products. Interface
Inc. set out to change this typical supply chain. Creating a
closed-loop supply chain, they
use their own post-consumer
waste (used carpet) as raw
material input to their produc-
than
133 million pounds of postconsumer carpet waste has been
Planet,
tion system.
sales of modu-
Interface Inc. reports more
environmental, and financial. Or,
in their words:
and
lar carpet squares. Since undertaking the goal of sustainability,
100 percent renewable
energy,
using virtually zero water in man-
ufacturing
processes
and
has
attained zero waste to landfill.
— A"caid Images/Alamy
newly
extracted
raw
Sourées Adapted tion Dave Gustashavrand Robert W:
materials, but they believe they are moving in‘the right
Hall, “From Lean to Green: Interface, Inc.” Target 24, no. 5
direction for achieving sustainability.
(2008), pp. 6—14 and interfaceglobal.com 2019.
Digital
Technologies
Emerging digital technologies that operations and supply chain managers are implementing include artificial intelligence, blockchain, 3D printing, and data analytics. Artificial
intelligence is used in manufacturing and services to perform complex tasks that require
human learning. Block chain technology is used to secure information that is passed along
the supply chain. 3D printing is used to rapidly make prototypes or custom products. All of
these technologies and analytics are discussed throughout this text.
Integration
A difficult opportunity and challenge facing all managers is cross-functional integration
of Decisions
within the organization. Some organizations are managing functions as separate departments with little integration across them. The best operations are now seeking increased
integration through the use of cross-functional teams, information systems, management
coordination, rotation of employees, and other methods of integration. Most of the implementation problems of new systems or new approaches can be traced to lack of crossfunctional internal cooperation. The same thing can be said about interorganizational
change in supply chains. Even when companies partner with their suppliers or customers
the partnerships are often not successful. Adequate information systems may also be lacking for supply chain integration.
Internally and
Externally
Chapter 1
Globalization
of Operations
Finally, the globalization of
Operations and supply chains is
and the Supply
3
a pervasive theme in business
today. One can hardly avoid
information on the accelerating nature of global business.
Strategies for operations and
its supply chain partners should
be formulated
with global
effects in mind. Even many
small businesses compete globally, sourcing or selling goods
and services in markets with
global
competitors.
Facility
location must be considered in
:
é
:
aa.
Chain
view of its global implications.
Introduction to Operations
17)
Coke is produced and sold globally.
— streetvu/Shutterstock
Technology can be transferred
rapidly across national borders. All decisions in operations and its associated supply chains
are affected by the global nature of business.
1.9
KEY POINTS AND TERMS
This text provides a broad overview of the challenging and dynamic field of operations
management and the supply chain. It stresses decision making in operations, its associated
supply chain, and the relationship of these decisions to other functions. The five major
decision categories—process, quality, capacity, inventory, and supply chain are the organizing framework for each of the five major sections in the text.
Key points emphasized in the chapter are these:
¢
Operations and its associated supply chain produces and delivers goods or services
deemed to be of value to customers in a global economy. Operations and supply chain
management is responsible for productivity, innovation and GDP growth in aggregate.
Without operations and supply chain management a firm, industry, and country cannot prosper.
¢
Operations and supply chain management focuses on decisions for the production,
sourcing, and delivery of the firm’s products and services. These decisions are intended
to maximize firm profitability and the value inherent in goods or services delivered to
customers throughout the entire supply chain.
¢
The supply chain is the network of manufacturing and service operations that supply
each other from raw materials through manufacturing to the ultimate customer. The
supply chain consists of the physical flow of materials, money, and information along
the entire chain of suppliers, production, and distribution and the reverse supply chain
of recycled and returned products and information. The supply chain connects many
different organizations.
¢
There are five key groupings of decisions in operations and supply chain management:
process, quality, capacity, inventory, and supply chain. These decisions need to utilize
analytics when appropriate and account for contingencies, or special situations, because
a best practice may not be best in all circumstances.
|
18 | Part One
Introduction
Key Terms
*
Operations and supply chain decisions are often cross-functional in nature. Decisions
may impact or be impacted by activities in other functions such as marketing and
finance. Often, cross-functional teams are formed to undertake complex decisions.
¢
We identify several opportunities and trends facing operations and supply chain managers that are emerging and will be important in the future. These are sustainability,
services, digital technologies, integration of decisions, and globalization of operations
and the supply chain.
Supply chain 2
Operations management 3
Supply chain management 4
Operations and supply chain
management 4
Gross Domestic Product
(GDP) 4
Productivity 5
Five major decision
responsibilities 8
Process 8
Quality 8
Capacity 8
Inventory 8
Supply chain 8
Analytics 11
Cross-functional decision
making 12
Transformation system 13
Internal and external
environments 14
Sustainability 15
Triple bottom line 15
Globalization 17
LEARNING
ENRICHMENT
What Is Operations Management?
https://youtu.be/IleEMOReAE2hk
Video
oyele
(for self-study —s
Supply Chain Management: A Force for Good
Video
Coca-Cola: Supply-Chain
https://youtu.be/UBSOiHUctrY
Video
PROX)
Sustainability
https://www.epa.gov/sustainability/learn-about-sustainability#what
Web Link
Globalization
Web Link
UE ELUMONeRCE
un)
https://youtu.be/BlOUhiOvrde
5:01
http://www.globalization101.org/what-is-globalization/
caer
areser
eer
SS
SS
SSS
SSS
SSS
SSS
SSS
SSS
SS
Discussion Questions
1. Why study operations management in the supply chain?
2. What is the difference between the terms “production
management” and “operations management’’?
3. What is the difference between operations management
and supply chain management?
4. How does the work of an operations manager differ
from the work of a marketing manager or a finance
manager? How are these functions similar?
. How is the operations management function related to
activities in human resources, information systems, and
accounting?
6. Describe the nature of operations management in the
following organizations. In doing this, first identify the
outputs of the organization and then use the five decision types to identify important operations decisions
and responsibilities.
a. A college library
b. A hotel
c. A small manufacturing firm
. For the organizations listed in question 6, describe the
inputs, transformation process, and outputs of the production system.
Chapter 1.
. Describe the decision-making view and the view of
operations as a process. Why are both views useful in
studying the field of operations management?
. Write a short paper on some of the challenges facing
operations management in the future. Use newspapers,
business magazines, or the Internet as your sources.
10. Review job postings from various sources for management positions that are available for operations management graduates. Summarize the responsibilities of these
positions.
. Describe how the view of operations as a process can
be applied to the following types of work:
a. Acquisition of another company.
b. Closing the books at the end of the year.
c. Marketing research for a new product.
d. Design of an information system.
e. Hiring a new employee.
What is the role of operations and supply chain management in national economic prosperity? How does it contribute to GDP, jobs, income, and society in general?
Introduction to Operations
| 19]
. Accompany has experienced the following changes in
sales, labor, and capital. Calculate the total productivity
percentage improvement from year | to year 2. Does
the total productivity measure make more sense in this
problem, since both labor and capital have been used?
Annual
Year1
Year2_
Inflation
Output (sales) $million
$260.5
$270.4
2%
Labor (payroll) $million
$110.4
$111.5
3%
Capital
$ 60.2
$ 61.0
1%
16. The Atlas company makes weight lifting equipment.
They are in the process of automating and have added
$5 million in capital equipment to their operations and
reduced the labor content. Has this resulted in increased
productivity based on the following numbers? What is
the productivity gain or loss as a percentage?
Year1
Year2_
Annual
Inflation
2%
What is the role of the operations function? How does it
contribute to profitability, return on investment, growth,
and other corporate objectives?
Output (sales) $million
$358.5
$361.4
Labor (payroll) $million
$110.4
$100.6
3%
For the following problem calculate the labor productivity improvement from year | to year 2.
Capital
$,,60!2e$ 65.2
0%
Annual
Year 1
Year 2
Inflation
Output (sales) $million
$103.4
$108.4
2%
Labor (payroll) $million
$ 19.6
$ 22.4
3%
Cobh
Ar PidiavEaR
Operations and Supply
Chain Strategy
Describe the elements of operations strategy and alignment with business and other
functional strategies.
ia op stations obj
LO2.6
Analyze two types of supply chain strategies.
aa .Illustrate how operations and supply ch incan become more sustainable.
LO2.1 Define
operations strategy.
There is an increasing awareness that operations and the supply chain contribute to the global
competitive position of a business and are not merely making a firm’s products or services.
This can be done by contributing distinctive capability (or competence) to the business and
continually improving the products, services, and processes. Operations strategies and decisions should fulfill the needs of the business and add competitive advantage to the firm. The
Operations Leader box on Southwest Airlines illustrates how operations adds competitive
advantage to its firm.
The operations function is a key value creator for the firm. Value is providing products
and services that customers want to buy at low cost. All functions of the firm must be
well coordinated for value to be created and competitive advantage to occur. The crossfunctional coordination of decision making is facilitated by an operations strategy that is
developed by a team of managers from across the entire business.
Chapter
2. Operations and Supply Chain Strategy
| 24)
OPERATIONS LEADER
Southwest Airlines
Southwest Airlines was founded
in 1971
as a regional
*
airline.It has achieved financial success with its 47th
consecutive year of profitability. In 2018 it reported
$3.5 billion in net income on $21 billion in revenue with
an annual return on invested capital of 25.9 percent.
It has achieved this amazing record through a consistent
business and operations strategy over its history.
The mission of Southwest Airlines is to provide the
highest quality customer service at the lowest cost
fares. This might have seemed impossible for a start-up
regional airline competing against existing large airlines
with massive economies of scale and purchasing power.
Southwest Airlines accomplished this by using an innovative operations strategy from the beginning with the
following features:
*
Purchasing only Boeing 737 aircraft. Operating a single
aircraft saved on maintenance training, spare parts stor-
age, and special prices from Boeing for aircraft. Today,
Southwest airlines operates 706 aircraft, all 737s.
e-| Southwest
—o
Markus Mainka/123RF
Flying out of regional secondary airports such as
Midway Field in Chicago and Love Field in Dallas
_where gate fees and operating costs are much lower,
and flights are more convenient and accessible for
customers.
*
Flying only point-to-point routes between pairs of cit-
ies avoiding large airport hubs and thus cutting turnaround time between flights to only 15 minutes with
very high aircraft utilization. High utilization of very
expensive capital investment is a key to their success.
*
Achieving high productivity from employees who
have significant profit sharing, training, excellent
working conditions, and low employee turnover.
+
Making flights fun for the aircrews and passengers
and giving passengers free bags when all other airlines charge a bag fee. This improves customer service and customer satisfaction.
Southwest Airlines has received numerous awards
for customer satisfaction including being ranked #1 by
the U.S. Department of Transportation, and receiving
the Best Customer Service Award by Freddie Awards
for frequent flyers. It has withstood fierce competition
from global airlines and other low-cost airlines such as
JetBlue. Southwest Airlines has achieved its success
by implementing a consistent operations strategy that
is well integrated with business strategy and marketing strategy.
The following definition of operations strategy is a starting point for our discussion:
Operations strategy is a consistent pattern ofdecisions for operations and the associated
supply chain that are linked to the business strategy and other functional strategies, leading
to a competitive advantage
for the firm.
This definition will be expanded throughout this chapter as a basis for guiding all decisions
that occur in operations and its supply chain with decisions in other functions.
We will use McDonald’s as an example in the next several sections to illustrate the
elements of an operations strategy. In 1955, Ray Kroc opened his first restaurant in Des
Plaines, Illinois, patterned after the McDonald brothers’ hamburger stand in California.
The McDonald’s service system was designed on the idea of a very limited menu and
fast production of standardized food and service with convenience and a low price. Never
before had customers been served food so fast in a clean and courteous environment. Using
a standard design for equipment, facilities, and employee training, the McDonald’s system
was replicated in many locations and rapidly expanded throughout the U.S. and then the
world.
|
22 | Part One
Jntroduction
McDonald’s is a
leading global service
firm with an operations
and supply chain
strategy.
Christopher Kerrigan/
McGraw-Hill Education
The McDonald’s system is a standardized service system designed to meet stringent specifications. Every detail of the system is designed to provide fast and efficient food and service.
McDonald’s has continuously adapted its service system and supply chain over the
years. For example, the menu has been expanded to offer many more food and beverage
items, but always within the capability of the existing restaurants. They have updated their
information systems in-operations, and responded to environmental challenges by replacing, for example, the foam boxes previously used for sandwiches with biodegradable paper
wrappers. In response to healthy food trends, they added salads, apple slices, and grilled
chicken. Nevertheless, McDonald’s still has its critics and sometimes is blamed for the
obesity of Americans and for having an adverse environmental impact.
McDonald’s is a global service firm. The operations strategy for global expansion has
been to replicate the service system design and supply chain in each country with minimum
modifications to the menu or processes. However, a few local international options are provided. For example, McDonald’s serves beer in Germany, McRice in Indonesia, soup in
Portugal, and salmon burgers in Japan. They have also extended their supply chain forward
by developing a franchise system that maintains strong control over the product and service.
Today, McDonald’s is the global leader in food service with more than 37,000 restaurants in 120 countries serving an average of 68 million customers each day. Next we will
describe the elements of operations strategy in detail and use McDonald’s to illustrate how
the elements of operations strategy support overall business strategy.
2.1
OPERATIONS STRATEGY MODEL
LO2.2 Describe
the elements of
operations strategy
and alignment with
business and other
functional strategies.
Operations strategy is a functional strategy along with the firm’s other functional strategies such as those of marketing, engineering, information systems, and human resources.
Since operations strategy is a functional strategy, it should be guided by the business
and corporate strategies shown in Figure 2.1. The four elements inside the dashed box—
mission, objectives, strategic decisions, and distinctive competence—are the heart of operations strategy. The other elements in the figure are inputs or outputs from the process of
Chapter 2.
FIGURE 2.1
Operations strategy
Corporate
process.
strategy
Operations and Supply Chain Strategy
| 23) a
Business strategy
Internal
analysis
ees
|eae
ate
engineering,
External
analysis
human resources, and
Consistent pattern
of decisions
developing operations strategy. The outcome of using the operations strategy is a consistent pattern of operations decisions that are well connected with the other functions in the
business and help provide a competitive advantage to the business.
Corporate
and Business
Strategy
Corporate strategy and business strategy are at the top of Figure 2.1. The corporate strategy defines the business that the company is pursuing. For example, Walt Disney Corporation considers itself in the business of “making people happy.” Disney Corporation
includes not only theme parks but the production of cartoons, movie production, merchandising, and a variety of entertainment-related businesses around the world.
Business strategy follows from the corporate strategy and defines how each particular
business will compete. Most large corporations have several different businesses, each competing in different market segments. Michael Porter describes three generic business strategies: differentiation, low cost, and focus. Differentiation is associated with a unique and
|
24 | Part One
Jntroduction
frequently innovative product or service,
while low cost is pursued in commodity
markets where the products or services are
imitative. Focus refers to the geographical or
product portfolio being narrow or broad in
nature. Focus can be combined with either a
differentiation or a low-cost strategy.
Operations
Mission
Every operation should have a mission that
is connected to the business strategy and is
coordinated with the other functional strategies. For example, if the business strategy is
differentiation through innovative products,
the operations mission should emphasize
new product introduction and flexibility to
adapt products to changing market needs.
Other business strategies lead to other
operations missions, such as low cost or
fast delivery. The operations mission is thus
derived from the particular business strategy selected by the business unit.
THE MAGIC KINGDOM. Disney Corporation is in
At McDonald’s, the operations mission _ the business of “making people happy.”
:
:
:
:
is to provide food and service quickly to
customers with consistent quality and low
cost in a clean and friendly environment.
Operations
Objectives
Phelan M. Ebenhack/AP Images
Operations objectives, sometimes called competitive priorities, are the second element of
operations strategy. The four common objectives of operations are cost, quality, delivery,
and flexibility. In certain situations, other objectives may be added, such as innovation,
safety, and sustainable operations. The objectives should be derived from the operations
mission, and they constitute a restatement of the mission in quantitative and measurable
terms. The objectives should be long-range-oriented (5 to 10 years) to be strategic in nature
and should be treated as goals.
Definitions of the four common operations objectives follow:
*
Cost is a measure of the resources used by operations, typically the unit cost of production or the cost of goods or services sold.
*
Quality is the conformance of the product or service to the customers’ requirements.
*
Delivery is providing the product or service quickly and on time.
*
Flexibility is the ability to rapidly change operations.
Table 2.1 shows some common measures of objectives that can be used to quantify
long-range operations performance. The four common objectives are listed along with a
fifth objective, sustainability, that is becoming more important to many firms. The objectives for five years into the future are compared to the current year and also to a current
world-class competitor. The comparison to a world-class competitor is for benchmarking
purposes and may indicate that operations is behind or ahead of the competition. However,
the objectives should be suited to the particular business, which will not necessarily exceed
the competition in every category.
At McDonald’s, each restaurant has specific objectives with respect to cost, quality, and
service times. These objectives are pursued using extensive standards and are frequently
Chapter
TABLE 2.1
Typical Operations
Objectives
2. Operations and Supply Chain Strategy
Current
Year
Objective:
5 Years
in the Future
| 2514
Current:
World-Class
Competitor
Cost
Manufacturing cost as a percentage of sales
Inventory turnover
55%
4.1
52%
eZ
50%
5.0
85%
99%
95%
3%
1%
1%
0.5%
1%
1%
90%
3 wk
95%
1 wk
95%
3 wk
10 mo
3 mo
6 mo
3 mo
8 mo
3 mo
400
200
400
30
iS
15
Quality
Customer satisfaction (percentage satisfied
with products)
Percentage of scrap and rework
Warranty cost as a percentage of sales
Delivery
Percentage of orders filled from stock
Lead time to fill stock
Flexibility
Number of months to introduce new products
Number of months to change capacity by +20%
Sustainability
Reduce carbon emissions (ppm)
Reduce workplace accidents per thousand
workers
measured for compliance. McDonald’s tracks the performance of each restaurant and compares the results with those of competitors.
Strategic
Decisions
TABLE 2.2
Examples of
Important Strategic
Decisions in
Operations
Strategic decisions constitute the third element of operations strategy. These decisions
determine how the operations objectives will be achieved. A consistent pattern of strategic
decisions should be made for each of the major operations decision categories (process,
quality, capacity, inventory, and supply chain). These decisions must be well integrated
with other functional decisions. This coordination and consistency is one of the most difficult things to achieve in business.
Table 2.2 indicates some important strategic decisions for operations. Note that these
decisions may require trade-offs or choices. For example, in the capacity area there is a
Strategic Decision
Decision Type
Strategic Choice
Process
Span of process
Automation
Process flow
Job specialization
Make or buy
Handmade or machine-made
Project, batch, line, or continuous
High or low specialization
Quality
Approach
Prevention or inspection
Training
Suppliers
Technical or managerial training
Selected on quality or cost
Capacity
Facility size
Location
Investment
One large or several small facilities
Near markets, low cost, or foreign
Permanent or temporary
Inventory
Amount
Distribution
Control systems
High or low levels of inventory
Centralized or decentralized warehouse
Control in greater or less detail
Supply Chain
Sourcing
Logistics
Insource or outsource products
National or global distribution
|
26 | Part One
Introduction
choice between one large facility and several smaller ones. While the large facility may
require less total investment due to economies of scale, the smaller facilities can be located
in their markets and provide better customer service. Thus, the strategic decision depends
on what objectives are being pursued in operations, the availability of capital, marketing
objectives, and so forth.
McDonald’s illustrates how a consistent pattern of strategic decisions is made in the
five operations decisions areas:
Process: Specialized equipment and work flows ensure meals are delivered to
customers quickly. For example, the special French fry scoop puts the right amount
of fries in each serving with little effort. Also, servers use information technology
to instantly communicate orders to food preparers.
Quality: More than 2000 quality, food safety, and inspections monitor food as it
moves from farms to suppliers to restaurants. McDonald’s requires that 72 safety
and quality protocols be conducted at each restaurant every day. Managers are
trained at “Hamburger U” in the McDonald’s system to ensure standards for
service, speed, food quality, cleanliness, and courtesy are met.
Capacity: Restaurant capacity is carefully designed to control customer waiting
times. Employees are scheduled to meet the fluctuating demand during the day.
Inventory: Just-in-time replenishment ensures food and packaging are available
when needed. Food and packaging are highly standardized across restaurants.
Supply Chain: Each restaurant is connected to its supply chain for fast replenishment. The supply chain is designed for frequent deliveries and to avoid stockouts.
Distinctive
Competence
All operations should have a distinctive competence (or operations capability) that differentiates it from the competitors. The distinctive competence is something that operations
does better than anyone else. It may be based on unique resources (human or capital) that
are difficult to imitate. Distinctive competence can also be based on an embedded organizational culture, proprietary or patented technology, or any innovation in operations that
cannot be copied easily.
The distinctive competence should match the mission of operations. For example, it is
a mismatch to have a distinctive competence of superior inventory management systems
when the operations mission is to excel at new product introduction. Likewise, the distinctive competence must be coordinated with marketing, finance, and the other functions so
that it is supported across the entire business as a basis for competitive advantage.
Distinctive competence may be used to define a particular business strategy in an ongoing business. The business strategy does not
always emanate from the market; it may be built instead on match-
Walmart has distinctive competencies to support
its low-cost strategy.
John Flournoy/McGraw-Hill Education
ing operations’ distinctive competence (current or projected) with a
current or potential new market. Both a viable market segment and
a unique capability to deliver the product or service offered must be
present for the firm to compete.
Walmart has a mission to be the low-cost retailer. To achieve this
mission it has developed a distinctive competence in cross-docking
aimed at lowering the costs of shipping. Using cross-docking, goods
from suppliers’ trucks are transferred across the loading dock to
waiting Walmart trucks and delivered to the stores without entering the warehouse. Walmart also has a sophisticated inventory control system and more purchasing power than its competitors and
Chapter
2. Operations and Supply Chain Strategy
| 27)
therefore can minimize inventories and related costs. These distinctive competencies help
Walmart compete on the basis of low cost.
McDonald’s early distinctive competence was its unique service and supply chain that it
designed. Since other firms have copied this system over time, the distinctive competence
has shifted to continuous improvement of the transformation system along with the brand.
McDonald’s system and its brand are now its distinctive competence.
2.2
COMPETING WITH OPERATIONS OBJECTIVES
LO2.3 Differentiate
the ways to compete
with operations
objectives.
We will now use the four operations objectives discussed above to describe different ways
to compete through operations. Most firms choose one or two objectives to focus on, so
that the strategic decisions made in operations can be aligned with and support these
focused objectives.
Suppose we start with the idea of competing through a quality objective. Delivering
quality means satisfying customer requirements. This assumes that marketing has identified a particular target market, and that operations understands these customers’ specific
requirements. Operations processes must be capable of meeting those requirements. If
competing through quality is the objective, strategic decisions related to product or service design, operations, and the supply chain must support customers’ expectations for
quality. For example, in a manufacturing company quality is improved by taking preventative measures in training workers, design of the process, and eliminating rework
and non-value-added activities. Customers who require quality expect a very low level
of defects.
Now, suppose we decide to pursue a low-cost objective. Perhaps the best way to achieve
low cost is to focus on conforming to customer requirements (quality) in product/service
design and operations processes by eliminating rework, scrap, and other non-value-added
activities. By preventing errors through continuous improvement, costs can be lowered
at the same time as quality improves. A low-cost objective may require more than just an
emphasis on conformance quality. Investment in automation and information systems may
also be needed to reduce costs.
If we select a delivery objective, strategic decisions should support fast or on-time
delivery, depending on the expectations of customers. Industrial (business) customers often
want on-time delivery because they schedule loading docks at warehouses or retail stores
and do not want several trucks delivering at the same time. Fast delivery may be desirable for consumers ordering products online (they will order from the company that can
deliver the product soonest). When quality improvement efforts reduce non-value-added
steps, the time to produce and deliver the product is indirectly reduced. Time can also be
directly reduced by improving process changeover times, simplifying complex operations,
and redesigning the product or service for fast production.
Finally, we could choose to emphasize a flexibility objective. If we reduce delivery
time, flexibility will automatically improve. For example, if it originally took 16 weeks to
make a product and we reduce production time to 2 weeks, then it is possible to change
the schedule within a 2-week time frame rather than 16 weeks, making operations more
flexible to changes in customer requirements. Other types of flexibility can be directly
improved by adding capacity, buying more flexible equipment, training workers to perform a wider variety of tasks, or redesigning the product or service for high variety.
We can see that operations objectives are connected. If we stress a quality objective, we
also naturally get some cost reduction, delivery improvement, and more flexibility. Quality is often a good place to start for improving operations. Then other objectives may be
tackled by directing strategic decisions to impact them.
While objectives are connected, an operation must still choose its number one or number
two strategic objectives to emphasize. For example, McDonald’s excels at low-cost and fast
(delivery) service. It does not emphasize flexibility, rather exactly the opposite. Zara, a giant
European fashion retailer, is able to achieve fast replenishment (delivery) of hot-selling
items within a few weeks by holding spare capacity and supporting rapid supply chain management practices. Zara emphasizes fast fashion, not low cost. Lexus is renowned for high
quality in its cars, which results from both its design and its manufacturing process. Chipotle
uses a highly flexible serving process that allows each customer to self-customize a meal.
2.3
LO2.4
CROSS-FUNCTIONAL STRATEGIC DECISIONS
Compare
product imitator and
innovator strategies.
Not only should objectives be linked to an operations mission, operations strategic decisions should be linked to business strategy and to marketing and financial strategies as
well. Table 2.3 illustrates this linkage by showing two diametrically opposite business
strategies that can be selected and the resulting functional strategies. The first one is the
product imitator (low-cost) business strategy, which is typical of a mature, price-sensitive
market with a standardized product (or service). In this case, the operations objective
should emphasize cost as the dominant objective, and operations should strive to reduce
costs through strategic decisions such as superior process technology, low personnel costs,
low inventory levels, a high degree of vertical integration, and quality improvement aimed
at saving cost. Marketing and finance also need to pursue and support the product imitator
business strategy, as shown in Table 2.3.
The second business strategy shown in the table is one of product innovator and
new product introduction (or product/service leadership). This strategy typically is used
in emerging and possibly growing markets where advantage can be gained by bringing to
TABLE 2.3
Strategy B
Strategic Alternatives
Business Strategy
Market Conditions
Product Imitator
Price-sensitive
Product Innovator
High volume
Standardization
Product-features-sensitive
Emerging market
Low volume
Customized products
Operations Mission
and Objectives
Emphasize low cost for mature
products
Emphasize flexibility to introduce
new products
Operations Strategic
Decisions
Superior processes
Superior products
Flexible automation
Fast reaction to changes
Mature market
Dedicated automation
Slow reaction to changes
Economies of scale
Workforce involvement
Economies of scope
Use of product development
teams
Distinctive Competence Operations
Low cost through superior process
technology and vertical
integration
Marketing Strategies
Mass distribution
Maximizing of sales opportunities
National sales force
Selective distribution
New-market development
Product design
Sales made through agents
Low risk
Low profit margins
Higher risks
Higher profit margins
Repeat sales
Finance Strategies
Fast and reliable new product
introduction through product
teams and flexible automation
Chapter 2.
Operations and Supply Chain Strategy
| 29) |)
market superior-quality products in a short amount of time. Price is not the dominant form
of competition, and higher prices are charged, thereby putting a lower emphasis on costs.
In this case, the operations and supply chain objective is flexibility to introduce superior
new products rapidly and effectively. Operations strategic decisions include the use of new
product introduction teams, flexible automation that is adapted to new products, a workforce
with flexible skills, and rapidly responding to marketplace changes. Once again, finance
and marketing also need to support the business strategy to achieve an integrated whole.
What Table 2.3 indicates is that drastically different types of operations are needed to
support different business strategies. Flexibility and superior-quality products may cost
more for the product innovator strategy. There is no such thing as an all-purpose operation
that is best for all circumstances. Thus, when asked to evaluate operations, one must imme-
3M corporate strategy
diately ask, what is the business strategy, mission, and objective of operations? Evaluating
operations is contingent on strategy. In other words, operations with different strategies
should adopt different best practices and decisions.
Table 2.3 also suggests that all functions must support the business strategy for it to be
effective. For example, in the product imitator strategy, marketing should focus on mass
distribution, repeat sales, a national sales force, and maximization of sales opportunities.
In contrast, in the product innovator strategy, marketing should focus on selective distribution, new-market development, product design, and perhaps sales through agents. It is not
enough for just operations to be integrated with the business strategy; all functions must
support the business strategy and one another.
Integrating marketing and operations and clearly laying out a particular mission and
objective for operations is essential to meet customer preferences for order winners and
order qualifiers. An order winner is an objective that will cause customers in a particular
segment that marketing has selected as the target market to choose a particular product or
service. In the product imitator strategy, the order winner for the customer is price; this
implies the need for low cost in operations, marketing, and finance. Other objectives in
this case (flexibility, quality, and delivery) are order qualifiers in that the company must
have acceptable levels of these three objectives to qualify to get the customer order. Insufficient levels of performance on order qualifiers can cause the business to lose the order,
but higher performance on order qualifiers cannot by themselves win the order. Only low
price/cost will win the order in this case.
In the product innovator strategy, the
order winner is flexibility to introduce
PA se? alee cine
superior products rapidly and effectively;
AI NEAIOR
the order qualifiers are cost, delivery, and
quality. Note how the order winner depends
on the particular strategy selected and that
zone
all functions must pursue superior levels
relative to the competition on the order
winner while achieving levels acceptable to
the customer on order qualifiers.
What is the order winner at Walmart?
It's low cost, and strategic decisions in
operations are made to keep costs down.
The same cannot be said about Nordstrom,
which competes on upscale merchandise
and superior customer service. Since the
order winners in these stores are different,
so are the operations strategies.
|
30.
2.4
LO2.5
Part One
Introduction
GLOBAL OPERATIONS AND SUPPLY CHAINS
Explain the
nature of global
operations and
supply chains.
Every day we hear that markets are becoming global in nature. Many products and services
are global in nature, including soft drinks, cell phones, TVs, fast food, banking, travel,
automobiles, motorcycles, farm equipment, machine tools, and a wide variety of other
products. To be sure, there are still market niches that are national or even local but the
trend is toward more global markets and products.
When operating in global markets, a company needs to be organized properly to produce
and market its products. As a result, the global corporation has emerged with the following characteristics. Facilities and plants are located on a worldwide basis, not country by
country. Products and services can be shifted between countries. Components, parts, and
services are sourced on a global basis. The best worldwide suppliers are sought, regardless of their national origin. The entire supply chain is global in nature. A few well-known
manufacturing companies that are global corporations are Ford, 3M, Nestlé, Philips,
Deere & Company, Coca-Cola, and Caterpillar.
The global corporation uses global product design and process technology. A basic
product or service is designed, whenever possible, to fit global tastes. When a local variation is needed, it is handled as an option rather than as a separate product or service.
Process technology is also standardized globally. For example, Black & Decker recently
designed worldwide
hand tools. Even fast food, clothing, and soft drinks have become
global products.
In the global corporation, demand for products or services is considered on a worldwide, not a local, basis. Therefore, the economies of scale are greatly magnified, and costs
can be lower. The iPhone came out as a worldwide product and was marketed globally. Its
demand and cost were scaled for a global market right from the start.
Logistics and inventory control systems are also global in nature. This makes it possible
to coordinate shipments of products and components on a worldwide basis.
Some services have taken on a global scope of operations. For example, consulting
firms, fast food, telecommunications, air travel, entertainment, financial services, and soft-
ware programming have global operations. Global consolidation has taken the place of
these formerly fragmented service industries. Certainly, not all service is global. There
remain services that are delivered on a local basis to serve local markets, but the trend
toward globalization is undeniable. A few of the service companies that are globally oriented are British Airways, Starbucks, Microsoft, and McDonald’s.
Some firms have adopted a hybrid approach with global economies of scale but a local
touch. In this case certain functions, for example, product design and manufacturing, are
handled on a local basis, while other functions, such as logistics and
inventory control systems, are standardized around the world. See
the Under Armour Operations Leader box for a strategy to compete
in global markets with local manufacturing.
The implications for operations strategy of this change toward
global business are profound. Operations and supply chains must
be conceived of as global in nature, including very broad searches
for both suppliers and customers. A global distinctive competence
should be developed for operations, along with a global mission,
objectives, and strategic decisions. Product design, process design,
Pfizer competes globally in 52 locations around
facility location, workforce policies, and virtually all decisions in
the world.
Ulrich Baumgarten/Getty Images
operations and the supply chain are affected.
Chapter 2
| 34)
Operations and Supply Chain Strategy
eae i
Se
SU eT
OPERATIONS LEADER
Under Armour Goes Local
of
Timing is everything in this
sports clothing and acces-
Under
Armour,
“fashion industry” for firms
like Under Armour.
Engineers and designers from around the world
will rotate and collaborate at headquarters in
Baltimore,
Maryland,
to
develop the new manufacturing processes.
Under
Armour hopes to improve
response times with their
sories,
has
maker
an
opera-
tions mission to develop
advanced
manufacturing
processes with the goal
of
on
producing
products
a small scale in local
markets where they are
sold. They plan to manufacture in the U.S. for U.S.
consumers,
in
Brazil
for
South American buyers, in
Europe for the European
market, and in China for
the Chinese market.
Under
Armour
currently
“local for local” model,
Justin Sullivan/Getty Images
manufactures
primarily
by
both producing and selling
within their global markets.
in
China, Jordan, and Vietnam. But shipping times to key
markets are typically weeks and sometimes longer
due to production delays or labor disputes at ports.
Source: L. Mirabella, “Under Armour Pushes to Localize
Manufacturing,” Baltimore Sun, reprinted in Star Tribune,
Oct. 26, 2015 and https://about.underarmour.com/investorrelations, 2020.
However, not all companies are following a global operations and supply chain strategy
with offshore manufacturing. Some companies that have moved some of their manufacturing Overseas are moving it back to the U.S., called reshoring. This is occurring as companies reconfigure their supply chains based on changing economic conditions, taxes, quality
considerations, and other factors that affect strategic operations and international supply
chain location decisions.
2.5
LO2.6
SUPPLY CHAIN STRATEGY
Analyze two
types of supply chain
strategies.
In the same way that operations are being expanded to a global context, operations strategy
can be expanded to supply chain strategy. Supply chain strategy includes consideration
of customers, suppliers, sourcing, and logistics in addition to operations. It requires a focus
on flows of inventory, materials, and information throughout the supply chain from the
suppliers to the ultimate customer.
Supply chain strategy takes into account not only the operations strategy of the firm
but also the strategies and capabilities of the suppliers and customers in the firm’s supply
chain. A supply chain strategy allows the supply chain to compete, not just the firm.
Supply chain strategy should be aimed at achieving a sustainable competitive advantage
for the entire supply chain. This advantage can be achieved by expanding the concepts
already covered in this chapter. For example, a supply chain should have a distinctive competence that is valuable and difficult to imitate or replace by competitors. This distinctive
competence should be based on what the firm does along with actions of its supply chain
partners. In a similar way, the supply chain partners and the firm should be working toward
|
32 | Part One
Jntroduction
the same mission and objectives in order to have a consistent supply chain strategy. Since
no single firm controls the entire supply chain, a coherent supply chain strategy can be difficult to achieve. Nevertheless, it is important to realize that supply chain partners that are
working toward differing goals will not be competitive with other supply chains that have
achieved a high degree of cooperation and consistency.
Just like we have already discussed for operations, there are two fundamental supply
chain strategies: imitation and innovation. Imitators have products or services similar to
those of their competitors and are oriented toward efficiency and low cost as a way of competing. In contrast, innovators differentiate their product or service as their way of competing and can charge higher prices. For example, Sport Obermeyer is in the fashion skiwear
business. Each year up to 95 percent of its ski, snowboard, and outdoor clothes are new
or redesigned. The company must plan production and forecast demand well in advance
of sales (more than a year), and as a result it often has stockouts or overstock situations
that result in markdowns at the end of the season. The problem facing Sport Obermeyer is
that its supply chain doesn’t match the nature of its product. The supply chain is oriented
toward efficiency and low cost, whereas Sport Obermeyer with its innovative products
needs a faster more flexible supply chain with better forecasting that responds to uncertain
demand. Sport Obermeyer appears to have the wrong supply chain.
To identify the proper supply chain, companies should first sort their products into
two categories: imitative and innovative. Imitative products are like commodities—they
have predictable demand and low profit margins. As a result, imitative products should
have a very efficient low-cost supply chain. Examples are toothpaste, oil, standard automobiles, and most food. In contrast, innovative products have unpredictable demand and
high profit margins. They need a flexible and fast supply chain to deal with uncertainty in
demand. Examples are fashion clothing, electric cars, new electronic products, and some
toys. The characteristics of these two types of products are shown at the top of Table 2.4.
Note that the product life cycle, contribution margin, average forecasting error, stockout
TABLE 2.4
Supply
Chain Strategies
Source: Adapted from
Fisher, M., What is the right
supply chain for your product?
Harvard Business Review,
Mar-Apr, 1997.
Product Differences
Imitative Products
Innovative Products
Product life cycle
Greater than 2 years
3 months to 1 year
Contribution margin
5% to 20%
20% to 60%
Average forecast error when
production is planned
10%
40% to 100%
Average stockout rate
1% to 2%
10% to 40%
Average forced end-of-theyear markdown
0%
10% to 25%
Supply Chain Strategy
Objective
Predictable supply at
low cost
Manufacturing
High utilization and
low-cost production
Respond quickly to
unpredictable demand.
Excess buffer capacity and
short throughput time.
May have low utilization
of capacity.
Inventory
High turnover
Significant buffers of parts or
finished goods. May have
low turnover.
Suppliers
Selected for cost and quality
Selected for speed, flexibility,
and quality.
Chapter
2. Operations and Supply Chain Strategy
| 33] [4
rate, and forced end-of-the-year markdowns are all dramatically
different for imitative and innovative products.
Firms often make the mistake of using one type of supply chain
strategy for both types of products. For example, General Mills, a
food company, might use an efficient supply chain with high inventory turnover and high utilization of its factories since most of its
products are imitative in nature. But General Mills needs a different supply chain that is highly flexible and responsive to meet the
uncertain demand for its new and innovative products. When firms
face this dilemma, they should not make the mistake of choosing
Sport Obermeyer needs an innovative supply
Shite
action sports/Shutterstock
2.6
only one supply chain strategy for all products.
The two types of supply chain strategies are summarized in the
bottom half of Table 2.4. The objectives of these two strategies are
different. While imitative supply chains should aim at a predictable supply at low cost, innovative supply chains should aim for quick response to unpredictable demand to minimize
stockouts, lost sales, and markdowns. In innovative supply chains the high margins can absorb
the higher costs of buffer capacity and buffer inventories needed to deal with uncertainty.
By using better forecasting, additional capacity, and drastic lead-time reductions for its
innovative products, Sport Obermeyer was able to cut the cost of both overproduction and
underproduction in half—enough to increase profits by 60 percent. Retailers were very
happy that product availability exceeded 99 percent, making Sport Obermeyer the best in
the industry for order fulfillment.
ENVIRONMENT AND SUSTAINABLE OPERATIONS
LO2.7 Illustrate
how operations
Sustainable operations has become an increasingly important part of operations and supply chain objectives and strategy. It refers to minimizing or eliminating the environmental
impact of operations along with social and financial viability of the firm for future gen-
and supply chain
erations. This is often referred to as the triple bottom line—environmental,
can become more
sustainable.
economic sustainability.
First, we describe the environmental part of the triple bottom line. In greening its supply chain a company should examine all opportunities, including product development,
sourcing, manufacturing, packaging, distribution, transportation, services, and end-of-life
management. This must be a cross-functional effort that involves all functions in the firm,
because it may require, for example, investments in new equipment or changes to product
packaging. Since reducing operations’ environmental impact is a strategic task, it must be
conceived as part of the business strategy.
Once the strategy has been set, it is usually best to start with a few focused initiatives
from the following list. Others can be attacked later.
social, and
. Curtail air, water, and landfill pollution.
. Reduce energy consumption.
. Minimize transportation and total carbon footprint.
. Work with suppliers to use recyclable and biodegradable packaging.
. Incorporate product reuse, end-of-life return, and recycling.
nA
=
WN
The first step that most companies take is to measure their environmental impact in
any or all of these five areas. Once this is known, cross-functional teams can be formed to
develop strategies and action plans to improve the measures. These teams can also include
supply chain partners to spread the effort up and down the supply chain. This should be
Es
Part One
Introduction
done in concert with the operations and supply chain strategy to ensure that environmental
impacts are minimized.
A Walmart effort in 1989 to use recyclable and biodegradable packaging ended in failure. Critics and suppliers believed the effort was intended to generate benefits for Walmart
at the expense of its suppliers. In 2005, their environmental effort was different. It was
launched by forming teams of representatives from suppliers, Walmart management, environmental groups, government, and academics. The teams set goals, developed measures
of environmental impact, and implemented programs to green the supply chain. Some
Walmart results from 2018 and goals for 2025 are as follows:
¢
¢
Increase renewable energy sources to 50 percent by 2025 from 28 percent in 2018.
Reduce emissions from Walmart’s supply chain by one gigaton by 2030.
*
Work toward zero landfill waste with a 78 percent reduction by 2018.
*
Various goals for employee diversity, engaging associates and communities, providing disaster relief, and relieving hunger. From 2015 to 2018 Walmart donated over
2.5 billion pounds of food to food pantries.
Sustainability also includes social responsibility by firms. This can take many forms
including hiring a diverse workforce, providing equal opportunities for all employees,
insuring safe and healthy working conditions, fair handling of employee grievances, ethical practices of every sort, and following all federal and state regulations. Accepting social
responsibility need not cost more, and actually can reduce costs and improve productivity.
This is achieved through lower employee turnover, lower training costs, and lower costs of
finding replacement employees. More highly engaged and satisfied employees can offer
better customer service and greater productivity.
The third part of the triple bottom line is economic sustainability.
It is achieved by a business and operations strategy that achieves
a sustainable competitive advantage. A firm without a competitive
advantage can easily lose market share and customers eroding its
top line along with increased costs and ultimately low profitability
that threaten firm survival.
Sustainable operations is an objective and strategy that firms and
operations can achieve with the help of their supply chain partners.
This objective need not cost more or reduce profits. In many cases,
but not all, through product redesign or process changes, costs can
be reduced and profits improved. British retailer Marks & Spencer
has met goals to send zero waste to landfills and is the first major
Marks & Spencer sets high sustainability goals.
retailer to be carbon neutral. Supply chain partners have been an
TEA/123RF
important source of support for this effort.
2.7
KEY POINTS AND TERMS
This chapter emphasizes achieving a competitive advantage through operations by developing an operations and supply chain strategy based on what the customers of the business
value. The key points are as follows:
*
Operations strategy consists of mission, objectives, strategic decisions, and distinctive
competence. These four elements must be tightly integrated with one another and with
other functions.
*
The operations mission should be aligned with the business strategy. Possible missions for
operations include low cost, fast new product introduction, fast delivery, or best quality.
Chapter 2.
Operations and Supply Chain Strategy
| 35] |
¢
The objectives of operations are cost, quality, delivery, and flexibility. One of the four
objectives should be selected as an order winner; the others are order qualifiers.
*
Operations strategic decisions indicate how operations objectives will be achieved.
Strategic decisions can be developed for each of the major decision areas (process, quality systems, capacity, inventory, and supply chain).
Key Terms
¢
The operations and supply chain strategy must be linked to the business strategy and
other functional strategies, leading to a consistent pattern of decisions, unique capability, and competitive advantage for the firm.
¢
The distinctive competence of operations should support the mission and differentiate
operations from its competitors. Possible distinctive competencies include proprietary
technology, embedded organization culture, and any innovation in operations that cannot be copied easily.
¢
The scope of operations and supply chain strategy has expanded to a global basis, particularly for businesses pursuing a global business strategy.
¢
In many situations the basis of competition is not the firm but the entire supply chain.
Supply chain strategy is an extension of operations strategy that considers not only the
firm but also the strategies and capabilities of its supply chain partners.
¢
Sustainable operations has become a critical objective and strategy. It should be
approached by forming cross-functional teams that include suppliers. Measuring and
reducing the environmental, social, and economic impact in all phases of design, operations, and distribution are the path of action.
¢
There is no one best strategy for all operations and their supply chains. The mission,
objectives, strategic decisions, and distinctive competence depend on whether a product
imitator, product innovator, or another strategy is being pursued by the business.
Operations strategy 21
Functional strategy 22
Corporate strategy 23
Business strategy 23
Mission 24
Operations objectives 24
Benchmarking 24
Strategic decisions 25
Distinctive
competence 26
Quality objective 27
Low-cost objective 27
Delivery objective 27
Flexibility objective 27
Product imitator 28
Product innovator 28
Order winners 29
Order qualifiers 29
Global corporation 30
Reshoring 31
Supply chain strategy 31
Sustainable operations 33
Triple bottom line 33
LEARNING
ENRICHMENT
What Is Operations Strategy?
https://youtu.be/j UKiuE4-aw0
Video
2:05
(for self-study or
Developing Supply Chain Strategy
Video
Introduction to Supply Chain Strategy
https://www.thebalancesmb.com/strategic-supply-chain-management2221231
Website
IKEA, a Leader in Sustainability-TED Talk
Video
instructor assignments)
https://youtu.be/cEyYBTEOAZ48
https://youtu.be/buH_vs7LFzw
5:06
(3218
ty
| 36 | Part One
.
Introduction
Discussion Questions
Find examples of operations and supply chain strategies. Write a few paragraphs describing the strategies
being pursued.
ie What are the reasons for formulating and implementing
an operations and supply chain strategy?
. Describe a possible operations mission that fits the following business situations:
Describe an operation where higher quality will cost
more. What is your definition of quality? Why does
higher quality cost more? If you use a different definition of quality, will higher quality cost less?
a. Ambulance service.
b. Production of hybrid automobile batteries.
c. Production of electronics products that have a short
product life cycle.
. An operations manager was heard complaining, “The
boss never listens to me—all the boss wants from me
is to avoid making waves. I rarely get any capital to
improve operations.”
10. What is the distinctive competence of the following
companies?
a. Starbucks Coffee Company
b. Hewlett Packard
c. Burger King
b. What should be done about the situation?
italy Explain how a distinctive competence in operations can
be the basis for competition in the company.
Define the following terms in your own words: opera-
Ey
a. Does the business have an operations strategy?
tions mission, order winner, order qualifier, and distinc-
tive competence.
How would you determine whether a company has an
operations and supply chain strategy? What specific
questions would you ask, and what information would
you gather?
Evaluate two local hospitals in terms of their emphasis on
the four operations objectives: cost, quality, delivery, and
flexibility. Are all departments focused on the same objectives? What are the order winners and the order qualifiers?
Define some of the strategic decisions that might be
required in a grocery store’s operation and its supply
chain depending on whether the strategy was emphasizing imitation or innovation.
Give two examples of a distinctive competence that can
be sustained and not easily duplicated. Explain why it is
hard to copy these distinctive competencies.
Give an example of a global business with which you
are familiar. How has globalization affected its operations and its supply chain?
14. What are the practical consequences of a lack of strategic linkage between the business and the operations
function?
IIS, Find a company that emphasizes sustainability. What
types of programs, measures, and objectives does this
company have?
16. Under what circumstances can a company pursuing a
sustainability strategy actually increase its profits and
under what circumstances might it not?
i
Explain the benefits of modular design.
4
New product development affects every part of a business, including operations and the
supply chain. New products provide growth opportunities and a competitive advantage for a
firm, so they are very important to the marketing and finance functions. Increasingly, there is
a challenge to introduce new products more quickly without sacrificing quality. For example,
the world’s automobile makers can now introduce a new model in two years, even with many
new technology features being designed and added, whereas it used to take four years or
longer. Personal computers have a very short product life cycle, sometimes less than a year.
New product design greatly affects operations by specifying the products that will be
made; it is a prerequisite for production to occur. At the same time, existing processes
and products can constrain the technology available for new products. Thus, new products
must be defined with not only the market in mind but also the production process that will
be used to make a product. New product design also affects the supply chain, as suppliers
will need to provide the various components that are needed, and sometimes be involved
in designing them.
°
°o
ead,
o
°
°
°
°
A
tain
°
°
2
° is
° haa
°°
6
°
ee
|
°
°
i
°
°
°
°
°
°
°
°
.
-
‘
Wi
°
°
37
TE
Part One
Jntroduction
The LEGO Group
The
product development
process
usually results in a
new product for the market, but the process is also useful for redesigning existing products. The LEGO Group
is investing over $100 million and hiring about 100 new
workers to redesign its current product. The goal is to
eliminate the use of petroleum-based plastics and make
the toys entirely from plant or recycled materials by
2030.
In this case, the product development process is
not designing a new toy. In fact, LEGO hopes customers will not notice any difference in their manufactured
“elements”—the small plastic pieces that kids (and adults)
around the world enjoy. Although a new raw material will
be used to manufacture the lego pieces, they should
have the same “click” together and easy separation as
the existing plastic pieces. The pieces also need to be
bright in color and stand up to unexpected trips through
the laundry!
The LEGO Group is based in Billund, Denmark, where
it produces
about
100
million
lego pieces
every
day.
Milosh Kojadinovich/12RF
They employ
more
than 250 product designers
in the
creative core of the company. They also coordinate with
educational institutions to carry out research on materials,
develop technology, and understand how children play.
While the cost and effort to develop a reliable new
raw
material
are significant,
LEGO
wants
“children,
the
builders of tomorrow, to inherit a healthy planet.”
Source: www.lego.com, 2020.
Product design follows from the development of a business strategy. The business
strategy includes a value proposition that defines the target market, the differentiation of
the product, and why the customer should buy it. This is the starting point for designing
a new product. Factors that will be considered are customer preferences, market size,
materials available for making the product, and issues concerning sustainability. See the
Operations Leader box on The LEGO Group for linkages among sustainability, strategy,
and new product development.
3.1
STRATEGIES FOR NEW PRODUCT
LO3.1 Compare
the three strategies
for new product
introduction.
INTRODUCTION
There are three fundamentally different ways to introduce new products. These approaches
are called market pull, technology push, and interfunctional view.
:
hee
:
f
eer
‘
Market pull. According to this view, the market is the primary
basis for determining
the products a firm should make, with little regard for existing technology. A firm
should make what it can sell. Customer needs are determined, and then the firm
organizes the technology, resources, and processes needed to design a product and
supply the customer. The market will “pull” through the products that are made.
Technology push. In this view, technology is the primary determinant of the
products the firm should make. The firm pursues a technology-based advantage by
developing superior technologies in their materials and components. The products
are pushed into the market, and marketing’s job is to create demand for these new
products. Since the products have superior technology, they will have a natural
advantage in the market.
Chapter 3
FIGURE 3.1
Product Design
| 39] |)
Lack of cooperation in designing a swing.
The Firm Designs a Swing
As used by the customer
What the customer wanted
Interfunctional view. This view combines some of the advantages of the first two
strategies, with products that fit the market needs (new or existing markets) and have
a technical advantage as well. To accomplish this, all functions (e.g., marketing, engineering, operations, and finance) should collaborate to design the new products needed
by the firm. Often, this is done by forming cross-functional teams that are responsible for the development of a new product. This is the most appealing of the three
views but also the most difficult to implement. Often cross-functional friction must
be overcome to achieve the degree of cooperation required for interfunctional product
development to succeed. If it can be implemented, the interfunctional approach usually
will produce the best results, and we emphasize it in the remainder of this chapter.
A humorous look at a lack of interfunctional cooperation is depicted in Figure 3.1.
3.2
NEW PRODUCT DEVELOPMENT
LO3.2 Describe
the three phases
of new product
development.
PROCESS
Most firms have an organized new product development (NPD) process that follows specific phases or prescribed steps. These phases may be formally defined in company documents and require sign-offs by senior management between phases. The purpose of this
process is to gain control of product development and ensure that all important issues are
addressed by the NPD team. ISO 9000 certification requires that a prescribed NPD process
be defined and followed by the company in the development of its products.!
'|SO stands for the International Organization for Standards. ISO 9000 is a standard that applies to new
product development and to production to ensure that quality products are designed and manufactured.
a | 40
Part One
Introduction
The typical phases followed by firms in developing new products are concept
development, product design, and pilot production/testing. The names of these phases
and the number of phases may vary by company. But there is a great deal of similarity
among the various approaches used, even across a wide variety of industries.
Concept
Development
This phase is concerned with idea generation and the evaluation of alternative ideas for the
new product and is sometimes referred to as the fuzzy front end. During this phase, several
product concepts usually are generated and evaluated. The physical product is not designed
during concept development; rather, different approaches to defining and meeting market
needs are considered and the best approach is selected by the company. For example, when
a new smartphone is developed, it must start with concept development. What features
should the new phone have, including the size of the screen, the amount of memory, the
type of cameras, and the features of the software to name only a few? What will appeal to
the customers and the market?
Among the several conceptual designs considered and evaluated, one will be selected
for the next phase of product development. The decision to proceed to the product design
phase ordinarily requires top management approval. At the time of approval, a crossfunctional team is established, if one does not exist already, to design the new product.
Product
Design
This phase is concerned with designing the physical new product. At the beginning, the
firm has a general idea of what the new product will be but not too many specifics. At the
end of the product design phase, the firm has a set of product specifications and digital
images (or engineering drawings) specified in sufficient detail that production prototypes
can be built and tested.
Product design requires consideration of many different trade-offs between product
cost, quality (features), and the schedule for bringing the new product to market. Engineers
are assigned to work on the various parts of the project. As they work, they make decisions
that ultimately will affect the product’s cost, its quality (features), and the schedule for
product introduction. It is easy to see why marketing, operations, and finance/accounting
must also be involved with engineering during this phase so that appropriate trade-offs can
be made for the greatest benefit of the entire business.
Engineering usually uses software to design the product and simulate its operation
before it is made. This will help ensure that the product works when it is produced. Virtual
prototypes, designed and tested using specialized software, are frequently used to speed
up and simplify the engineering design tasks. Computer-aided design (CAD) systems are
also used to design and view the product digitally and, in most cases, eliminate the need for
paper blueprints or drawings. At the end of this phase,
the designs are transmitted to production as a basis for
pilot production. In the smartphone example above, all
of the physical components and the software will have
detailed specifications for manufacturing.
Process design consists of planning the resources,
tasks, and training needed for manufacturing the
new product. It should be taking place simultaneously with product design. Manufacturing should
not wait for the final design to be completed before
process design begins. As a matter of fact, it is better if process design is done in parallel with product
This is a CAD system used for product design.
design so that changes can be made in the product to
Shutterstock/Gorodenkoff
facilitate the production process before finalizing the
FIGURE
3.2
New product design
process.
Chapter 3 Product Design
| 44)
Product design
Preliminary process design
|
Pilot production/testing |
Final process design
a Concept development
Se
*
:
product design. It is also a good idea for the product designers to have some manufacturing experience so that they are aware of the process options available and the pitfalls
of designs that can lead to poor production processes. Figure 3.2 shows how process
design should proceed in parallel with product design.
Pilot
Production/
Testing
In this third phase, products require testing of production prototypes before they are put
into production. For example, in the design of a new laptop computer, several laptops
would be built as prototypes and tested for their ability to meet the product specifications.
This may include performance tests of hardware and software and lifetime tests of reliability of the laptop. Similar pilot production and testing is done for aircraft, automobiles, new
cereals, and many other new products.
During this phase, the process for production is finalized. Since the product design is
nearing completion, the process can be designed in great detail and tested for its capability to
make the product that has been designed. Process and product modifications should be considered so that the process is optimized before full-scale production and market introduction
begin. To facilitate full-scale production, a set of documents should be finalized that contains
not only product specifications but also process design specifications, training procedures for
operators, and test results. This will facilitate the transition from design to production.
3D printing or additive manufacturing has made physical prototype development
faster and easier. Almost any three-dimensional shape can be made by laying down successive layers of metal, plastic,
or ceramics. An inkjet type of
printer is used to make successive and additive layers of
the three-dimensional _prototype. Prototypes for testing
can be produced in a few hours
instead of days or weeks with
older technologies.
As 3D printing technology
and the related software continue to improve, more complex designs become possible.
Airbus, for example, is experimenting with 3D printing metal
Scientists use 3D printing technology to “print” human cells
of spinal cord tissue. While these prototypes are still being
components for its jets, allowtested, they will soon be ready for implanting into patients.
ing their engineers a variety Of — Robert Clark/Getty Images
|
42
Part One
Introduction
new options in the design process. One such prototype is 30 percent lighter than the manufactured component it would replace, a desirable characteristic in their industry. In the medical industry, scientists are experimenting with printing prototypes built of human cells, as
well as a variety of other new materials.
The rapid prototyping enabled by 3D printing is revolutionizing new product development processes in many industries. Helping firms speed up this portion of the design process will enable them to create more and better product designs.
3.3
CROSS-FUNCTIONAL
PRODUCT
DESIGN
LO3.3 Evaluate
how concurrent
The new product development process is one of frequent misalignment. No matter how
excellent the advanced planning or the technology is, misalignment between marketing,
engineering deals
product design, and operations is a common
with misalignment.
variety of reasons.
Misalignment in marketing occurs when the product does not meet market needs. This
can occur because of poor market intelligence or a misinterpretation of market needs. As a
result, sales may be low and the product may need to be redesigned or eliminated.
Technology misalignment occurs when the product designed by engineering cannot
be made by operations. This happens when technologies are new or unproven or are not
well understood. Operations can be misaligned with the new product in terms of materials
needed, labor skills, quality assurance, and scheduling. Finally, reward systems throughout the organization may reinforce the use of current technology and production practices
rather than the new processes needed.
To overcome these problems in NPD, a concurrent marketing, engineering, and production approach can be used. The traditional approach proceeds in stages or steps, as shown
in part (a) of Figure 3.3. In a sequential process, each function completes its work before
the next one starts.
Figure 3.3 (b) illustrates a simultaneous development process, also called concurrent
engineering (or simultaneous engineering). All functions are involved from the beginning,
frequently by forming an NPD team, as soon as concept development is started. In the first
stage, marketing has the major effort, but other functions also have a role. During the product design phase, marketing reduces its effort, but not to zero, while engineering has the
major role. Finally, operations picks up the lead as the new product is tested and launched
into the market.
FIGURE 3.3
occurrence. Misalignments can occur for a
Sequential and concurrent approaches.
Marketing
Effort
Engineering
Operations
Effort
e
Time
Time
Sequential approach
Concurrent approach
(a)
(b)
Chapter 3.
Product Design
| 43)
The traditional approach is more like a relay race, while the concurrent approach is like
rugby. In a relay race, each runner picks up the baton for one portion of the race. In rugby,
the entire team runs down the field together, pushing and shoving in a group, to advance
the ball toward the goal.
Concurrent engineering is used extensively by NASA (National Aeronautics and
Space Administration) to coordinate both internally and externally with customers. At
their Integrated Design Center, design teams develop new tools and technologies for a
variety of purposes. The cross-functional teams develop tools such as telescopes and
hyper-sensitive imagers, along with space flight designs and space architecture. NASA
scientists and engineers work side-by-side with other functions as well as customers who
may later manufacture the designed products. Their concurrent approach allows them to
improve their product designs and to shorten development schedules.
Concurrent engineering is not always effective. It is less likely to be successful in projects with high uncertainty (e.g., unfamiliar product, market, or technology). Concurrent
engineering is more likely to improve performance for product extensions or products that
serve familiar markets with current technology. Its main advantages, of helping to reduce
misalignment and shorten project completion times, make this method useful in a variety
of industries.
3.4
SUPPLY CHAIN COLLABORATION
LO3.4 Describe the
criteria for selecting
suppliers for
collaboration.
— Just as internal collaboration is important, so is external collaboration with supply chain
customers and suppliers. While relationships with customers and suppliers often are established in new product development, collaboration is something different, requiring actual
participation in the design process.
Collaboration with customers (either consumers or business customers) means tapping
into their knowledge and expertise to design products they are willing to buy. The collaboration can take many different forms, including the following:
¢
Asking customers the right questions. What can we do to help you make your lives
easier or more productive?
e
Aligning incentives for customers to share their knowledge with the design team. Incentives could include merchandise, monetary rewards, and first access to new designs.
¢
Creating a collaborative technology platform to share information. This can take many
forms, including computer networks or software to enable collaboration. For example,
National Semiconductor created software that allows customers to design circuits by
using National Semiconductor’s products.
¢
Including customers as advisors to the design team.
While collaborating with customers provides benefits, it also requires a change in
attitude from “controlling the design” to working in partnership with customers. This
requires a different mindset. For example, the designers cannot simply ask the customers what they want or need, since customers don’t always know. A more sophisticated
approach is to observe customers using current products to find limitations. The members
of the design team should also ask themselves, What do customers do that we can do better? When they work with customers collaboratively in this way, new product designs are
improved.
The second aspect of collaboration in the supply chain is working with suppliers. Since
purchased materials and components often account for more than 5O percent of the cost
of goods sold, suppliers should collaborate to design the product. This is particularly
44 | Part One
Introduction
important when the product involves new technologies in which the company does not
have expertise.
Suppliers can be asked to join the NPD team or to provide input at critical points in the
design process. Their role is to offer improvements in the design or alternative approaches
that leverage their own expertise. When a supplier is considered a potential collaborator,
the following criteria should be considered:
*
*
*
*
Technical expertise: Does the supplier have technical expertise that the company does
not have?
Capability: Can the supplier meet targets for cost, quality, and product performance?
Capacity: Can the supplier meet the product development schedule and the ramp-up to
production?
Low risk: What is the risk that the supplier will not perform as expected?
Supplier collaboration can provide improvements of 10 to 20 percent in cost, time,
quality, and product performance. But should all suppliers be considered for collaboration? No, it is best to include only those who are critical to the design and have some-
thing useful to offer the process. One example of successful supplier collaboration is TPI
Composites wind turbine blades, as shown in the Operations Leader box. When there is
collaboration with critical suppliers and customers in the supply chain, the design process
is greatly improved.
OAR ROARS a AT
ASAE
ag]
eI
i “ et
cna
wed
-
1 ao
i
TPI Composites
TPI Composites, the largest U.S. independent manufacturer of composite wind blades, is a major supplier in the
wind energy industry. Their expertise in producing high
quality blades makes them an important supplier to other
manufacturers of wind turbine systems. Collaboration
within their supply chain is extensive.
The U.S. market
has grown
substantially, with over
60,000 turbines collecting wind energy for conversion
to electricity, and then sharing on the power grid (an
additional 60,000 turbines are used for local energy
consumption). Modern wind turbines are cost effective
and reliable, but also complex, with roughly 8000 parts,
including blades up to 75 meters (250 feet) in length and
towers over 80 meters (262 feet) high, roughly the height
of the Statue of Liberty.
Due to the size and complexity of turbine blades, TPIl’s
manufacturing process is costly and labor intensive. A
GYRO PHOTOGRAPHY/amanaimagesRF/
Getty Images
collaborative partnership with the Department of Energy,
Sandia
National
Laboratories,
lowa State University, and
their blade customers helped TPI reduce production times
for a single blade by 37 percent (from 38 to 24 hours).
As the demand
for renewable
energy
persists and
wind turbines are designed in ever larger sizes, TPI must
find ways to overcome
infrastructure constraints. These
constraints include highway underpass heights limiting
the size of wind towers and difficulty transporting longer
wind blades. Addressing these challenges through future
collaboration will be needed to advance the industry.
Source: www.energy.gov, 2020 and www.tpicomposites
.com, 2020.
Chapter 3. Product Design
| 451 |
3.5
QUALITY FUNCTION
DEPLOYMENT
LO3.5
Evaluate
The new product development process is aided by many different tools and techniques,
an example of
Quality Function
Deployment.
some of which are covered in the remainder of this chapter. Quality function deployment
(QFD) is a tool for linking customer requirements to technical product specifications. QFD
is very useful in translating the ordinary language obtained from the customers to technical
requirements understood by engineers. It also facilitates cooperation and communication
between marketing, engineering, and manufacturing.
QFD was first used in 1972 at the Mitsubishi shipyard in Japan. It spread from there to
Toyota and to American companies. Now many companies throughout the world are using
QFD in industries such as automobiles, electronics, home appliances, and services. QFD is
very useful as a communication tool.
When using QED, the firm begins by identifying various customer attributes. Each of
these attributes can be met by one or more engineering characteristics of the product. By
using the matrix shown in Figure 3.4, the customer attributes on the left side of the matrix
can be related to the engineering characteristics on the top of the matrix. When the matrix
is completed, it is called the house of quality.
The house of quality illustrated in Figure 3.4 will be explained in some detail by using
a bicycle example. We will work through this example one step at a time, beginning with
the customer attributes.
FIGURE 3.4
Relationship matrix.
Engineering characteristics
Relationships
Ff
@) Strong positive
\
Positive
X
Negative
«&) Strong negative
Peer fo
ome
Ce
a ee
P=
|jel-le
Helle
EZTEEELE
os
Customer
attributes
of
Number
(#)
gears
Cruising
(mph)
speed
(#)
paint
of
Coats
(ft/Ibs)
frame
of
Strength
B
a
1
Customer perceptions
2
3
A Our bike
O Competitor A
C] Competitor B
4
5
ea
|
46 | Part One
5
J/ntroduction
Customer
Attributes
The customer attributes (CAs) shown on the left side of the matrix in Figure 3.4 represent the voice of the customer. These attributes are determined through market research
involving potential customers of the bicycle to define the important attributes of the prod-
uct. Therefore, a target market must be defined so that the appropriate types of customers
can be contacted. Suppose, in this case, the bicycle is being designed for a very specific
market: use by college students on campus. College students would be interviewed to
determine what they consider important features or attributes of a bicycle. Suppose that the
students would like a bicycle that is easy to pedal, is strong and durable, has fast acceleration, has a low cost, and looks nice. Note that these CAs are not very specific at this point
and need further definition by means of the QFD process.
A few more things are now added to the house of quality. After the CAs have been listed
on the left side of the matrix, they are rated on their relative importance by customers to
sum to a total of 100 points. This is shown on the “chimney” column of the house of quality in Figure 3.4. On the right side of the matrix is a comparison of how the company’s
current bicycle compares to competitors’ offerings on each of the CAs.
Engineering
Characteristics
The next step in QFD is to translate the customer attributes into engineering characteristics
(ECs). This is done by determining how each of the customer attributes can be met by the
new bicycle design. Engineering characteristics must be measurable and specific and are
closely related to the final design specifications for the product.
For the bicycle design, some of the ECs might be number of gears, weight of the bicycle
in pounds, strength of the frame, cruising speed, and number of coats of paint on the
frame. These characteristics are listed across the top of the matrix in Figure 3.4 and then
related to each of the customer attributes. For example, the CA “easy to pedal” is strongly
related to the number of gears on the bicycle. Generally, the more gears, the easier it is to
pedal the bicycle in different conditions and situations. Also, “‘easy
to pedal” is inversely related to the weight of the bike. Various symbols are placed in the matrix (see Figure 3.4 for the key) to indicate
the nature of the relationship between each particular CA and the
ECs. This can be done by conducting engineering tests or by using
generally understood relationships.
Next, we switch to Figure 3.5, which adds a roof to the house of
quality. The roof shows how each EC is related to the other ECs.
This makes it possible to study any of the trade-offs that may be
required between
one EC and another. For example,
we see that
the weight of the bicycle will negatively affect its cruising speed.
Also, the number of coats of paint will have a mild positive effect
on bicycle weight.
Finally, on the bottom of the matrix in Figure 3.5, we have indi-
cated the value of each EC achieved by the competitors’ bicycles.
We have also shown a target value that we have set for each EC in
our new bicycle design. The target value is determined by design
decisions, based on the importance of various customer attributes,
the linkages to ECs, and the desired performance of the new bicycle
relative to those of the competitors. The ultimate result of the house
of quality is a translation of the CAs into target values for ECs on
the bottom of the matrix.
The house of quality is very useful in increasing cross-functional
Her bicycle can be designed by use of QFD.
communications because it neatly connects the market requirements
Paul Bradbury/age fotostock
with the design characteristics that engineers must consider. Thus, a
Chapter 3.
FIGURE 3.5
im
Relationships
12)
aoe
Q) Strong positive
s
S
—
—
5
S
Ov
me
\
Positive
X
Negative
E ge y
vo
attributes
| 47) |4
House of quality.
g
Customer
Product Design
=
op
=
Peer
4si5
Sa).
8
Pe
Za
Ss
aS
€ a
Ss
<
vo
- op
=)
=a)
eg
(x) Strong negative
;
Customer perceptions
5
Competitive
A Our bike
evaluation
© Competitor A
| | Competitor B
Targets
design can be developed that will meet the needs of the market while still considering all
the design trade-offs required. The house of quality is a visual tool that captures a variety
of information needed by various business functions during the new product development
process.
QFD also can be applied to service industries in much the same way it is applied to
manufacturing. To illustrate, suppose a pizza shop is considering adding a take-out delivery service for its pizza.
The CAs for this new service have been determined from customers to be fast, courte-
ous, and reliable service. Also, the delivery agent should have a clean-cut appearance and
the order should be delivered complete (no missing items) with hot pizza. The CAs are
listed on the left side of the QFD matrix in Figure 3.6.
For services it can be difficult to identify the ECs, which are sometimes hard to define
and measure.
In this case, the ECs are delivery time (minutes),
customer
satisfaction
(from a periodic survey of customers), proportion of orders delivered on time, and the
temperature of the pizza when it is delivered. Note that the customer survey will measure
intangible CAs such as a clean-cut appearance, courtesy, order completeness, and general
satisfaction with the service.
The CAs are now related to each of the ECs in the same way as in the bicycle example.
Also, the roof of the house of quality, customer perceptions, competitive evaluations, and
targets are added to complete the analysis, as shown in Figure 3.6. While QFD for service
may be measured somewhat differently than it is for manufacturing, the same general principles apply.
48
| Part One
FIGURE 3.6
Introduction
QED for Pizza U.S.A. delivery.
Relationships
v
E
Ey
Q) Strong positive
eae
\ Positive
hice
E
3.
=
2
X
&
o
Negative
(x) Strong negative
fa
minutes
Customer perceptions
5
Customer
Traeseie
|
Fwnawnaree
Pen rene |
[cansaanise
[0
ns
he
attributes
Complete order
Competitive
evaluation
A
B
A Our position
© Competitor A
(ej Competitor B
3.6
MODULAR
LO3.6 Explain the
benefits of modular
design.
DESIGN
Usually products are designed one at a time without much regard for commonality of parts
or modular properties that can aid production and still meet customer needs. Modular
design makes it possible to have relatively high product variety and low component variety
at the same time. The core idea is to design a series of basic product components, or modules, that can be assembled into a large number of different products. To the customer, it
appears there are a great number of different products. To operations and the supply chain,
there are only a limited number of basic components and processes.
Controlling the number of different components that go into products is of great
importance to operations, since this makes it possible to produce more efficiently
by standardization of processes and equipment. In the supply chain, few components
means fewer suppliers to manage and fewer parts to be purchased. A large number of
product components will greatly increase the complexity and cost of operations and
supply chain.
Modular design offers a fundamental way to change thinking about product design.
Instead of designing each product separately, the company designs products around standard component modules and standard processes. When this is done, the product line is
carefully analyzed and divided into basic modules. Common modules should be developed
that can serve more than one product line, and unnecessary product frills should be eliminated. This approach will still allow for a great deal of product variety, but the number of
unnecessary product variations will be reduced.
Chapter 3 Product Design
| 49) 1
The modular design approach
can best be illustrated by an example. A large manufacturer of beds
used modular design ideas, with four
basic sizes: twin, double, queen,
and king. The inside construction of
the mattresses was limited to only a
few different spring arrangements
and foam padding thicknesses. A
moderate variety of mattress covers
were used to meet consumer preferences for color and type of design.
Robert Wilson/123RF
This approach greatly reduced the
number of mattress components while providing substantial variety for the customer. For
example, with four bed sizes, three types of spring construction, three types of foam, and
eight different covers, a total of 288 different mattress designs were possible.
4x3x3
x 8 = 288 combinations
Not all combinations were produced, since some might be unacceptable to the customer
(e.g., the expensive springs with the thin foam pad). Although there are still many product
combinations in this example, the number of components has been limited.
Most automobile manufacturers use modular design for their vehicles. Even MINI
Cooper’s most basic model, for example, has many optional choices: four body colors,
three wheel designs, eight interior finish options, two engines, and eleven style options.
The theoretical number of different cars that can be produced is as follows:
ASR BSE 16
S< IU SS I?
Modular design provides an opportunity to streamline production at MINI while offering
abundant consumer choices. The company further benefits by using some of these modules in its other car models, saving significant time and cost in the module design process.
3.7
KEY POINTS AND TERMS
New product design has a great impact on operations and supply chain, and other functions
as well, since it determines the specifications for future products. Likewise, operations can
constrain a firm’s ability to develop new products and make them more costly to produce.
As a result, operations should be deeply involved in new product development.
¢
There are three ways to develop new products: market pull, technology push, and interfunctional. The interfunctional approach is usually the best since it includes both market
and technological considerations in the new product design.
¢
The NPD process is often specified in companies as having three phases: concept development, product design, and pilot production/testing.
¢
Products should be designed from the start for manufacturability. This is done by
considering design of the production process as part of product design and utilizing a
concurrent engineering approach.
*
Concurrent engineering uses overlapping phases for product design rather than a
sequential approach. Typically, an NPD team is formed with representation from all
major functions (marketing, engineering, operations, and finance/accounting) to ensure
cross-functional integration.
|
50 | Part One
Introduction
*
¢
*
*
Key Terms
3D printing, or additive manufacturing is useful for creating product prototypes. The
speed and flexibility of this new technology enhances product design opportunities.
Supply chain collaboration in NPD is essential. This should be accomplished by collaborating with both customers and suppliers in the NPD process.
QED is used to connect customer attributes to engineering characteristics. This typically is done through a technique called the house of quality that can be used for both
manufacturing and services.
Modular design is used to minimize the number of different parts needed to make a
product line of related products. This can be done by designing standard modules and
considering only the combinations of options that have significant market demand.
Market pull 38
Technology push 38
Interfunctional view 39
Concept development 40
Product design 40
Pilot production/testing 40
Process design 40
Production prototypes 41
3D printing 41
Additive manufacturing 41
Quality function deployment 45
House of quality 45
Customer attributes 46
Misalignment 42
Sequential process 42
Concurrent engineering 42
Collaboration 43
Engineering characteristics 46
Trade-offs 46
Target value 46
Modular design 48
LEARNING
What Is the New Product Development Process?
Video
ENRICHMENT
https://youtu.be/vQZjNIRpuFg
2:49
(for self-study i
Tools for Accelerating a Cross-functional Design Process
Web Link
instructor assignments) —hetps://www.mckinsey.com/business-functions/operations/our-insights/
accelerating-product-development-the-tools-you-need-now
Prototyping
https://youtu.be/SSWt-TS YD08
Video
220
3D Printing Prototype Example
https://youtu.be/RpFTRT8FkPO
Video
3:02
Sustainability in New Product Development
https://youtu.be/-HS-sIU-XTc
Video
3:56
Discussion Questions
1. Why is cross-functional cooperation important for new
product design? What are the symptoms of a possible
lack of cooperation?
. In what circumstances might a market-pull approach or
a technology-push approach to new product design be
the best approach?
. Describe the steps that might be required in writing and
producing a play. Compare these steps to the three steps
for NPD described in Section 3.2. How are they similar?
4. Why has there been an increase in product variety in
global markets?
5. How can modular design help to control production
variety and at the same time allow product variety?
. What is the proper role of the operations function in
product design?
. What form does the product specification take for the
following firms: a travel agency, a beer company, and a
consulting firm?
Chapter 3. Product Design BEXYY”
8. Find examples of modular design of products in everyday life.
y Work with one of your classmates as your customer;
you are the supplier. Have your customer select a product and specify the customer attributes (CAs) that are
desirable. Then you specify the engineering characteristics (ECs) required to meet the customer’s needs.
Complete the house of quality matrix by specifying the
relationships in the matrix. Ask your customer if the
resulting ECs will meet his or her needs.
. What are the essential benefits of using a QFD
approach to product design? Also, identify any negative
effects that might apply to the use of QFD.
Atle A student would like to design a backpack for student
books and supplies. The CAs are a (1) comfortable
backpack that is (2) durable with (3) enough room
and (4) not too heavy to carry. Think of some ECs
that can be used to measure these customer attributes.
Then construct a QFD matrix showing the positive
and negative relationships that you expect to see in
this case.
WP, An entrepreneur is designing a sub sandwich shop that
would be located on campus. Define the CAs that you
would like to see for the service (not the product) delivered at this location. Then specify some ECs that can be
used for measurement of the service.
13: Suppose a car you want to buy has five choices for interior colors, three types of stereos, three engine choices,
two battery types (regular and heavy duty), 10 exterior
colors, two transmission choices, and four types of
wheel covers. How many possible combinations of the
car are possible for the manufacturer? What can be done
to limit the number of combinations without limiting
customer choice?
oat)
sSsudoaves
yoouvrus
gestsPitot
Oh nes ai Gepaed
ilizog a} Quiwode xetinin cies i Bit
shan att
Mi S82 Of 5502S tiny lanl anldanoglex :
svi
~T Dily
ert is,
i
Va
¢ asc
‘ie his ev ioe
mre
lag joie Holyeboe doe @ secebeant oa Se
it a“)
oft aly
SS rh —
uy fet 2A oft snieG Suqdtiien bile satire
-vilol (gubow, od) 100) salvaSt wil Geaeet Ahhh bidew
U9,
oe
ae
GF og Sail
eres
itiie
yeea han
e renee
ati
nF Elva:
220s Snitt> sul) .osns tee ht) Merl alte atl
:
wonaies
Of yinb vrxsd bre paleo efi
: Srlenan yn ras
2
9
-=
P="
Tes ltwio-rairoa bass.
Sago
7
S
:
:
—
oH
90 6 1 sdt erie
inobate eit Aoapdond & egies ae94iF, bi w Wee
<
urcertil Maldive Janitaltaiowietedoui: di duit o1 Ss
~
aye
attitT Oi Vess
remind ond
ta 2oqy} mes bas., ayieuis aiiseineney iu TOKE
ste enoitenidinor. Hieng ynem wo ctvratxule
Mey Perity siats Td aasbiytoi mie Sdiveor so
ps
pe
ime ines cee
20 E> tad) 6S Sine Wisse Mila pao wanes bets.
IPSs ae MeJeane, tT bay a
“Sind tot asuinr
ls cast madd vir PIU
et
x~— rae i
RIS
TT PY na
Aaa
fee
Nibet Hgnone (€) aie>I
neyneranioh tySg +
:
7
a
—
LEARNING
mar
36
=
-
=
ior
nein
=
5
1 Press
~
.
249
=
Web Cink=
Miss ite
%
iL-
7
(Pwr
trees Chee
7
”
™ Vide
i
a
> ~-4
ee
-
vyleo
3:02
~
4
Vicker
3:56~
~
~
Se
———
:
-
vi
—-
i
’
a
The)ri ries
omy |fete bryreba oi
ne
Cay
ontee
aah
ee
PART
Process Design
nw
mm
ew
BY
Process Selection
Service Process Design
Process-Flow Analysis
Coe
a Lean Thinking and Lean Systems
Among the most important decisions made by operations managers are those
involving the design and improvement of the process for producing goods and
services. These decisions include choice of process and technology, analysis
of flows through operations, and the associated value added in operations. Two
themes underlie and unify Part Two: first, the idea of designing and improving a
process to enhance the flows of materials, customers, and information; second,
the idea of eliminating waste in processes. These principles can be used to
design and manage a process that not only is efficient but provides value for
the customer.
(Stab
7 te ae Je. 1m
Process Selection
BRBeEe
LEARNING
OBJECTIVES
eee
ee
After reading this chapter, you should be able to: |
bar
i Contrast and compare the five types of product-flow processes. oo
LO4.2
Describe the differences among order fulfillment processes.
_ Explain how companies should make process selection decisions.
LO4.4
Correctly place examples of products on the product-process matrix.
LO4.5
3 Describe the features of focused operat
LO4.6
Discuss the uses of mass customization and 3D printing.
Pai Eye
LO4
Ts
saee Th
7:35: Contrast pollution prevention, pollution control and pollution practices: sans
Process selection decisions determine the type of process used to make a product or service.
For example, automobiles are made using an assembly-line type of process, wine making is
a batch type of process, and a tailor shop is a job shop type of process. The important considerations required for process selection include the volume of the product and whether the
product is standard or custom. Generally speaking, high-volume products that are standard
will be made on an assembly line, and low-volume custom products will be made using a
batch or job shop process.
This chapter describes the various types of processes that can be selected and the corresponding product or service characteristics when one process or another is preferred. Two
main types of classifications are provided. The first is by the product flow, including continuous flow, assembly line, batch, job shop, and project. The second is by the type of order
fulfillment: whether the product is made-to-order, assembled-to-order, or made-to-stock.
°o
e
°
“
o
ee
°
°
°
°
°
o
°
°
LS
°
°
©
eed
ee
2 ° o
°
°
Pe
oe
-
@
°
|
°
°
°
°
°
Chapter 4 Process Selection
| 55] 1)
After process types and order fulfillment, we discuss focused operations, mass customization, and 3D printing as process choices. We also discuss environmental concerns
and the prominent role process selection plays in determining the impact of a firm on its
natural environment.
4.1
PRODUCT-FLOW CHARACTERISTICS
LO4.1 Contrastand — There are five types of product flow: continuous process, assembly line, batch, job shop,
compare the five
and project. In manufacturing, the product flow is the flow of materials, since materials
types of product-flow
are being converted into the product. In services, there is a work flow, such as a flow of
processes,
customers or information.
Continuous Processes
Continuous process refers to the so-called process industries, such as sugar, paper, oil, and
electricity. Here, the output is made in a continuous fashion and tends to be highly standard
and with very high volumes of production. Often continuous flow products are liquids or
semisolids that can be pumped or that flow from one operation to another. For example, an
oil refinery consists of miles of pipes, tanks, and distillation columns through which crude
oil is pumped and refined into gasoline, diesel, oil, lubricants, and many other products.
Continuous processes tend to make commodity products. Since it is difficult to differentiate the product, low cost becomes the “order winner” for manufacturing to compete in
very price-sensitive markets. Therefore, continuous processes tend to be highly automated,
operate at nearly full capacity, and minimize inventories and distribution costs to reduce
the total cost of manufacturing. While cost per unit of output is low, flexibility to change
product mix or product type is very limited in continuous processes.
Assembly Lines
Assembly lines are used to make only one or a few different products using inflexible
equipment and labor. Later in this chapter, we describe mass customization that allows
much more flexible assembly lines. Assembly-line flow is characterized by a linear
sequence of operations. The product moves from one step to the next in a sequential manner from beginning to end. Unlike continuous processes, in which the products are liquids
or semisolids, assembly lines make discrete products such as automobiles, refrigerators,
computers, printers, and a vast array of mass-produced consumer products. Products are
moved from one operation to the next, usually by a conveyor system.
Figure 4.1 shows how an assembly-line process is used to make a metal bracket. The
first step in production is to cut a rectangular metal blank in the required dimensions of
the bracket. At the second workstation, two holes are drilled into the metal blank. Then the
bracket is bent at a 90-degree angle, and finally, it is painted. Notice how the workstations
are placed in the proper sequence needed so that the product moves sequentially from one
end of the line to the other. Each workstation likely has a single worker performing a task,
so most workers have limited skills.
Like continuous processes, assembly-line operations are very efficient but also very
inflexible. The assembly-line operation requires high-volume and standard products. At
the same time, it is difficult to make changes in the product itself or the volume of flow,
resulting in inflexibility of operations. For example, it takes several weeks to change over
a traditional automobile assembly line to a new model. Also, the line runs at a constant
speed, and so the volume can be altered only by changing the number of hours worked or
redesigning the entire line.
bi 56 | Part Two
Process Design
FIGURE 4.1
Assembly-line flow.
The Product
(a metal bracket)
Task or workstation
eee
Product flow
Assembly-line operations generally require large amounts of capital investment and
must have high volume to justify the investment. For example, a modern plant that makes
semiconductor wafers costs over $2 billion in initial investment, and an automobile assem-
bly plant costs about $1 billion. An automobile assembly plant completes the production
of a car every minute, or about 350,000 automobiles a year, if operated on a two-shift
basis. Because of the significant amount of capital required, finance is concerned with
the choice of an assembly-line process and works closely with operations in making these
investments. Also, marketing must be geared toward mass appeal to justify producing in
high volume.
Most electronics
manufacturing
processes are
assembly lines.
nikitabuida/Shutterstock
Chapter 4
Process Selection
| 57)
Batch Flow
Batch flow is characterized by production of the product in batches or lots. Each batch of
the product travels together from one operation or work center to another. A work center is
a group of similar machines or processes used to make the product.
Figure 4.2 shows various low-volume brackets that are made using a batch process. In
this simple example, three differently shaped brackets—A, B, and C—flow through the
four work centers. Notice how bracket A requires work in all four work centers; bracket B
requires only cutting, bending, and painting; and bracket C requires cutting, drilling, and
painting. One characteristic of a batch operation is that it can be used to make many different types of products, and so more variety is typical than on an assembly line. Batches
of each of these products can have a different flow path, and some products actually skip
certain work centers. As a result, the flow is jumbled and intermittent. Contrast this to the
flow of a line process, which is regular and sequential.
Batch operations often use general-purpose equipment that is not specialized to make
just one particular product. This offers flexibility. Labor is more skilled and flexible in its
ability to make different products. As a result, a batch operation is configured with both
equipment and skilled labor to be more flexible than an assembly-line process. Lot sizes
can vary, from hundreds down to as few as one unit. As a result, batch processes can be
configured
to handle low-volume orders.
The jumbled flow of a batch operation results in considerable production scheduling
and inventory challenges. When loaded to nearly full capacity, the batch operation will
typically have high inventories as lots wait in line to be processed. High capacity utilization
will cause interference between the various lots as they wait for labor or equipment that is
assigned to another job at the time. This results in a loss of efficiency in a batch operation.
A batch operation uses a process layout because the machines and labor are organized by process types into work centers. The assembly-line process, however, uses a
FIGURE
4.2
Batch flow (metal
brackets).
CREATES
Product A
Product B
RRA
Product C
Batch A
=
Task or work center
= >
e@ =>
Batch B
Batch C
ri
i ||58 | Part Two
.
Process Design
product layout because the machines and labor are organized according to the product
flow itself. An example of a process layout is the typical high school, where classrooms
are organized according to subjects (or processes) such as math, chemistry, and music. The
students flow through the facility in batches (classes), going from one process to the next.
Batch operations are used when the volume is not high or there are many different products. Examples are furniture, boats, dishware, and other products with large variety and
low to moderate volumes. Furniture making, for example, requires many different styles
and options. Dining tables may be available in different woods and with different chair
styles. Each of these can be produced in a small batch.
Job Shop
Job shops make products to customer order by using a process layout. Thus, we consider
the job shop a special case of the batch process. In a job shop, the product is made in
batches, usually in small lot sizes, but the product must be made to customer order.
Like the batch process, a job shop uses general-purpose equipment and has a jumbled
flow. It has high flexibility for product mix and volume of production, but the costs are
generally higher since the volume and standardization are low. Typical products produced
in a job shop include plastic parts, machine components, electronic parts, and sheet metal
parts that are made-to-order.
Project
The project form of operations is used for unique or creative products. Examples of projects are concerts, construction of buildings, and production of large aircraft. Technically
speaking, the product does not flow in a project since materials and labor are brought to
the project site and the project itself is stationary. Projects are characterized by difficult
planning and scheduling problems since the product may not have been made before.
Also, projects are difficult to automate, though some general-purpose equipment may be
used. Labor must be highly skilled because of the unique nature of the product or service
being made.
Each unit is made individually and is different from the other units. Projects are used
when the customer desires customization and uniqueness. Generally speaking, the cost of
production for projects is high and sometimes difficult to control. It can be challenging up
front to specify all of the work that will be needed to complete a project.
Boeing makes large aircraft by using a project process. Each airplane is assembled at a
fixed site within the factory with materials and labor brought to the site. A complex schedule is made that must balance work across all the different aircraft being produced. The construction industry uses projects to construct buildings, roads, and bridges. Service industries
also use projects for fund-raising events, political campaigns, concerts, and art fairs.
Discussion of Process Types
The characteristics of the five processes we have been discussing—continuous, assembly
line, batch, job shop, and project—are summarized in Table 4.1. This table makes direct
comparisons among the types of processes. Notice that continuous and assembly-line operations have relatively low-skilled operators and high automation, whereas batch, job shop,
and project operations are the opposite. Capital varies from high to low across the various
process types, and the objectives and process characteristics also vary.
One way to measure the efficiency of a process is the throughput ratio (TR):
TR = Total processing time for the job
——
:
x 100%
Total time in operations
Chapter 4
TABLE 4.1
Process
Characteristics
Process Selection
| 59] |p
Continuous and
Characteristics
Assembly Line
Batch and Job Shop
Project
Batch
Single unit
Product
Order type
Continuous or very
large batch
Flow of product
Product variety
Market type
Volume
Sea
—
Sequenced
Low
Mass
High
Jumbled
High
Custom
Medium to low
None
Very high
Unique
Single unit
Labor
Operator Skills
Low
Task type
Repetitive
_ High
Nonroutine
High
Nonroutine
Pay
Medium
High
High
High
Low
Special purpose
Medium
High
General purpose
Medium
Little
General purpose
Flexibility
Cost (per unit)
Quality
Low
Low
Conformance
Medium
Medium
Conformance
High
High
Conformance
Delivery
On time
On time
On time
Capital
Investment
Inventory
Equipment
Objectives
In the numerator of the throughput ratio is total processing time for the job, which includes
only the time the job actually spends being processed by machines or labor, excluding any
waiting time between operations. The denominator includes the total time the job spends
in operations, including both processing and waiting time. Most batch and job shop operations have TRs of 10 to 20 percent, rarely higher than 40 percent. This means that a typical job spends most of its time waiting to be processed relative to the actual processing
time. In contrast, continuous and assembly-line processes have TRs of 90 to 100 percent.
The throughput ratio represents the proportion of time in operations during which value is
actively being added to the job.
At this point, examples from the housing industry may help solidify some of the process choices. At the project end of the continuum is the custom-built house. A unique plan
for it may be drawn by an architect, or existing plans
may be modified for each house built. The process is
labor-intensive, time-consuming, and costly, but it is
very flexible. General-purpose equipment is used to
complete the work.
The batch process is characterized by the production of similar houses in groups. In this case, the customer can select one of several standard houses with
only minor options such as paint colors, light fixtures,
and carpets. Such a house is usually less expensive per
square foot than a custom-built project house, but there
is less flexibility in operations to allow customers to
choose among options.
The assembly-line method of house production is
House construction can be completed using various types
characterized
by modular or factory operations. Stanof processes.
dard houses are produced in sections, in a factory, by
Alex Potemkin/Getty Images
1) |60 | Part Two
Process Design
relatively cheap labor. The use of expensive plumbers, carpenters, and electricians is largely
avoided by installing complete electrical and plumbing systems at the factory. After being
built on an assembly line, the house sections are brought to the site and erected in a few
days, using a crane. These modular houses are typically the least expensive of all and provide the least flexibility in customer choice.
Obviously, a company faces a major strategic decision in choosing the type of process
to use for the construction of houses. Most companies choose only one type of process,
unless separate divisions are formed for different processes.
4.2
APPROACHES
LO4.2
Describe the
TO ORDER FULFILLMENT
Another critical decision for operations is how the orders from customers are fulfilled:
differences among
whether
order fulfillment
advantages and disadvantages to each of these. A make-to-stock (MTS) process can provide
faster service to customers by delivering orders from available stock and at lower costs than a
make-to-order (MTO) process. But the MTO process has higher flexibility for product customization. An assemble-to-order (ATO) process is like a hybrid of these, enabling relatively
fast service to customers because there is limited work to complete once the customer order
is received. It is also flexible because the customer can specify some types of customization.
In the MTO process, the cycle of production and order fulfillment begins with the cus-
processes.
the product is made-to-order,
assembled-to-order,
or made-to-stock.
There
are
tomer order. After the order has been received, the design must be completed, if it is not
already done, and materials are ordered that are not already on hand or on order. Once the
materials begin to arrive, the order can be processed as materials and labor are added until
the order is completed. Then the order is delivered to the customer. Once the customer
pays for the order, the cycle is completed.
The key performance measure of an MTO process is the length of time it takes to design,
make, and deliver the product. This is often referred to as the lead time. Another measure
of performance in an MTO environment is the percentage of orders completed on time.
This percentage can be based on the delivery date the customer originally requested or the
date that was promised to the customer. The date requested by the customer provides, of
course, a stricter criterion.
In contrast, the MTS process has a standard product line specified by the producer, not
by the customer (see Figure 4.3). The products are carried in inventory to fulfill customer
demand immediately. Operations produces inventory in advance of actual demand in order
to have the proper products in stock when the customer order arrives. The critical management tasks are forecasting, inventory management, and production planning.
The MTS process begins with the producer specifying and producing the product. The
customer then requests a product from inventory. If the product is available in inventory, it
is delivered to the customer. If it is not available, a back order may be placed or the customer may cancel the order. A back order allows the firm to fill the order at a future date
but requires the customer to wait for the order. Ultimately, once the order is received, the
customer pays for the product and the cycle is completed.
In an MTS process customer orders cannot be identified during production. The production cycle is being operated to replenish stock. What is being produced at any point
in time may bear little resemblance to what is being ordered by customers. Production is
geared to future orders and replenishment of inventory. See the differences in the production cycles in Figure 4.3.
The key performance measure for an MTS process is the percentage of orders filled
from inventory. This is called the service level or fill rate and is typically targeted in the
Chapter 4
FIGURE 4.3
Make-to-Stock
Process Selection
| 61] |
Make-to-Order
MTS, MTO, and
ATO comparison.
ae
eee
ee
Customer
order
Customer
order
Product
Product
Assemble-to-Order
Customer
order
Product
range of 90 to 99 percent. This means, for example, that when a customer places an order,
it is immediately filled from existing stock 99 percent of the time. Other relevant performance measures are the length of time it takes to replenish inventory, inventory turnover,
capacity utilization, and the time it takes to fill a back order. The objective of an MTS
process is to meet the desired service level at minimum cost.
In summary, the MTS process is keyed to replenishment of inventory with order fulfillment from inventory, whereas the MTO process is keyed to fulfilling individual customer
orders. An MTO process can provide higher levels of product variety and has greater flexibility. The performance measures of these two fulfillment options are completely different.
The MTS process is measured by service level and efficiency in replenishing inventory,
and the MTO process is measured by its response time to customers and the ability to meet
promised customer delivery dates.
Assemble-to-order (ATO) processes are a hybrid of MTO and MTS. The subassemblies are made-to-stock, but the final assembly is made-to-order. The ATO process builds
subassemblies in advance of demand. When the customer order is received, the subassem-
blies are taken from inventory and assembled together to fill the customer order. Figure 4.3
shows how the subassemblies are built to a forecast and placed in inventory. The product
must be designed in a modular fashion for ATO to be used.
1] |62 | Part Two.
NaS
NNT
Process Design
aS
ie
Sh
si a
SPN
aie
eee
PERATIONS LEADER
co
"7
ae
ae 8 8
Culver’s
From
its
1984
beginning
in Sauk
Center,
Wisconsin,
there are now more than 700 Culver’s restaurants in the
U.S. The restaurant chain, famous for its ButterBurgers
and
frozen
menu
custard
provides
desserts,
examples
continues
of items
to grow.
that
are
Their
make-to-
stock, assemble-to-order, and make-to-order.
Some food items, such as daily soup options and salj
ads are made-to-stock. They are prepared using
batch
processes,
and
then
held
in stock
at each
.
|
EN JURGERS
yeti ts&
restaurant.
These items are ready to be served quickly once custom-
ers order them.
ButterBurgers, on the other hand, are made-to-order.
They are cooked only once a customer has ordered one,
and they require some preparation lead time. Customers
can specify which toppings to include, so the sandwich is
truly custom to their preferences.
Finally, frozen
custard
assemble-to-order
desserts
process.
are
Several
made
flavors
using an
of custard,
along with a variety of toppings, are in stock and ready
for the customer order. Once the customer specifies their
custard flavor and topping, the dessert can be quickly
assembled and served.
Culver’s uses three different types of processes—
MTS,
ATO,
and
tively.
The
choice
MTO—to
operate
of which
efficiently
use
and
depends
effec-
the
characteristics of the food item and on Culver’s strategy.
Source: www.culvers.com, 2020.
to
on
designs by Jack/Shutterstock
Some operations are moving toward assemble-to-order and make-to-order processes
for standardized products, whenever possible, by reducing production lead times. If the
standard product can be made quickly, it need not be produced ahead of time and placed in
finished-goods inventory; instead, it can be made or quickly assembled when ordered by
the customer. For example, Allen-Bradley Co. can make and ship a motor starter unit in
over 300 different configurations in one day from when it is ordered. This product, which
previously had been made-to-stock, can be assembled-to-order with large savings in inventory and improved customer service. Allen-Bradley assembles the product with a fast and
flexible assembly line.
An example of the three types of processes involves the production of diamond rings for
the jewelry business. The MTS process is used for rings that are carried in finished-goods
inventory by the jewelry store. In this case, the customer buys one of the rings from the jeweler’s stock. The ATO process is used when the customer selects the stone and then makes a
separate selection of a ring setting. The jeweler will then assemble the ring components. The
MT process is illustrated by jewelers who make rings to the customer’s design. The setting
and the stone are designed, fabricated, and assembled into a unique ring. See the Operations
Leader box on Culver’s to understand how they use these order fulfillment approaches.
Chapter 4
FIGURE 4.4
Process Selection
| 63] |]
Order penetration point.
MTO
aoe
MTO
ATO
MTS
ff
Assembly
Distribution
The type of customer order, whether MTS, MTO, or ATO, determines the order
penetration point in the supply chain where the product is linked to a specific customer
order.' There are four possibilities for the placement of the order penetration point, as shown
in Figure 4.4. For MTS operations, the order point is after final assembly is completed; therefore, the customer can only select the product from what is available in inventory. For ATO,
the order penetration point is after fabrication and before final assembly. Since the product
is assembled after the order is placed, the customer can specify some customization in terms
of the modules he or she selects. For MTO operations, the order penetration point is either
before fabrication or before ordering materials from the supplier in cases in which unique
materials or components are needed. For MTO, many types of customization are possible,
but the lead time to the customer can be longer and the product is typically more costly.
4.3
LO4.3
PROCESS SELECTION DECISIONS
Explain how
companies should
make process
selection decisions.
We have been discussing two dimensions that can be used for process classification purposes: product flow and approaches to order fulfillment. These dimensions are used to
construct the six-cell matrix shown in Table 4.2. This matrix contains the six combinations
used in practice. Multiple combinations may be used by a single firm, depending on the
products and volumes required by the market.
All six combinations are encountered in industry. Although it is common for an
assembly-line operation to make-to-stock, it can also assemble-to-order. For example, an
automobile assembly line is used to produce a large variety of different automobile options
for particular customers, as well as cars that are being made for dealer stock. Similarly, a
TABLE 4.2
Process
Characteristics
Matrix
Make-to-Stock
Automobile assembly
Oil refining
Cannery
Cafeteria
Automobile assembly
Laptop computers
Electronic components
Fast food
Batch and Job Shop
Machine shop
Wine
Glassware factory
Costume jewelry
Machine shop
Restaurant
Hospital
Custom jewelry
Project
Speculation homes
Commercial paintings
Noncommissioned art
Buildings
Movies
Ships
Continuous and
Assembly Line
' Sometimes also called the customer order decoupling point.
ni i64 | Part Two
Process Design
project form of process commonly is used to make-to-order. However, a construction company can build a few speculation houses to stock that are sold later.
In discussing the process selection decision, we shall begin with an example and generalize from there. Let us consider the construction company mentioned in Section 4.1,
which can choose to build houses using the project, batch, or assembly-line process. With
any of these processes, the company can also choose to make the houses to stock or to
order. What, then, are the factors that should be considered in making this choice?
First, the company should consider market conditions. The assembly-line approach
requires a mass market for inexpensive houses, the batch process requires a lower-volume
market for medium-priced houses, and the project process requires a market for expensive houses. Which one of these is chosen will require discussions between marketing and
operations, which are cross-functional since both market and process are affected. In the
end, matching the process to the market will be a key strategic decision.
Second, the company should consider capital requirements. The assembly-line process
will require a great deal more capital than will the project or batch flow. The assembly
line requires capital for a factory and equipment and to finance the partially completed
or finished houses. By contrast, construction of custom project houses requires much less
capital since only one house or a few houses are being built at any one time and no factory is needed. The finance function will be intimately involved with operations in making
these capital decisions.
The third factor that should be considered is the availability and cost of labor. The
project and batch processes require costly skilled labor, such as plumbers, electricians, and
carpenters. The factory line approach requires relatively inexpensive, low-skilled labor.
Unionization may affect both the supply and the cost of labor. The human resources function will be involved in these decisions with operations because of the employee selection,
training, and compensation issues involved.
Finally, the company should consider the state of technology for both process and product. Are innovations likely to come along that will make a process obsolete before the costs
are recovered? Assessment of these conditions is part of risk evaluation for the process.
In summary, four factors appear to influence process selection:
1. Market conditions
2. Capital requirements
3. Labor
4. Technology
4.4
LO4.4
PRODUCT-PROCESS
Correctly
place examples of
products on the
product-process
matrix.
STRATEGY
To this point, we have been treating process decisions as static, as if the organization
makes the decision one time and then uses the selected process forever. Actually, process
decisions are dynamic, since processes evolve over time. Furthermore, process decisions
are Closely related to product decisions, and both products and markets change over time.
Hayes and Wheelwright have proposed a product-process matrix that describes the
dynamic nature of product and process choices (see Figure 4.5). On the product dimension (horizontal) of the matrix is the life cycle of a typical product, ranging from a unique,
one-of-a-kind product to a high-volume, standardized product. A product typically evolves
from the left side to the right side of the matrix.
On the process dimension (vertical) of the matrix the various processes are represented,
ranging from the project to a continuous process. The process has a life cycle similar to the
Chapter 4
Process Selection
| 65] |
FIGURE 4.5 _ Product-process matrix.
Source: Adapted from Hayes, Robert H., and Steven C. Wheelwright. “Line Manufacturing Process and Product Life Cycles.” Harvard Business Review, January—February
1979, pp. 133-140.
Unique,
one-of-a-kind
product
Low volume,
low
standardization
Low volume,
multiple
products
Higher volume,
few major
products
High volume, high
standardization,
commodity
Project
Job Shop
Assembly
Line
Continuous
product life cycle, evolving from a unique project type of production at the top of the matrix
to a continuous process at the bottom. Many products have followed the product and process
life cycle. Automobiles were made in a job shop environment in the early 1900s before
Henry Ford invented the moving assembly line. Electronics often are produced in batches
until the volume becomes sufficient to support an assembly-line process. Most product life
cycles will not require processes to evolve from project all the way to continuous, but processes often do need to change one category or more among those listed in Figure 4.5.
Most organizations should position themselves on the diagonal of the matrix. This
means that a low-volume product with high variety would be produced by a project or
job shop, but a highly standardized product with high volume would be produced by an
assembly-line or continuous process. The diagonal of the matrix represents a logical match
between the product and the process.
The product-process matrix represents the strategic choices available to firms in both
product and process dimensions. Often, strategy is represented as consisting only of product choice. But the process can provide a unique capability that helps the firm compete
in the market. Thus, a position on the matrix represents a strategic combination of both
product and process. This type of strategic position requires cross-functional cooperation
between marketing and operations to ensure that both product and process choices have
been considered. For example, if a firm typically produces standard beverages in high
|
66 | Part Two
Process Design
volume, moving into a new market that requires small batches in a wide variety of flavors
may be at odds with its current production capabilities. The firm may need to invest in new
equipment, alter existing equipment, or reconsider whether it is capable of competing in
this new market.
A firm might be tempted to move down the diagonal ahead of its competitors and thus
gain competitive advantage at lower cost. This can be a good idea if the customer is ready
to accept a more standardized product. If the customer prefers more customization, the
firm may be forced to move back up the diagonal to remain competitive.
All firms in any particular industry, however, do not occupy the same spot on the diagonal
of the product-process matrix. Some firms may choose to stay in the upper left-hand corner
of the matrix; others may move down the diagonal. One example of this behavior is the handheld calculator business. Hewlett-Packard has chosen to stay with low-volume, high-variety,
and high-priced calculators while the rest of the industry has moved down the diagonal
toward highly standardized, high-volume, and low-priced calculators. The Hewlett-Packard
calculators, which are suited to various specialized market niches such as accounting, surveying, and electrical engineering, can command high prices at relatively low volumes.
4.5
LO4.5
FOCUSED OPERATIONS
Describe the
features of focused
operations.
Often a company will have products that are produced in a variety of volumes and with various levels of standardization. When a company mixes all these products in the same factory,
it can lead to disaster. Skinner, who originated the idea of the focused factory, tells the story
of an electronic instrument company that made low-volume custom autopilot instruments
and high-volume standard fuel gauges in the same plant.” After years of losing money on
the fuel gauges, management decided as a last resort to separate the fuel gauge production
from the autopilot production by building a wall down the center of the plant. They also
assigned separate quality control and materials management staff to each product as well
as separate direct labor, supervision, and equipment. As a result of these changes, the fuel
gauges became profitable in four months and the autopilots also improved their profitability.
The problem essentially resulted from two different sets of objectives being mixed in
operations: one of low cost for the fuel gauges and one of superior product performance and
innovation for the autopilot. The autopilot production needed more stringent requirements
in quality, materials, and skills than did the fuel gauges. Before focus, the fuel gauge costs
were inflated and efforts were not directed to each product separately. When their production
processes were separated, and thus the specific production needs of each product were separated, each process could better respond to its particular customer and market requirements.
Services can also be focused. An example of a focused factory for services is Midwest
Orthopedic Specialty Hospital near Milwaukee, Wisconsin, which specializes in providing
medical treatments on bones and joints. Their specialists perform surgeries and provide
nonsurgical treatments for problems ranging from arthritis to sports injuries. With their narrower range of services, they are a focused factory when compared with a general hospital,
which provides more wide-ranging services. By narrowing the types of customers treated,
the orthopedic hospital narrows its need for different types of rooms and the equipment
needed to treat various illnesses. The needed range of employee skills is also more limited.
Lack of focus in manufacturing plants and service operations has resulted from attempting to accomplish too many different goals with the same facility or operation. In some
cases, product proliferation in the markets served by the company has led to incompatible
2 W. Skinner, “The Focused Factory,” Harvard Business Review (May—June
1974), pp.
113-21.
Chapter 4 Process Selection
aed
| 671 |
G
——
products being produced together in the same facility. The solution
may be to arrange each product as a plant-within-a-plant (PWP),
which entails setting up a process for one product or product line
separately from other processes within the same facility. This is
done by physically separating product flow, using separate workGlobal Headquarters
forces and equipment, and separate management and support staff.
The result is two or more small plants within the larger plant, like
the Skinner example earlier in this section. The firm may sacrifice
Honeywell uses focused factories to compete
some economies of scale but will do a better job of meeting market
by using several PWPs within one building.
requirements and improving profitability.
©Spencer Platt/Getty Images
:
:
;
A refrigeration
company used the concept of focused operations
by organizing two PWPs inside the same building. It separated
compressor production for high-volume, standardized, and mature products from compressor production for low-volume and customized products. The two PWPs used many of the
same components and shared their shipping department, but separating their processes
enabled both to improve their performance. In this case, one factory was divided into two
separate focused factories.
Service operations also can be focused by assigning different types of service products to different facilities. For example, in an insurance business selling both high-priced,
service-intensive policies and low-priced, commodity-style policies, using the same set of
workers to serve both could cause trouble. The high-priced policies might get too little service while the low-priced policies receive too much service. The solution to this problem is
to separate the service for these policies into two different facilities or two different PWPs
with separate workforces and appropriate expectations for each type of policy.
4.6
MASS CUSTOMIZATION
LO4.6 Discuss
the uses of mass
customization and
3D printing
To this point, we have been discussing traditional forms of production processes. However, with new technologies that enable flexible manufacturing, mass customization is now
possible. Mass customization is a strategy to provide high volumes of custom products
in lot sizes of one, that is, each production unit is unique. At first blush, mass customiza-
tion (custom products and high volume) appears to be an oxymoron, two words that are
incompatible, like jumbo shrimp or a deafening silence. But this dichotomy between mass
production and customization can be overcome by using modern technologies, including computers, robotics, modular design, and the Internet. See the Operations Leader box
regarding Nike’s mass customized shoes.
Traditional mass production is built on economies of scale by means of a high-volume
standardized product with few options. With economies of scale, the more production the
lower the average unit cost. By contrast, mass customization depends on economies of
scope—that is, a high variety of products from a single process. Economies of scope also
reduces the average unit cost due to production of many products by the same process.
Consequently, mass customization comes from a different economic basis, a common process rather than a common product.
Customization refers to making a different product for each customer with the cost
of this usually being high. But mass customization is providing customized products at
approximately the same cost as mass production. This is a stringent requirement and means
that some products cannot be mass customized because the cost would be too high.
One of the early examples of mass customization is Paris Miki, a global retailer of
standard and mass customized eyeglasses. The mass customization process begins with
It||68 | Part Two
Process Design
OPERATIONS
LEADER
Nike Does It
Nike
among
and
other
Nike-owned
athletic
offer customized
shoe
shoes
Shoes with a customized “fit”
are significantly more elusive. While
Converse,
brands,
that can
be
ordered online for a reasonable price
and delivered within a few weeks.
Customers can select from numerous
fabric or leather colors and patterns
Sen
on various pieces of the shoe, as well
obsession 24k/StockimoiAlamy, Stock Photo
roe
EE
ar
as the colors of laces, stitching, and
soles. These shoes, with their many customizable options,
are an example of successful mass customization.
Cou
hee
a
there are firms offering custom-fitted
shoes, they sell at prices reflecting
the significant work to individually
size the shoe, most likely performed
in a job shop. These shoes are custom, but not mass ee
Mass customization gives custom|ers many options, as well as the enjoyment of designing
and using a product with their own personal stamp on it.
design of the eyewear by taking a digital photo of the customer and recommending a style
of lenses that fits the customer’s face. An optometrist then adjusts the lenses to fit the customer’s preference. Finally, the customer selects options for nose bridges, hinges, and arms
for the frame. The customer receives a photo of the proposed eyewear. Finally, a technician
makes the lenses and frames at the store within one hour.
There are three forms of mass customization:
1. Modular production and assemble-to-order (ATO).
2. Fast changeover (nearly zero setup time between orders).
3. Postponement of options.
Modular production can provide a variety of options by using an assemble-to-order
process. For example, when Dell receives a computer order, the company assembles standard modules or components rapidly to fulfill the customer’s order. The order is then
shipped and the customer receives it in a few days. But this requires modular design, as
well as modular production. Dell also uses the same process to make standard computers
for stock and shipment to retail stores.
Fast changeover is the form of mass customization used by Paris Miki for its glasses,
where moving from one customer order to the next is very quick, with little or no setup
between them. In this case, it is critical that each order is uniquely identified by a bar code,
or other identifier, that specifies the customer’s options. It is also essential to have nearly zero
changeover time on equipment so that a lot size of a single unit can be produced economically.
Postponement is used to defer a portion of the production until the point of delivery.
For example, customized T-shirt shops can put a unique design on a T-shirt at the point
of purchase. Hewlett-Packard printers receive their final configuration for various voltages and power supplies at U.S. or overseas warehouses before delivery. Postponement
makes it possible to ship standard units anywhere in the world and customize them at the
last minute.
From a manufacturing point of view, mass customization has changed the dynamics
of the product-process matrix. Flexible automation makes it possible to make small lot
sizes along with large lot sizes without a great cost penalty. Thus, with mass customization a firm can operate over a wider range of product choices without major changes to its
Chapter 4
Process Selection
| 69] |
process. This amounts to a wider horizontal operating
position on the matrix.
There
are limits to mass
customization,
too. As
some early mass customizers discontinue their customized products, there is disagreement on whether
mass customization is a viable strategy for most firms.
For example, automobile producers have struggled to
make mass customization a reality. While they have
successfully modularized the design of their cars and
reduced changeover times, they have been less successful in linking customer preferences to the production process. In most states, manufacturers cannot sell
directly to customers, and so the factory must comMass customization is not for everyone. Automobile makers
municate
have struggled to mass-customize cars.
Also, customers
with customers
via the dealership network.
©Monty Rakusen/Getty Images
but painting occurs early in the production process.
Customers like getting a new car right from the dealer
like to choose the color of the car,
lot and not waiting a few weeks for the car to be made and shipped to them. These issues
illustrate the significant challenges to successful mass customization.
Cross-functional integration is a key enabler of mass customization. Modularity and
other forms of mass customization create an increased need for information exchange and
cooperation among R&D, manufacturing, and marketing functions. Most new practices
that involve both cross-functional work and new technologies are successful only in certain
Situations and conditions.
4.7
3D PRINTING AND ADDITIVE MANUFACTURING
3D printing is rapidly growing as a production technology. Also called additive
manufacturing, this technology deposits successive layers of plastic, metal, or ceramic
material to build a 3-dimensional solid object. An inkjet type of printer sprays the material on thin layers until the product is built up to its normal size. It can make complex
shapes with holes, interior spaces, irregular contours, or other difficult dimensions. Prior
to 3D printing these objects would have been machined from metal or made from expensive plastic injection molds. Regarding process type, additive manufacturing fits as a job
shop in the product-process matrix.
3D printing has applications in a variety of fields including medical implants, aerospace, apparel, architectural display models, prototypes, and art objects to name only a few.
Spare parts are another application since they can be made at the point of use rather than
carried in inventory.
In medical science, 3D printing is playing a major role in the industry shift toward personalized medicine. Dental implants are made to fit a particular patient. Orthopedists are
testing designs and materials for printing customized joints for joint replacement surgeries.
Ophthalmologists are 3D printing semiconductor polymer materials that they use to create “bionic” eyes for blind patients. And transplant specialists are even developing options
to print human cells into entire organs, such as hearts and kidneys, for patients whose own
need replacement.
In additive manufacturing the product design is transmitted via a digital 3D model
from the design department to the 3D printer. As a result the technology can be used for
170
| Part Two.
Process Design
3D printer and printed objects
cookelma/Getty Images; Maruna Skoropadska/123RF
distributed manufacturing, as designs can be transmitted to remote locations and produced
there instead of shipping the parts or products long distances.
Additive manufacturing is in its infancy. Rather than replacing traditional manufacturing it will probably complement it in specific situations. The ability to transmit a
design around the globe and make it in many different countries is appealing. The price of
3D printers has been dramatically reduced to less than $1000 for the most basic machine
to hundreds of thousands for more advanced equipment.
Already, UPS is experimenting with 3D printing. They have purchased 100 industrial grade 3D printers for their Louisville, Kentucky, air hub to make everything from
airplane parts to iPhone parts. Since 3D printing can shorten supply chains and reduce
inventory, UPS wants to understand whether to treat 3D printing as a threat or a service
opportunity. FedEx and Amazon have their own studies under way. UPS also has 3D
printers at some of its retail shops, where it can make products and prototypes directly
from customer designs.
As the price of additive manufacturing is reduced it can support many low volume customized parts or products. Additive manufacturing will permit relocalization of production in both developed and developing countries. Countries and industries can build their
additive manufacturing around digital technologies. This will result in shifts in investment,
labor, and location of manufacturing.
4.8
LO4.7
ENVIRONMENTAL
Contrast
pollution prevention,
pollution control,
and pollution
practices.
CONCERNS
More than any other area of the firm, operations affects the natural environment. From the
selection of inputs into the transformation process to the process outputs and by-products,
decisions about the production process can have a significant environmental impact. With
increasing regulation and consumer scrutiny, firms are continuously considering how to
meet the demands of these stakeholders.
When firms are making process selection decisions, they must consider the environmental impact. We can think about three areas of decisions that affect environmental impact.
First, there are technologies for pollution prevention. These structural investments
reduce or eliminate pollutants from the production process. Process choice decisions, for
example, using solar power versus coal-based power, determine the types of pollution output the process will create. Pollution prevention investments might include designing the
Chapter 4 Process Selection
| 74) i
process to waste fewer raw materials, redesigning products or processes to reduce pollutants, or investing in equipment that requires less energy input. These are instances of
preventing pollution from occurring.
Second, pollution control technologies are also structural investments. They differ
from pollution prevention technologies in that they are used to treat or dispose of pollutants and harmful by-products that are outputs from the process. These technologies most
commonly are added to existing processes that were designed and purchased in the past,
perhaps when firms were less concerned. Pollution control extends an existing process by
adding another step to deal with the waste by-products from the process.
The third category of process decisions related to environmental impact is pollution
practices that affect the way processes are used. Such practices include retraining workers
to use the existing process in a new manner and increasing cross-functional coordination
to seek creative and innovation improvements in environmental impact. These practices
also include monitoring and reporting systems related to the way processes operate. Other
practices may involve changes in the supply chain, for example, selecting new suppliers
who certify their materials are obtained using sustainable methods.
Many processes have been improved greatly over the years to lessen their impact on the
environment. From changes in process inputs—for example, using consumer waste paper
rather than trees to manufacture new paper—to innovative use of used materials—for
example, making park benches out of recycled plastic bottles—operations plays a major
role in managing environmental issues.
Examples of firms employing these process decisions are abundant. Shipping giant
Maersk is pursuing a zero emissions goal, General Mills uses significant water saving
strategies, and Toyota has a net positive impact plan—environmental performance even
beyond zero emissions. For more details see the sustainability link in the Learning Enrichment box at the end of this chapter.
Other environmental concerns related to process choice include the following:
1. Recycling outputs—finding uses for process by-products.
send their food waste to pig farms.
Example:
Public schools
2. Recycled inputs—using materials from other processes as inputs into a process. Example: Andersen Windows uses wood shavings that are by-products of one process as
input together with a thermoplastic polymer to form Fibrex, a composite material used
to make window frames and other products.
3. Remanufacturing—restoring and reusing some product components in the production
of new products. Example: Caterpillar salvages and reuses many parts in its industrial
equipment.
4.9
CROSS-FUNCTIONAL
DECISION MAKING
There are many cross-functional interactions in process selection decisions. Marketing has
a large stake in process selection decisions due to their understanding of the needs and
sizes of markets. Process choices require large capital investments and thus make it difficult to change the process quickly. In many cases, the markets the firm faces may be
changing faster than the firm can recover the capital investment from process choices.
Thus, marketing should work closely with operations in these decisions to ensure that both
current and future market demands can be met, along with environmental impacts.
The critical role of marketing in estimating and managing future demand is apparent.
Although forecasting is always an inexact science, some scenario planning should be done
ti| 72 | Part Two.
Process Design
to estimate the appropriate product and process response for different levels of demand.
This will make it possible to manage the risk inherent in process choices and the associated
capital investment. Marketing should also be aware of the process choice implications by
managing demand to support the process choices that are made.
Finance has a key role in process selection decisions because of the capital investment
required. Process selection choices should be subjected to standard cash-flow and present
value analysis. This will ensure that any contemplated process choices will provide the
required returns on capital at an acceptable risk. Finance will also be required to raise the
capital once the process selection decision is made and to provide capital for future investments as the product, process, and environmental challenges evolve over time.
The human resources function plays a key role in providing the human capital that is
consistent with the process selection choices. Different processes require different labor
from unskilled to highly skilled and different labor specialties. The human resources function must hire, train, and guide management of the workforce so that it is coordinated with
the process choices made by operations.
Information systems and accounting professionals should be aware that different processes
have different performance measures and different data requirements. The information and
accounting system designed for an MTO process will not work in an MTS operation; a job
shop information system will not work in an assembly-line environment, since the information used for scheduling and inventory control depends on the type of process selected.
Because large investments are required in hardware and software, information systems and
accounting decisions must be closely coordinated with process selection choices.
We have shown that process selection choices affect all parts of the firm. They are
strategic decisions that determine the future capabilities of the firm and thus involve all
functional areas along with general management. With proper cross-functional coordination, the processes selected can offer competitive advantage to the firm and will be supported by all functions.
4.10
KEY POINTS AND TERMS
This chapter has emphasized how process design and selection can meet the strategic needs
of the business. These key points have been made in the chapter:
*
*
There are five types of processes: continuous, assembly line, batch, job shop, and project. The continuous and assembly-line processes are suited to high-volume standard
products that are produced at low cost with limited flexibility. The batch and job shop
processes are suited to low- to moderate-volume products that are customized or produced in a high variety. The disadvantage of batch processes is the jumbled flow, which
reduces throughput and efficiency. The project process is best for unique or creative
products that are made one at a time. It requires intensive planning and scheduling and
generally results in costly products or services.
The second dimension of process is the type of order fulfillment: make-to-stock, maketo-order, or the hybrid assemble-to-order. With MTS, the replenishment cycle for inventory is separate from the customer order cycle. In contrast, the MTO process is set
in motion by customer orders and geared to delivery performance. The MTS process
provides standard products, whereas the MTO process is suited to custom orders. The
ATO process makes subassemblies in advance for inventory and assembles them into a
final product when ordered by the customer.
Chapter 4 Process Selection
REAR
¢
The order penetration point determines the point at which the customer order enters the
production process. This is related to whether the process is designed to be MTS, ATO,
or MTO.
¢
The combination of product flow and type of order fulfillment provides six types of
processes. Selection from among these six requires consideration of market conditions,
capital requirements, labor, and technology. Taking into account these factors, the process selection decision is always strategic and cross-functional in nature.
Key Terms
¢
The product-process matrix provides a dynamic view of the process selection decision
by considering the life cycle of both products and processes. Strategy is defined by a
position on the matrix for the firm’s product and process. The matrix helps provide
coordination between marketing decisions about product and operations decisions concerning the process.
*
Focused operations are used to separate products and processes that have different requirements in terms of the production process or the markets served. Each
type of process or product family should be assigned to a different facility or
plant-within-a-plant.
¢
Mass customization is the ability to make a customized product at approximately the
same cost as a mass-produced product. This can be done for some products by using
flexible automation, robotics, modular design, and information systems. There are three
types of mass customization: modular production/assemble-to-order, fast changeover,
and postponement.
¢
3D printing and additive manufacturing can rapidly produce prototypes or unique custom products from a 3D digital design. Rather than replace traditional manufacturing
it will complement it. Product designs can be transmitted and products can be printed
where they are needed.
¢
Environmental concerns are a major challenge related to process design. When making decisions about processes, firms must consider whether they will develop structural processes to prevent or control pollution or use pollution practices to manage
these matters. Process decisions must account for plans to recycle and remanufacture
products.
¢
Process selection decisions are highly cross-functional in nature because they affect
human resources, capital, information systems, and the ability of the firm to deliver
products to the market. Therefore, all functions should be knowledgeable about process
choices and the impact of process selection on their particular functional area and the
environment.
Continuous process
Assembly line 55
Batch 57
General-purpose
equipment 57
Jumbled flow 57
Process layout 57
Product layout 58
Job shop 58
Project 58
Throughput ratio 58
55
Make-to-stock 60
Make-to-order 60
Lead time 60
Back order 60
Service level 60
Assemble-to-order 61
Order penetration point 63
Product-process matrix 65
Focused factory 66
Plant-within-a-plant 67
Mass customization 67
Economies of scale 67
Economies of scope 67
Modular production 68
Fast changeover 68
Postponement 68
3D printing 69
Additive manufacturing 69
Pollution prevention 70
Pollution control 71
Pollution practices 71
Part Two
Process Design
LEARNING
ENRICHMENT
Jelly Belly Factory Tour
https://youtu.be/tSPi2qS4hsQ
(for self-study or
Kroger Bakery Tour
instructor assignments)
https://youtu.be/uhBS02j73wk
Chrysler Virtual Reality
https://youtu.be/INNSEIkcpvM
Project to Build the Icelandic Super Dam
https://youtu.be/anpbr4mms0E
Make-to-Stock versus Make-to-Order
https://youtu.be/k_PkOh-vxDU
Factory Home Construction
https://youtu.be/LDU3s9KfOHY
Mass Customization
https://youtu.be/IxqJdo YSWfw
Zero Emissions Factory
https://youtu.be/Ogg7uLInOCU
Discussion Questions
ike Classify the following types of processes as continuous,
assembly line, batch, job shop, or project:
a. Doctor’s office
Automatic car wash
College curriculum
Studying for an exam
Registration for classes
Ck
mos
ec
Electric utility
. Why are assembly-line processes usually so much more
efficient but less flexible than batch processes? Give
three reasons.
. The rate of productivity improvement in the service
industries has been much lower than in manufacturing.
Can this be attributed to process selection decisions?
What are the challenges involved in using more
efficient processes in service industries?
. Several industries—including those that produce
furniture, houses, sailboats, and fashion clothing—
have progressed very little down the diagonal of the
product-process matrix toward more standardized and
efficient. Why do you think this is so?
. Compare an expensive restaurant, fast-food restaurant,
and cafeteria in terms of process characteristics, such as
product type, labor, capital, and operations objectives.
. A company is in the business of making souvenir
spoons to customer order. The customers select the
size of the spoons and may specify the design to
be embossed on them. One or more spoons may be
ordered. The company is considering going into the
make-to-stock spoon business for souvenir spoons
and everyday tableware as well. What will it have
to do differently in terms of planning production?
How is the business likely to change?
. What are the strategies of the following organizations?
Is the strategy defined in terms of product or process
or both?
a.
McDonald’s
b. AT&T Inc.
c.
General Motors
d. Harvard Business School
. Suppose that a firm is considering moving from a
batch process to an assembly-line process to better
Chapter 4 Process Selection
meet evolving market needs. What concerns might the
following functions have about this proposed process
change: marketing, finance, human resources, account-
ing, and information systems?
Give an example of mass customization not discussed
in the chapter.
10. What techniques or approaches can be used to achieve
mass customization in practice?
Ik, Search the Internet to find applications of 3D printing.
What industries and what product examples do you
find?
1
What is the difference between economies of scale and
economies of scope? How do firms consider these when
investing in processes?
. What are some of the classic signs of an unfocused
operation?
14. What are the pros and cons of organizing a plantwithin-a-plant?
15. Why should operations be concerned with environmental issues?
16. What are the main ways in which processes are
managed to accommodate environmental goals?
CHE AUR ATA ESR
Service Process Design
oi
ot
it it ei et i
A
a
LO5.4
Define the attributes of the service-profit chain.
nn
nner
The service economy accounts for more than 80 percent of jobs in the U.S. and in most
industrialized economies in Europe and Asia today. Yet service production often receives
little emphasis in many business and operations management courses. Increased emphaSis on service process design is needed to reflect the importance of services in modern
economies.
Unlike manufacturing processes, we see service processes every day. As customers, we
participate in the process and immediately know whether we are receiving good or bad service. Unfortunately, world-class service is rare. For example, was your most recent retail
Chapter 5
Service Process Design
771
OPERATIONS LEADER
City of Fort Collins
The city of Fort Collins, Colorado, is a community of
160,000 people in the front range of the Rocky Mountains, and it was a winner of a Malcolm Baldrige National
Quality Award in 2017. The city has 2400 employees
and 2200 volunteers. Its largest employers are Colorado
State University, Hewlett Packard, Poudre Valley Health
System, Poudre School District, Eastman Kodak, and
Avago Technologies.
The city has excellent living conditions, ranking in the
top 10 percent of cities nationally on the best place to
live, best place to work, quality of culture, air quality, and
visual attractiveness.
Fort Collins is financially sound, with a credit rating of
Aaa by Moody’s—a rating maintained by only 4 percent
of governments. It has a positive ratio of revenue to budget, low debt and increasing tax revenues. The city’s tax
rates are among the lowest in Colorado.
Fort Collins is a national leader for environmental
goals. Community energy usage decreased by 12 percent
annually for the past three years, despite a population
growth of 7 percent.
Fort Collins has the wide participation of citizens with
citizen involvement on 27 advisory boards and commissions. Senior leaders use a “co-creation model” to work
collaboratively
with
residents
and
businesses
on
solu-
Source: City of Fort Collins
Fort
Collins
built around
has
an
16 processes
improvement
that advance
approach
the goals of
the city. It uses public engagement
every two years to
set the Mission, Vision, and Values
(MVV). This is fol-
lowed by identifying measurable outcomes and financial
plans.
sites provide government transparency and accountabil-
The city adopts its financial plans based on the Budgeting for Outcomes (BCO) process that involves city
staff members at all levels and citizens to discuss priorities that support the MVV.
ity including the Community Dashboard and Scorecard of
City’s performance.
Source: https://www.nist.gov/baldrige/city-fort-collins, 2019.
tions to urban challenges.
Information is widely available to citizens online. Online
service pleasant? Do you enjoy waiting in the doctor’s office? And what do you think of
most airline service?
Most business graduates will be employed in service industries and work closely with
operations and supply chain professionals. It is important to learn about service since it 1s
highly relevant to all students, managers, and consumers.
Services are delivered by a wide variety of organizations—businesses selling services to
consumers (restaurants, appliance repair) and to other businesses (consulting, accounting),
non-profit services (health care, education), and government services (licensing, police
protection). The operations function in these organizations varies widely in the resources
used to produce these services, but there are common elements that allow us to study the
processes used in service organizations. For an example of world-class service, see the
Operations Leader box on Fort Collins, Colorado.
What can be done to improve services? Service process design is an essential ingredient of better service delivery. We take the ideas of process selection and extend them to
|
78
Part Two
Process Design
services. We expand the discussion into the domains of service product offerings, service system designs, globalization of services, service guarantees, and the important role
played by employees and technology in service organizations.
5.1
LO5.1
DEFINING SERVICE
Differentiate
the characteristics
of a service
organization from
a manufacturing
organization.
Most definitions of service stress the intangibility of the offering. Services are indeed
intangible; that is, their processes create value for customers by performing transformations that do not result in a physical entity (product). However, services can be difficult to
define and cannot be easily quantified; for example, do hospital patients consume one service or multiple services as they receive tests and treatments—perhaps numerous in quantity? Rather than specify a formal definition of a service, it is important to consider the
characteristics of such processes and their implications for both managers and customers.
Simultaneous production and consumption is a critical characteristic of services
because it means that the customer may be in the production system while production takes
place. The customer can introduce uncertainty into the process by placing demands on the
service provider at the time of production. Also, the simultaneity of production and consumption means that most services cannot be stored; and the service must often be located
near the customer so that the customer can travel to the service provider or vice versa.
Exceptions are communications such as call centers, TV, Internet services, and electricity
services that can be provided over long distances.
It is important to distinguish between service processes that are front office and those
that are back office. Processes that require the presence of or interaction with the customer
are front office service processes. The importance of simultaneous production and consumption therefore applies to front office services because the customer is participating in
the process. For example, dental assistants and dentists provide front office services when
interacting with customers. This interaction within the service process between providers
and customers is critical to service process design but quite foreign to manufacturers.
Back office services, in contrast, can be performed separately from their consumption
by the customer and, therefore, do not have to accommodate interaction with the customer.
Most transaction processing in banks and testing patient samples in medical offices are
back office processes that are not produced and consumed simultaneously but become
valuable to the customer some time after the work is performed.
Because characteristics of services vary widely and the extent of interaction between
the provider and the customer can also vary greatly, it is difficult to generalize about services. However, they are clearly different from products that are outputs of manufacturing.
Some of the important contrasts between products and services are shown in Table 5.1.
TABLE
5.1
Differences Between
a Product and a
Service
A Product
A Service
A product is tangible
Production precedes consumption
A service is intangible
Production and consumption are simultaneous
A product can be transported
A product can be stored in inventory
Ownership is transferred at the time of purchase
A product can be resold
A product can be demonstrated before purchase
The seller produces
*Exceptions are electricity and communications services.
A service cannot be transported (though
producers can be)*
A service cannot be stored
Ownership generally is not transferred
No resale is possible
A service does not exist before purchase
The buyer can perform part of the production
Chapter 5
5.2
Service Process Design
| 79} |
SERVICE-PRODUCT BUNDLE
LOS.2 Explain
the elements of a
service-product
Bunce!
Before the process to deliver a service is designed, the service-product bundle must be
defined. The service-product bundle consists of three elements:
1. The service (explicit service).
2. The psychological benefits of the service (implicit service).
Oo
. The physical goods (facilitating goods).
Most
services
come
bundled
with explicit services,
implicit services, and facilitating
goods. For example, when customers go to a fast-food restaurant, they receive both an
explicit service, which they hope is fast and accurate, and a facilitating good, the food. In
this case, the implicit service is how customers feel about the interaction and the pleasantness of the surroundings. Many services have fixed facilitating goods, such as the building
and equipment that are used but not consumed during delivery.
In the case of a subway ride, the explicit service is the transportation from one place
to another and includes customer perceptions and experiences, such as the sound, sight,
smell, and feel of the ride. The implicit service is the sense of well-being and security that
the subway ideally provides. Finally, the subway car is the facilitating good. It is important
in the design of the service not to overemphasize one piece of the service-product bundle
and neglect the other elements.
Most services require a more complex design than a subway ride. Consider the explicit
services, implicit services, and facilitating goods for a luxury hotel. The explicit services
include both basic services and amenities. These would be provided by the bellhop, concierge, restaurant, maid, room
service, front desk, and Internet. The implicit service is a
sense of safety, a caring staff, and the atmosphere of a luxury hotel. The facilitating goods
are the hotel building, the food, the beds, the room, and physical surroundings.
Figure 5.1 provides more examples of a variety of service-product bundles. Notice that
some bundles are mostly all service with few goods (e.g., consulting and haircuts) while
the automobile industry provides mostly goods with only a little service. Here, we include
the automobile as an example of a service-product bundle because the purchase of a new
auto includes several service elements that customers recognize and pay for. The auto bundle includes not only the physical product but also the ability to test-drive and finance the
product at the dealership in addition to the manufacturer warranty that covers the auto.
The combination of these service elements with the
product makes up what we consider a service-product
(or product-service!) bundle. One might also consider
Luxury hotel guests expect a full service-product bundle.
the maintenance and repair service offered by the dealer
after the sale.
The task for the operations function, before delivering any services to customers, is to design the service
delivery system. That system includes all the processes
that will be used to deliver services, including details
such as the technology used in the process design, the
types of employees needed, and even the appearance of
the employees and facilities. While operations can control both the explicit service and the facilitating goods,
implicit services are obviously harder to control (and
may
vary
sais greatly from one customer to another). ThereJ
Jupiterimages/Getty Images
fore, it 1S important
that management
use
the means
it
Ea
Part Two
Process Design
FIGURE 5.1
Comparison of various
products and services
packages.
Services
Products
100%
75%
50%
25%
0%
25%
50%
75%
100%
McGraw-Hill Education
has available (e.g., technology or employees) to design the implicit services into the service delivery system.
An important point to emphasize is that the design of the service to be delivered and
the design of the service delivery system are intertwined and, often, should be designed
concurrently. Also, the delivery of a service is a simultaneous marketing and operations act
that requires both the right visual cues and well-functioning processes. Therefore, crossfunctional cooperation is essential to service design and delivery.
Like products, services have supply chains, although they may be less concerned with
the flow of the physical product and more concerned with the flow of work, customer,
and information. Services cannot be stored, but they do use inventory, and so they rely
on product-based supply chains to provide that inventory. For example, a hospital patient
requires service processes for explicit services (surgery, perhaps) and additionally the
work flow of outsourced
lab tests, information
and financial flows from insurers, and
coordination of work and information as the patient is discharged from the hospital to a
rehabilitation center. Such a complex network of supply chain activities mirrors the activities of product-based supply chains but usually includes both tangible product flows and
intangible work and information flows.
5.3
LO5.3
SERVICE DELIVERY SYSTEM MATRIX
Organize a
variety of service
offerings into the
service delivery
system matrix.
There are many ways to think about services, the options they offer customers, and the variety of ways in which they can be delivered. Some services can be delivered in only one
standardized way and every customer gets more or less the same service. Other services are
highly customized to customer requests, and exactly the same service is virtually never
repeated for another customer. The challenge for management is to design the right service
process so that the service delivery system is matched to the requirements of its customers.
Both customer preferences and design requirements of the service delivery system are
contained in the service delivery system matrix, which is shown in Figure 5.2. On the top
of the matrix is the dimension of customer wants and needs, which captures the service
package (or service-product bundle) customers are seeking. This dimension incorporates
the uniqueness of demands from one customer to another, an indication of the uncertainty
Chapter 5
FIGURE 5.2
Service Process Design
| 84] |]
Service delivery system matrix.
Source: Collier, D-A. and Meyer, S.M., A service positioning matrix, International Journal of Operations and Production Management, 18(12): 1223-44.
Customer Wants and Needs in the Service Package
Highly customized with
unique process
sequence. Customer has
great decision-making
power.
Standard with options,
using moderately
repeatable sequence.
Customer has some
decision-making power.
Standardized with
highly repeatable
process sequence.
Customer has low
decision-making power.
Many process
pathways.
Jumbled flows,
complex work
with many
exceptions.
Moderate
number of
process
pathways.
Flexible flows
with some
dominant paths,
moderate work
complexity.
Design
SystemLimited number
Delivery
Service
of process
pathways. Line
flows, low work
complexity.
and variation introduced into operations by individual customers. Customers with basically the same wants and needs can be served by processes that are highly standardized and
routinized, whereas customers with unique wants and needs must be served by processes
that allow variety and high levels of customization.
The vertical side of the service delivery system matrix represents the operations
service system and denotes the number of different pathways that service customers can
take in the service process. This can vary from a single pathway or small number of pathways to a virtually infinite number. A small number of pathways allows few options in how
the service is delivered; however, an infinite number of pathways allows the service to be
different each time it is delivered.
When both dimensions of the service delivery system matrix are considered, three
types of services can be identified. Customer-routed services are those in which customers want a unique, highly customized experience. Customers have a great deal of
4 Eg
Part Two
Process Design
decision-making power to determine the components of the service as well as how and
when and the sequence in which they are delivered. For these services, each customer
wants a different set of experiences, and the service process must allow a great deal of
personal discretion and interaction with the customer. These services are carried out using
highly flexible processes and may rely on highly trained workers to deliver the right set of
experiences to match customer wants and needs. Personal trainers, Internet shopping, and
museums are examples of customer-routed services. Customer-routed services are similar to those delivered by job shops for manufacturing products, in terms of allowance for
customization.
In the midranges of both customer wants and needs and service delivery system design,
co-routed services offer customers a moderate number of choices, using moderately standardized processes. Medical and stockbroker services fit in this category. A golf course
is another example of a co-routed service in which management has designed the course
to be played in a standardized sequence (from Hole | to Hole 18), but within the service
delivery system customers have a reasonable degree of decision-making power in how they
choose to play.
Finally, highly standardized services are delivered using a design for provider-routed
services. These services are characterized by processes that allow few options during service delivery and are designed for customers whose needs are very similar to one another.
Grocery store self-checkout is a service delivery system with a very limited number of
pathways from which customers may choose. It provides a limited set of services, and there
is little customer discretion in using it. Customers whose needs are not met by self-checkout
must use employee-operated checkout stations. Eating at McDonald’s and getting a blood
test are other examples. Provider-routed services are therefore similar in nature to a manufacturing assembly-line process. We refer to them as provider-routed because the provider,
either an individual or an organization, decides how a particular service will be carried out.
The service delivery system matrix is intended not only to classify the different types of
services but also to indicate how the operations function task differs among services. For
example, provider-routed services may require management attention to automation and
capital investment, but customer-routed services may require more attention to management of human resources and flexible technology issues.
The service delivery system matrix suggests that service firms will generally be
located on the diagonal, indicating alignment between the service package and the service
process. Both the choice of which customer segments to serve (horizontal dimension) and
decisions regarding design of the service delivery system (vertical dimension) are strategic in nature. Marketing, operations, and human resources functions must work closely to
ensure that external opportunities and internal capabilities have been considered during
strategic planning.
The major difference between the service delivery system matrix and the productprocess matrix that guides the selection and design of manufacturing processes is that the
design of the service delivery system generally does not vary with customer volume. In
the product-process matrix, the volume and customization of the product offering are the
major factors in determining the most appropriate production process. In contrast, services
often are delivered using the same process whether they are produced in small or large
volumes. For example, very similar processes are used for a medical service such as setting
a broken leg regardless of whether the service is delivered at a large 2000-bed hospital,
which has many such patients, or in a smaller 120-bed hospital. Similarly, fast-food restaurants treat customers the same way regardless of the number of customers they serve
and regardless of customer order size. To increase volume, fast-food restaurants simply
open more locations, but the service process is the same. The degree of customization of
Chapter 5
Service Process Design
| 83] iy
a service, rather than volume, is the main characteris-
tic that affects the design of the service process and the
way the service is delivered.
Self-service by customers is also a consideration in
service delivery system design. Customers may serve as
labor at key points in a service process, such as bagging
their own groceries, or they may complete an entire
service process independently, as occurs when they
fill their tanks at a self-service gas station. Self-service
usually benefits the firm as customers provide “free”
labor during service delivery. For self-service to be a
successful component of service delivery system design
firms must design their service processes carefully for
Grocery self-service is a provider-routed service.
both simplicity and customer satisfaction.
Syda Productions/Shutterstock
Self-service is possible for any of the types of services defined in the service delivery system matrix,
from simple standardized services to highly customized services. A key issue for operations managers is designing self-service opportunities that customers are both willing and
able to perform. While the relatively simple self-service offered at ATMs appeals to a wide
range of customer segments, having to pull one’s retail selections from warehouse shelving (e.g., at IKEA stores) may limit the appeal of the retail service for some segments. An
understanding of the needs of a firm’s target customer segments must serve as a guide to
the right service delivery system design.
5.4
CUSTOMER
LO5.4 Describe the
effect onthe service
delivery system of
customer contact.
CONTACT
| We now look at interactions between customers and service organizations in detail to
— understand how the extent of customer contact relates to service processes. With lowcontact services, it is possible to separate a service into two portions: a service creation or
production portion and a service consumption or delivery portion. By doing so, the customer can be removed from the service creation portion. Separating the customer from the
service production portion allows for greater standardization of processes and therefore
better efficiency. Examples of low-contact services are processing of online orders and
ATM transactions. As indicated above, these services are usually designed using a
provider-routed approach. See Figure 5.3, in which low-contact services are referred to as
buffered core because these services are designed to be buffered or removed from interactions with the customer.
At the other end of the contact spectrum, high-contact services involve the customer
during the production of the service. Examples are dentistry, haircutting, and consulting.
In these services, the customer can introduce uncertainty into the process with a result-
ing loss of efficiency. For example, a customer may impose unique requirements on the
service provider, resulting in a need for more processing time. In this case, the service
delivery system design typically will be customer-routed unless customization has been
limited by the provider. These interactions are referred to as reactive in Figure 5.3 because
the service delivery system must react to customer requests.
In the middle ground of customer contact, permeable systems have processes that are
penetrated by customers in fairly restricted ways, usually via telephone or limited face-toface contact. Here, limited interaction with customers allows some customer preferences to
be met. But such accommodation 1s restricted to maintain process efficiency.
S ||384| Part Two
FIGURE 5.3
Process Design
Customer contact matrix.
ed, 2017.
Source: Adapted from Jacobs, F.R. and Chase, R.B., Operations and Supply Chain Management, McGraw-Hill Education, 15th
Degree of customer/server contact
Reactive
system (much)
Permeable
system (some)
Buffered
core (none)
Production
efficiency
Sales
opportunity
Worker
Requirements
Clerical
skills
Helping
skills
Verbal
skills
Procedural
skills
Trade
skills
Diagnostic
skills
Focus of
Paper
handling
Demand
mgmt.
Scripting
calls
Flow
control
Capacity
mgmt.
Client
mix
Office
automation
Routing
methods
Computer
databases
Electronic
aids
Selfserve
Client/
worker teams
Operations
Innovations
Operations managers must be concerned with the amount of customer contact because
higher levels of customer contact can introduce variability into a process. Variability is a
challenge for operations managers because it makes capacity planning more difficult and
can result in waiting lines. Table 5.2 defines five types of customer-introduced variability.
Service firms that try to accommodate all types of customer-introduced uncertainty may
find that the cost of delivering the service begins to spiral out of control. Instead, they must
learn to manage the uncertainty, either by using creative means to reduce it or by finding
low-cost means of accommodating it.
A few examples provide insight into managing variability. Arrival variability results in
empty restaurant seats at certain times of day and full seats and a waiting line at other
times. Customer arrivals are somewhat random, but usually clustered around standard meal
times. A reservation system can help to manage arrival variability by shifting some customers to somewhat earlier and later than peak standard times. Thus, reservation systems can
be effective for managing customer arrival uncertainty. Capability variability, on the other
hand, is observed in hospital patients’ varying abilities to move about, feed themselves, and
take care of their basic needs such as getting a drink or using the bathroom. Hospitals usually hire low-wage staff to assist with these needs to keep costs low while reserving more
expensive labor (like nurses) for tasks that require more extensive licensing.
Chapter 5
TABLE 5.2
Types of Customer-
*
Introduced
*
Variability
shed
+
’
Source: F. Frei, “Breaking the
Trade-off Between Efficiency
and Service,” Harvard Business
Review, November 2006,
pp. 92-101,
Paras
aD
rege
Service Process Design
eae
| 35] |)
Noes
Arrival variability—uncertainty in when customers will arrive to consume a service.
=
Request variability—uncertainty in what customers will ask for in the service-product bundle.
Capability variability—uncertainty in the ability of customers to participate in a service.
*
Effort variability—uncertainty in the willingness of customers to perform appropriate actions.
*
Subjective-preference variability—uncertainty in the intangible preferences of customers in how
y
:
3
;
service is carried out.
&
me
re.
5
i
_
The relationship between customer contact and process efficiency can be stated as
follows:
Potential inefficiency = f (degree of customer contact)
The measure of the degree of customer contact is the amount of time it takes for the service
to be produced, delivered, and consumed by the customer. As this time increases, the delivery process is increasingly inefficient. High contact may be costly in terms of lost efficiency, but it may offer opportunities to increase sales to customers, resulting in increased
revenue for the service firm, as shown in Figure 5.3. For example, consultants often have a
high degree of contact with clients, and such interactions provide them with opportunities
for additional consulting work and therefore additional revenue.
When possible, high-contact and low-contact portions of service delivery systems
should be separated to create front office (high contact) and back office (low contact) processes. Front office operations require intensive customer interactions, whereas the back
office can operate more efficiently away from the customer. The separation of high-contact
and low-contact services is an application of the principle of focused operations.
There are several characteristics of high-contact and low-contact services:
¢ Low-contact services are used when face-to-face
interaction is not required, for example, shipping
operations or check processing in banks.
*
Low-contact services should use employees with
technical skills, efficient processing routines, and
standardization processes. High-contact services
require employees who are flexible, personable, and
willing to work with the customer (the smile factor).
Airline service is a high-contact service with some customer
variability.
Ryan McVay/Getty Images
*
Low-contact operations can work at average demand
levels and smooth out the peaks and valleys in demand.
Providers of high-contact service must respond immediately as demand occurs in peak situations.
¢
High-contact services generally require higher prices
and more customization due to the variability that
customers introduce into the service.
While customer contact is an important ingredient of
service delivery system design, it is not the only consideration. Contact with customers becomes increasingly
challenging to manage with increases in the total duration of interactions and the richness of the information
exchanged during interactions. The t nature of uncer=
tainty introduced by the customer is also of critical
i} |86 | Part Two
Process Design
importance. For example, contact can be high, but if the customer interface is standardized
or the customer provides self-service, efficiency is still possible. In fast-food restaurants
the degree of customer contact is relatively high, but the nature of the contact is highly
controlled in contrast to a fine-dining restaurant, where there is more uncertainty in what
the customer may request. Thus, high contact by itself is not always inefficient; it becomes
inefficient when customers introduce uncertainty or do not provide self-service.
5.5
SERVICE RECOVERY AND GUARANTEES
LO5.5
Explain service
recovery and service
guarantees.
Service recovery is an important element of service management when there is a service
failure—in other words, when something goes wrong during the delivery of a service. Service recovery consists of the actions necessary to compensate for the failure and restore, if
possible, the service requested by the customer. For example, when there is a power failure, service recovery includes the time it takes for the electric company to restore power. In
a restaurant, if the waiter spills soup on a customer’s lap, service recovery includes helping
to dry the clothes with napkins, an apology, and perhaps an offer to dry-clean the clothes at
the restaurant’s expense. Often when the service recovery is swift, properly performed and
appropriate in the customer’s eyes, the customer accepts the service failure and recovery
and is satisfied with the overall service experience. Because service failure from time to
time is nearly inevitable, service firms must design recovery processes to ensure that such
actions are taken consistently. See the Operations Leader box on UPS for an example of a
satisfying service recovery.
ag
hal tall
:
4
/
Mh
:
a=
e am
ho .
UPS Service Recovery
A customer who usually received packages at the office
ordered a large storage unit and needed
it delivered to
her house. Tara Hunt, then an Intuit executive, called UPS
to check on it and was told that during the holiday rush,
some packages are not delivered until as late as 9 p.m.
Agitated,
she
posted
a message
on
Twitter
about
waiting for UPS and that she could not walk her dog
while she waited, not wanting to miss the delivery. Tony
Hsieh, CEO of Zappos, was following her Tweets after
having met her previously. He was having dinner with the
UPS president and relayed her frustration to him. Five
minutes later, the UPS executive called and connected
her with an operations manager to arrange delivery for
the following morning.
At 9 a.m. sharp, the doorbell rang. Not only was the
package delivered, but the UPS employee brought flow-
ers and chocolates, along with treats and toys for her
dog! Ms. Hunt says she now goes out of her way to use
Hii
20g Pought shoes at Zappos the vety Next day
Tara Hunt has since moved on to a successful career
as a digital marketing
professional,
researcher,
author,
roberto galan/123RF
and speaker. She is passionate about helping businesses
grow by leveraging social networks.
Source: Adapted from: “A Social Networker’s Story,”
—BusinessWeek March 2 2009, p. 30; wwwtarahunt.com
2019,
Chapter 5
Service Process Design
| 37] |
Many companies solidify their service recovery processes by offering service guarantees as a way to define the service and ensure its satisfactory delivery to the customer. A
service guarantee is like a product guarantee, except customers cannot return a service if
they do not like it. For example, if you do not like your haircut because it’s too short, you
have to live with it until your hair grows out.
A service guarantee has two components: (a) a promise of what service will be delivered and (b) what the payout or service recovery will be if the promise is not fulfilled. The
value of the guarantee to the service firm offering it is that the promise defines exactly
what service needs to be delivered correctly. The firm must design its processes and train
its workers to meet the expected service every time it is delivered.
FedEx Corporation, for example, has a money-back service guarantee for its shipping
services within the United States. Packages will be delivered by the published delivery time
(or quoted time, as in the case of FedEx SameDay) or the service price will be refunded to
the customer. This service guarantee defines exactly what the organization must achieve
and what happens when a service failure occurs. Another example is Atlantic Fasteners,
a distributor of hardware in Massachusetts, whose service guarantee for on-time delivery
is: “We deliver defect-free in-stock fasteners on time as promised or we give you a $100
credit.” Atlantic Fasteners has an incredible 99.96 percent reliability and accuracy rating in
meeting its service guarantee. Other companies may offer somewhat less-precise service
guarantees. For example, hotels may give you a free night’s stay if you are not satisfied. A
restaurant server may give you a free dessert or a free meal if you are not satisfied with the
food. These service guarantees are not as precise in guiding operational activities as the
FedEx or the Atlantic Fasteners service guarantees, but they are, nonetheless, better than
not having a service guarantee at all.
The service guarantee is not an advertising gimmick or simply a way for customers to
get their money back if they are not satisfied. It is an assurance that the service provider
will perform as promised. And if a customer is not satisfied and requests the payout from
the service guarantee, the service provider can use the request as feedback to understand
both what customers expect from the service and how the service delivery system must be
changed to better match customer expectations.
A carefully crafted service guarantee benefits both customers and the service provider.
For the former, a service guarantee reduces the risk in purchasing the service. Customers
know exactly what service they are buying, when a service is deemed to have failed, and
how they will be compensated. For the latter, a service guarantee clarifies explicitly what
is to be achieved by the service. Such clarification of intent helps guide the design of the
service delivery system, specifies the extent of service recovery required upon service
failure, and presents a clear vision to motivate employees to deliver high service quality. A
service guarantee can, moreover, reward the service provider with loyal customers.
5.6
TECHNOLOGY
LO 5.6 Evaluate
the role of
technology in service
management.
FOR SERVICES
Technology in services takes many forms. First, there is the technology embedded in the
service system itself through equipment and automation. Almost all services have been
automated, at least to some extent, in industries from medicine to airlines to retail. See
Table 5.3 for examples of automation in various industries.
Pushing automation forward for services 1s not the right design for all types of services.
While the resulting service may be more efficient (more customers served for a lower percustomer cost), automation can change the nature of the service and reduce the number
of selling opportunities. The market ultimately will determine how much automation is
oy t=t=4)
«Part Two
Process Design
TABLE 5.3
Examples of
Technologies
Supporting Service
Delivery
Service
Examples of Technology
Medicine
MRI scanners, electronic
Intensive care unit monitoring systems,
medical records, automated diagnostic testing, pacemakers,
robotic surgery, telemedicine
Telecommunications
Cellular phones, TV, video conferencing, satellite communications,
Retail
Point-of-sale scanners, inventory control systems, radio-frequency
e-mail, Internet, cloud-based storage and computing
identification (RFID), electronic payment systems, self-service
Education
Legal
checkout
Digital libraries, Internet, interactive learning
Digital searches, databases for evidence
Hotels
Fast checkout, keycard security, reservations systems, heating/
Airlines
ventilation controls, Internet access
Air traffic control system, electronic cockpits, reservation systems,
in-flight technology services
reasonable and how automation is used. For example, with fast food available everywhere,
many customers still prefer traditional restaurants with high human touch, better food, and
a larger variety of choices.
Some argue that managers should view employees, not automation, as the center of the
service delivery system. They suggest that service companies do the following:
¢
Use automation to support front-line employees, not to monitor or replace them.
¢
Value investments in employees, for example, hiring a highly educated workforce.
¢
Make recruitment and training as important for front-line workers as for managers and
staff employees, and link compensation to performance for employees at every level.
This employee-focused approach emphasizes people as the center of service. It would be
appropriate for some customized services that have co-routed or customer-routed service
delivery systems.
Artificial
Intelligence
Artificial intelligence is a second type of technology that is rapidly being implemented
in services. Artificial intelligence (AI) is software and hardware that is programmed to
exhibit aspects of human intelligence. Al mimics human cognitive functions such as problem solving, learning, and creativity. Recently, Al has been providing services that imitate
human service employees. Some of these applications are discussed here along with the
potential for future use. AI promises to save human effort while reducing the costs of delivering service. When advanced AI is used it may be difficult to tell the difference between
a human service provider and an AI agent or AI bot (bot is short for web-enabled robot).
There are four levels of AI development:
¢
Routine: Repetitive tasks
*
Analytical: Problem solving and learning
¢
Intuitive: Think creatively
* ‘Empathetic: Respond emotionally
Technology enables service delivery.
Ariel Skelley/Blend Images/Getty Images
These four levels are generally hierarchical from routine
at the lowest level of AI to empathetic at the highest
level.
Routine AI deals with repetitive tasks in service that
can be provided with a minimal degree of learning or
adaptation. It can rely on observations to act and react
Chapter 5
Service Process Design
| 89} |]
repetitively as it performs simple, standardized, repetitive, and transactional service tasks.
An example is the service provided by a vacuum cleaner robot that replaces the human
effort in vacuuming a room or the entire house. The vacuum cleaner follows an embedded
program but can make adjustments for obstacles and obstructions. Other examples of routine AI are health care diagnosis assistance to physicians by pre-programmed scripts, AI to
flag bank checks outside the norm, and AI to prevent fraud in credit cards. The advantage
of routine AI relative to humans is the consistency of service that can be provided by AI
agents, although the services are limited to routine types.
A common use of routine AI is customer phone or chat service provided by a conversational bot that answers. According to the latest figures from IBM, companies spend
$1.3 trillion on customer service calls each year.' They estimate that up to 80 percent of
the 265 billion calls or chats have routine questions that can be answered by a conversational bot; the remainder could be handled by customer service agents. This leaves agents
available to handle delicate questions and complex calls increasing customer satisfaction
and lowering agent workload. These conversational bots will become even more useful in
the future as they become more sophisticated. But some customers may prefer talking to a
human, so this option must always be available. As a matter of fact, some companies may
provide humans answering the phones as part of their marketing and service strategy, even
though it is more expensive.
Analytical AI systems learn and adapt systematically based on data. They use rulebased logic to learn using information processing and logical reasoning. Employees who
use Analytical AI software require technical training and skills on data and analysis. Big
data analytics utilizing very large data sets is part of this type of AI. Often, but not always, a
combination of software and human interaction is required for analytical AI. For example,
H&R block uses the IBM Watson computer to assist tax preparers. The computer system
combs through 74,000 pages of the U.S. tax code along with thousands of yearly tax law
changes. By combining the power of tax professionals and Watson’s technology, improved
service to the customer is provided uncovering every deduction the customer is entitled to
receive. Another example of Analytical AI is IBM’s Deep Blue chess player. In this case,
the software analyzes all of the possibilities for a given chess move and provides a suggestion for the best move. Deep Blue has beaten some of the best chess masters in the world.
Big data analytics AI is widely used in sports in conjunction with analysts to provide
support for decision making including recruiting players for specific positions and decisions about the best strategies in particular games. Practically every professional sport
has data analytics software and employees who know how to use it. This type of AI is
appropriately used when data analysis leads to learning as part of the service provided.
Ultimately, analytical AI can be applied to services such as accounting, financial analysis,
and engineering.
Intuitive AI learns and adapts intuitively based on understanding. Neural networks are
used to provide deep learning. This type of AI is useful for complex, chaotic, and idiosyncratic tasks that cannot be easily completed. Using neural networks, the software can
learn things it has not been taught before based on intuition and experience, the same way
that humans learn. It must learn based on the context and can extract ideas from similar
situations.
Intuitive AI can support professionals who engage in creative thinking or problem
solving such as doctors, lawyers, travel agents, and consultants who offer services to
customers. The Associated Press reported that it is using AI bots to cover minor league
' https://www.business.com/articles/future-of-conversation-bots/.
a ||90 | Part Two
Process Design
baseball games. Automatically generated reports are
being offered based on statistics from 142 games in 13
leagues. It took a year for baseball experts to monitor
the reports for accuracy and intuition being produced by
the AI agent. After one year, the AI agents were used to
actually report on the games with confidence they were
as accurate and intuitive as human reporters.
Empathetic AI is useful in decision making that
requires emotions. The AI agent can learn to adapt
empathetically based on experience. It is important
for high touch service that includes highly interactive
social and emotional communications. One example is
the humanoid robot Sophia that interacts with a client using multiple visual and hearing sensors to act
as a human would in a conversation including blinkPepper interacts with a customer.
ing eyes, hand and arm motions, and moving lips for
VTT Studio/Shutterstock
speaking. It is reminiscent of C-3PO in Star Wars.
Sophia is a platform for general intelligence R&D
intended to evolve to scalable commercial production. Another empathetic bot, named
Pepper, can recognize the principal human emotions and adapt its behavior to the mood
of the client. To date more than 140 SoftBank Mobile stores in Japan are using Pepper as
anew way of welcoming, informing, and amusing their customers. Pleasant and likeable,
Pepper represents a humanoid companion or friendly service representative. Pepper can
recognize your face, speak, hear you, and move around. Pepper behaves as if it likes to
interact with customers.
Artificial intelligence will have amazing advances and applications in the future. To
replace or assist humans the customer should believe the response is from a human service provider. While AI will provide some services more efficiently at an acceptable level
of interaction, other services will remain to be provided by humans. Two strategic questions for all service operations are: (1) In what situations will customers accept service
by AI agents, and (2) what should be the nature of the interaction between human service
providers and AI agents? These questions need to be answered by management in using
AI software to provide service.
5.7
GLOBALIZATION OF SERVICES
LO 5.7
Appraise
how globalization
has affected
services.
We turn to the outsourcing and offshoring of services. Outsourcing of services is placing
services outside the firm that were previously done inside the firm, such as HR and accounting. Offshoring, in contrast, is the export of services to other countries. Both of these practices have been enabled largely by advances in information technology and communications.
As a result, seamless delivery of services can be provided globally that are transparent to the
user. Services include call centers, finance and accounting, IT infrastructure, HR, medical
services, and knowledge services such as product support, R&D, and analytics.
Offshoring of services presents many of the same opportunities and challenges as
offshoring manufacturing. Potential upsides include lower costs and focusing on core
competencies; downsides include coordination costs and loss of direct control. One difference from manufacturing is due to the intangibility of services. While physical products can be inspected upon receipt from a supplier, services can be more challenging to
inspect and therefore require additional mechanisms to ensure that service performance
Chapter 5
Service Process Design
| 94} |
standards are met. One possibility is to request customer feedback immediately after the
service is performed.
Service offshoring requires attention to both people and processes. The expectations of the
target customers are a major factor in selecting processes for offshoring. Management must be
sure to link customer wants and needs (refer back to the service delivery system matrix) with
the appropriate process. Understanding the desires of the customers will lead to a set of strategically linked decisions on the appropriate mix of processes, technology, and service workers.
Services should be offshored with both the advantages and disadvantages in mind. If
a service is offshored, it should be done in partnership with the supplier to ensure quality
rather than merely seeking the lowest cost. Offshoring requires an expectation that the supplier can continue to fulfill the service requirements and provide the quality required, not
only the lower cost. It is a lasting relationship that is desired.
Shifting cost and the quality effects due to international borders can be a concern. What
appears to be a significant cost advantage could disappear because of trade differences
that occur. For example, wages are bound to rise in low-cost countries over time. Also, the
cost advantages can be drastically overestimated owing to the need to monitor and support
overseas suppliers, or the quality of service can suffer. Thus careful evaluation is necessary; otherwise the firm may find itself reshoring services due to cost or quality concerns.
Economists have engaged in heated debates over offshoring, which often overlap with
free trade arguments. Theoretically, free trade benefits both the origin and destination
country. The origin country benefits from lower cost imports and the destination country benefits from jobs. On the other hand, offshoring has sparked controversy in the origin country due to job losses and wage erosion. Economists who are against offshoring
argue that currency manipulation, unfair trade regulations, and tariffs benefit the destination country to the disadvantage of the origin country. Therefore, managers must consider
not only the cost advantages of offshoring, but the possible backlash and negative social
effects of job loss in the origin country, when trade is not actually free.
There are two types of offshoring for services. Some services such as call centers and
accounting have been commoditized and are routinely outsourced. Many of them have
been moved to low-cost countries such as India and the Philippines. These are transactionbased services where prices are driven down leading to outsourcing in an offshore location.
The second type of offshoring is professional services that have been outsourced and
moved offshore. One example is lawyers in India that were hired by the Tusker Group in
Austin, Texas, to review 400,000 documents
for $25 per hour. Similar work could cost
more than $125 per hour in the United States. A second
example is reading scans by radiologists. When radiologists at Altoona Hospital in Altoona, Pennsylvania,
Legal work is being outsourced to India.
Mustafa Quarishi/AP Images
could not keep up with reading scans, they hired a group
based in Bangalore, India. The Indian radiologists were
trained in the U.S. and costs were substantially lower
with no loss in quality, plus scans could be read at night
due to time differences and be ready the next morning
in the U.S.
Globalization of services is ongoing. Many of the
same issues as manufacturing outsourcing must be
considered but many services, outside of those already
mentioned, are inherently immune to offshoring and
must be produced in the country of origin. When offshoring services, both the advantages and disadvantages must be considered.
| |92 | Part Two
5.8
Process Design
SERVICE PROFITABILITY AND EMPLOYEES
Service profitability has been linked to focusing on customers and employees as paramount in importance. Managers should focus in particular on the front-line employees who
deliver the service, the technology that supports them, training, and customer satisfaction.
When these factors are linked and improved, profits will be improved according to the
service-profit chain shown in Figure 5.4.
Customer loyalty is the key to revenue growth and profitability. Although focusing on
increasing market share is sometimes touted as the key to profitability, customer loyalty
is an additional factor of equal or greater importance. Even a 5 percent increase in loyal
customers can increase profits in many industries by 25 to 85 percent.
The service-profit chain shows that customer loyalty is driven by satisfied customers. Naturally, if customers are satisfied, they will likely provide repeat business and also tell others
about their positive experiences. Customers who report very high levels of satisfaction affect
profitability through their loyalty much more than do customers who are merely satisfied.
External service value leads directly to satisfied customers. External service value is the
benefit customers receive minus their cost incurred in obtaining the service, which includes
LO5.8 Define the
attributes of the
service-profit chain.
not only the price but also the costs of finding the service, traveling to the service location,
waiting for the service, and correcting any service problems encountered. For example, Progressive Corporation, an insurance company, has created CAT (catastrophe) teams that fly
to the scene of major accidents to quickly provide support services such as transportation,
housing, and claims handling. By avoiding legal costs and putting money quickly into the
hands of the insured parties, the CAT team more than makes up for the costs of travel and
maintaining the team. The CAT team provides value to customers, helping to explain why
Progressive has one of the highest margins in the property-and-casualty insurance industry.
Next, we see the important role that front-line, customer-facing employees play in the
service-profit chain. Employee productivity, retention, and satisfaction work together
to create high service value to customers. Productive employees lower the costs of
operations and ensure satisfied customers when supported by management and appropriate
FIGURE 5.4
The service-profit chain.
Source: Heskett, J.L., et al., “Putting the Service-Profit Chain to Work,” Harvard Business Review, March-April 1994, p. 166.
t
in
2
Tg)
ABS
| Internal
service
quality
ks
é
Ps
;
:
Revenue
growth
*
Employee
satisfaction | |
|
nal |)
Customer
satisfaction
[| | Customer
|
loyalty
Profitability
O Job design
O Employee selection
and development
O Employee rewards
and recognition
O Technology for service
O Service concept:
Results for customers
O Retention
O Repeat business
O Referral
O Service designed and
delivered to meet
targeted customers’ needs
:
Chapter 5
Service Process Design
| 93] |
technology and systems. For example, Southwest Airlines has the most productive employees in the airline industry. As a result of short routes, fast turnaround, and productive
employees, Southwest has 40 percent greater aircraft and pilot utilization than its competitors. Employee retention and low employee turnover help drive productivity and customer
value. Traditionally, the cost of employee turnover considers only the cost of recruiting,
hiring, and training replacements. In reality, the greatest cost of turnover is the lost productivity and decreased customer satisfaction associated with new employees. At Southwest
Airlines, customer perceptions of service value are high, based on low fares, on-time service, and friendly and helpful employees.
Employee retention and productivity are driven by having satisfied employees. For
example, a study of insurance company employees found that 30 percent of dissatisfied
employees intended to leave the company, a potential turnover rate three times that of satisfied employees. Satisfied employees are the result of internal service quality. This includes
the employee selection process, job design, reward systems, and the technology used to
support service workers. Focusing management’s attention on improving internal service
quality systems to provide support for employees in conducting their work can improve
employee satisfaction, productivity, and turnover. Employees will be satisfied with their
jobs when they feel that they can act on behalf of customers. This will lead to both employee
and customer satisfaction. This is achieved in part by giving front-line employees latitude
to use resources to meet customer needs immediately. For example, in Ritz-Carlton Hotels,
front-line employees are authorized to spend up to $2000 to satisfy a customer need.
The service-profit chain illustrates the central role of employees in delivering services to
customers. During the delivery of services, employees are the “face” of the company and
their satisfaction 1s directly observed by customers, often influencing customer perceptions
of the service as well. This differentiates services from manufacturing, since manufacturing
employees rarely have direct contact with customers. A manufacturing employee’s effect
on customer satisfaction is through the product that the customer may receive days, weeks,
or months later. However, the morale, attitude, and satisfaction of service employees is
directly—and immediately—trelated to customer satisfaction and loyalty. There is no buffer
zone between service employees and customers in high- or medium-contact services.
The service delivery system design should reflect this direct contact between service
employees and customers. This can be done by providing real-time (during the service
delivery) tools such as access to customer information to help service employees perform
their jobs. For example, bank tellers who can quickly scan relevant portions of a customer account while the customer is face-to-face or on the
telephone with them can present banking products that fit
the customer profile. Such personalized selling opportunities
tend to be more successful than
low-contact marketing, such
as mail or e-mail. Services
can also be improved through
so-called “smile training” in
which service workers are
trained to be nice to customers
and seek their satisfaction even
in pressure situations. Service
Warrah’s uses superior customer service to increase profits.
workers should be rewarded for — Leonard Zhukovsky/123RF
f ||94 | Part Two
Process Design
both productivity and customer satisfaction. The service-profit chain shows that these two
measurements are not in conflict; rather, productivity can actually drive customer satisfaction through faster and better service at lower cost. Both satisfaction and productivity
can be achieved not by trying harder but by improving the use of people, technology, and
service process flows.
An interesting application of the service-profit chain is being made at Harrah’s in Las
Vegas, Nevada. Traditionally, the gambling industry catered to high rollers. A former
professor at the Harvard Business School, Gary Loveman, Harrah’s then-CEO, revolutionized the gaming industry by showing that the key to profitability and growth is not
only catering to the high rollers but also providing exceptional service to all customers.
“People don’t understand that gaming itself is fundamentally entertainment,” says Loveman. Gamblers from all walks of life like to play the odds again and again . . . and again!
This will be done by having satisfied employees provide exceptional service to satisfied
and loyal customers.
5.9
KEY POINTS AND TERMS
This chapter emphasizes the design of service delivery systems. The key points are as follows:
¢
A front office service is defined by simultaneous production and consumption. This
makes it impossible to store a service for later use, and a service often must be located
near customers, with the exception of technology-delivered services such as communication and electricity. The customer is part of the service process during production and
may introduce inefficiencies, but at the same time sales opportunities.
*
Back office services can be buffered from the uncertainty introduced by customers and
therefore can be designed for higher efficiency.
¢
Services consist of bundles of services and goods, including explicit services, implicit
services, and facilitating goods. It is important to provide the right mix of these three
elements and not overlook the psychological (implicit) component of service.
¢
The service delivery system matrix is formed by juxtaposing customer wants and needs
in terms of customizing a service against the service delivery system. The combination
of service package and service process design elements results in three main service
types: customer-routed services, co-routed services, and provider-routed services. Each
of these service types has different requirements for operations managers to meet.
*
Customer contact depends on the duration and degree of interaction between the provider and customer. Generally, high-contact services are performed in the front office;
low-contact services are performed away from the customer in the back office. In addition to contact, the degree of uncertainty introduced by the customer will have an impact
on efficiency within the service system.
When services are not delivered as promised, the firm should provide quick and helpful
service recovery. A service guarantee can be offered to ensure that the customer understands what is promised and what constitutes an error in service delivery. The service
guarantee provides a way for operations to know exactly what is required.
Technology allows the automation of services for greater efficiency that can result
in lower costs and more uniform quality. Artificial intelligence offers the prospect
of providing some services that are indistinguishable from human service providers.
*
*
Chapter 5
Service Process Design
| 95] |]
Nevertheless, both AI and service employees will be needed depending on the type of
service provided.
Key Terms
LEARNING
ENRICHMENT
(for self-study or
instructor assignments)
*
Service outsourcing and offshoring are trends that present opportunities and challenges.
Offshoring often is used to obtain talent from widespread global locations, particularly
for information or communications intensive services. A strategic approach should be
taken to offshoring and outsourcing, not just chasing low-cost labor, since changing
costs, quality, and reliable suppliers in the long-run should be considered.
*
The service-profit chain indicates how value provided to the customer drives customer
satisfaction and loyalty, which leads to revenue growth and profitability. External customer value is the result of employees who are productive, satisfied, and retained by
the firm. These employees must be appropriately selected, trained, and rewarded. The
service-profit chain indicates the crucial role of employees in delivering services and
financial results.
Intangibility 78
Simultaneous production and
consumption 78
Front office 78
Back office 78
Service-product bundle 79
Explicit services 79
Implicit services 79
Facilitating goods 79
Service delivery system
matrix 80
Operations service system 81
Customer-routed services 81
Co-routed services 82
Provider-routed services 82
Self-service 83
Customer contact 83
Customer-introduced
variability 84
Service recovery 86
Service guarantee 87
Artificial intelligence
(AD 88
Routine AI 88
Analytical AI 89
Intuitive AI 89
Empathetic AI 90
Outsourcing of services 90
Offshoring 90
Service-profit chain 92
Customer loyalty 92
External service value 92
Productive employees 93
Employee retention 93
Satisfied employees 93
Internal service quality 93
What Is a Service Export?
http://www.tradeready.ca/20 16/trade-takeaways/service-exportssuddenly-important/
—What Is Service Recovery in Health Care?
Web Link
Web Link
https://www.ahrq.gov/cahps/quality-improvement/improvementguide/6-strategies-for-improving/customer-service/strateg
y6pservice-recovery.html
Machine Learning and AI in Financial Services
https://youtu. be/xefkx 1O7p6A
Video
6:00
AI Products with User Centered Design — J.P. Morgan
https://youtu.be/Ku8x lina_N4
Video
3:20
The Service System Design Matrix
www.mhhe.com/servicesystem
Video
10:58
i ||96 | Part Two.
Process Design
Discussion Questions
Li Classify the following services by their degree of
b. Trucking firm
customer contact (high, medium, or low). Also,
c. Grocery store
determine how much uncertainty the customer
introduces into the service delivery system by the
d. Appliance repair firm
Define a possible service guarantee for each of the
following services:
ability to make customized service demands (high,
medium, or low).
a. College classes
a. Check-clearing in a bank
b. A theater performance
b. Bank teller
c. Buying a used car
c. Bank loan officer
Give an example of the service-profit chain for movie
theaters. Define each of the components in the chain
and explain how you would measure each.
Locate each of the following services on the service
delivery system matrix:
a. Vending machine
Why is the service-profit chain important to operations
management?
b. Housecleaning service
c. Appliance repair
. How do the managerial tasks differ among the services
described in question 2?
Describe the service-product bundle for each of the
following services:
a. Hospital
b. Lawyer
c. Trucking firm
. Critique the customer contact model. What are its
strengths and weaknesses?
Identify the front office and back office services for
the following organizations. Could these services be
improved by increasing or decreasing the degree of
customer contact? By separating low- and high-contact
services?
a. Hospital
10. Find some service guarantees in everyday life and bring
them to class for discussion.
11. What attributes are required of a service guarantee to
make it effective?
1
What are the pros and cons of having a service
guarantee?
ile} How can we use the service delivery system matrix to
improve service operations?
14. What does it mean for a service firm to outsource some
of its services?
ilaye What key factors are most firms seeking when they
offshore services?
16. In what situations will customers accept service by
Al agents?
Whe What should be the nature of interaction between
human service providers and AI agents?
Clade
PAu ly iE IR
Process-Flow
Analysis
Calculate process-flow cap abilities
Explain the principles of process redesign.
A customer walks into a home improvement store, selects a paint color from the multitude
of sample options, and gives the selection to the store employee. The employee enters
information about the selection into a machine that automatically dispenses the appropriate
set and quantities of pigments into a can of white paint. Another machine is used to shake
the can, resulting in consistent color throughout the can. The customer walks away with
a virtually customized product in a matter of minutes. The simple process used to create
the customized paint combines the customer’s preference, employee skills, and automated
technology.
Processes are encountered in all parts of a business, not only in operations and supply
chain management. Accountants use many processes, including collecting transactions,
posting to the ledger, trial balance, adjusting entries, financial statements, and closing
entries. Marketing managers also use numerous processes, including strategic planning,
marketing research, advertising, selling, and customer relations. All of these processes can
be improved by the ideas of process-flow analysis presented here.
ny
Ou
SS
ee -
i
a
aoy
H] | 98 | Part Two
Process Design
This chapter is about understanding processes and how they are used to produce and
deliver products and services. It is also about determining what a process is capable of producing. Process-flow analysis requires viewing and analyzing the transformation process as a
sequence of steps connecting inputs to outputs. It is used to discover better methods or procedures for producing and delivering a product or a service deemed to be of value to customers.
Measuring process flows is essential to process-flow analysis and to improving transformation processes. We describe several process measures, including processing time,
throughput time, flow rate, inventory, and capacity. We also define bottlenecks and provide
analyltic methods for calculating these measures.
The flowchart (often referred to as a process map) is an essential tool to facilitate
process-flow analysis. Flowcharts should consider not only process flows but customers,
suppliers, and employee inputs in designing better processes. A flowchart for a high-contact
service process, such as consulting, often reflects the customer’s perspective, mapping the
activities performed for the customer. In manufacturing, a flowchart often shows the activ-
ities performed on materials as they move through the production system.
To truly understand process-flow analysis, we begin this chapter with process thinking.
This is a very powerful idea in business education and in practice.
6.1
LO6.1
PROCESS THINKING
Describe
process thinking and
system boundaries.
Process thinking is the point of view that all work can be seen as a process. It begins by
describing the process of interest as a system. A system is defined by its boundaries, inputs,
outputs, suppliers, customers, and system flows. System definition is needed before
detailed measurement and process flowcharting can begin.
A system is a collection of interrelated elements whose whole is greater than the sum of
its parts. The human body, for example, is a system. The heart, lungs, brain, and muscles can-
not function without one another. They are interrelated, and the function of one part affects
the others. The whole of the body is greater than any of its individual parts or components.
A business organization can also be viewed as a system. Its parts are the functions of marketing, operations, finance, accounting, human resources, and information systems. Each of
these functions accomplishes nothing by itself. A business cannot sell what it cannot produce,
and it does no good to produce a product or service that cannot be sold. The functions in an
organization are highly interactive and have value as a system that they do not have separately.
Within operations, the transformation or conversion system is made up of workers,
equipment, customers (for services), and the activities that carry out the transformation.
The transformation system can be analyzed by first specifying the system boundaries.
The boundaries delineate the resources and activities in the system being analyzed from
those that are outside of the analysis and decision area. Identification of the system boundaries is always difficult and somewhat arbitrary, but it must be done to separate the system
being analyzed from the larger system or organization in which it operates. In this sense,
the boundaries of a firm separate the firm from the larger supply chain in which it resides.
To illustrate these concepts, consider the case of a bank that is installing a new information system. The new system will replace the current one, with larger capacity, new hardware, and new software. Training will be required to operate the new system, and so human
resources can be considered part of the system. Operations will be affected by the new software and must be included within the system boundaries. Each part of the organization that
is affected by the new information system should be included within the system boundaries,
and functions that are not affected can be excluded as being outside the system boundaries.
In this way, the appropriate system boundaries can be identified for purposes of analysis.
Chapter 6 Process-Flow Analysis
| 99] |]
A cross-functional team should be formed, consisting of the functions that are affected by
the conversion to the new information system. This team will be responsible for overseeing
the conversion from each of their functional perspectives and should handle the interactions
between functions. If this is done by a cross-functional team rather than workers from a single
function, a systems view of the project will be taken. This sort of process thinking considers
all the interacting functions within the system boundaries when making the conversion.
6.2
THE PROCESS VIEW OF BUSINESS
LO6.2
Explain
One of the most important contributions of process thinking is that a business can be
how the process
view of business is
cross-functional.
viewed as a system that consists of a collection of interconnected processes. The process
view of a business is horizontal in nature; the functional view is vertical. This is shown
graphically in Figure 6.1.
As an illustration of interconnected processes in a business, consider a scenario. A sales
team has a process for creating the customer order, while at the same time interacting with
operations to ensure adequate capacity is available to fill the order. Other marketing personnel use a process for pricing the customer order. Once operations receives the order, the
necessary processes are used to produce enough output to fill the order. The shipping area
has a process for securing the order for delivery, and transportation is scheduled to deliver
the order to the customer. Finance uses its own processes to bill and receive payment from
the.customer, while relying on pricing information from marketing and order size and
delivery confirmation from operations.
Viewing a business as a collection of processes emphasizes the cross-functional nature
of decision making. It illustrates that functions must make handoffs to one another in executing a process. As a result, time and information can be lost between processes. In some
cases, the number of steps in a process is so large that the system cannot function in an
efficient and effective manner. See the example in the Operations Leader box of the complex and time-consuming process of setting up clinical trials to test newly developed drugs.
Another example comes from an MRI (magnetic resonance imaging) facility, where
a backlog of patients had developed. The backlog was difficult to solve because the MRI
facility did not control its own schedule. An outside service provider was contracted to
schedule patients, and their scheduling caused several problems. The scan indicated on the
schedule did not always match with the actual scan that the patient needed, and sometimes
not enough scan time was scheduled for certain patients. Further, there was a need to hire
FIGURE 6.1
The process view of
business.
Source: V. Grover and M.K.
Malhorta, “Business Process
Reengineering: A Tutorial on
Concept, Evolution, Method,
Technology and Application,”
Horizontal organization
Journal of Operations Manage-
ment 15 (1997), p. 200.
Customer
request
Order
fulfillment
: ||100| Part Two
Process Design
Process Analysis to Improve Clinical Drug Trials
The Office of Clinical Research at the University of South
Florida (USF) wanted to improve the complex process of
sa
ear
starting a clinical trial. A clinical trial is a research study
ar’
ay
used
to conduct
a controlled
test of
a new
drug
on
humans. Patients volunteer to participate in such studies,
and the results are used to determine whether the new
drug will be approved for use in the marketplace.
USF assembled a team of experts to map and improve
the process.
The administrative process comprised five subprocesses, 30 activities, 11 decision points, five loops, and
eight participants. The mean activation time was 76.6
days. The typical clinical trial goes through these five
subprocesses: Initial Preparation, Contract Negotiation,
Budget Negotiation, Preparation for Submission, and Trial
Activation.
They used flowcharting to create an overview of the
administrative process starting from initial preparation
and ending with trial activation. Additionally, activity time
stamps were used to compute the duration of each sub-
process and of the overall process. From the flowchart
and reduced the time taken by many
they identified
activities in the administrative process.
Based on these studies, time-limiting subprocesses
were those of contract and budget negotiation. Reducing
Ek !sakson/Blend Images LLC
the time taken by these processes
reduced the activa-
tion time by 28 percent.
Source: D.A. Martinez, A. Tsalatsanis, A. Yalcin, D.J. Zayas-
Castro, and B. Djulbegovic, “Activating Clinical Trials:
A Process Improvement Approach,” https://doi.org/10.1186/
$13063-016-1227-2, open source, 2016.
an additional technician to conduct scans, and such technicians were in relatively short
supply in the local labor market. Dealing with the backlog of patients waiting for scans is
an operations problem, but processes outside of operations interact with processes within
operations to cause the problem. Solutions to the backlog problem need to account for the
larger system of activities within and outside of operations.
This example illustrates how operations is a part of a larger organization that includes
many other functions and outside organizations. Nearly all operations decisions are related
to at least one other function in the organization. The process view of business provides a
vehicle for understanding the interactions between various organizational functions, and
sometimes extending beyond organizational boundaries. These interactions can be streamlined and improved by process flowcharting, as is described next.
6.3
PROCESS FLOWCHARTING
LO6.3 Construct a
process flowchart for
a given process.
Process flowcharting is a tool for beginning to understand and improve processes within a
larger system. This is a very commonly used tool in a wide variety of industries. It can be
useful for almost any type of process, to gain understanding of the activities that must
occur for the process to successfully produce a product or service.
Process flowcharting refers to the creation of a visual diagram to describe a transformation process. Flowcharting is known by several names: process mapping,
flow-process charting, and in a service operations context as service blueprinting. Value
Chapter 6 Process-Flow Analysis
1104} |)
stream mapping is yet another approach to process flowcharting popularized by firms that
implement lean systems and lean thinking. Creating a visual diagram can be invaluable in
documenting what happens within a transformation process. This pictorial documentation,
when it includes process measurements, such as time or cost, can help to identify how the
transformation process can be improved by changing some or all of the following elements:
Raw materials
Product or service design
Job design
Processing steps or activities used
Management control information
Equipment or tools
ae
Ue
SOSuppliers
While there are many different specific forms of the flowchart in use, the most common
is the systems flowchart. An example of a systems flowchart for the “selecting a supplier”
process is shown in Figure 6.2. In this example, the systems flowchart is drawn from the
perspective of the buyer within an organization and shows the discrete steps, along with
decision points and flow sequences, in selecting a supplier.
Another example appears in Figure 6.3, which shows
the service provided to help a customer select and have
altered a suit from a retail store. This systems flowchart
depicts a service context with the customer being in the
system and interacting with the service provider and
is, as such, also called a service blueprint. Moreover,
because the service blueprint captures the perspectives
of different people—customer, sales associate, and
tailor—it is also known generally as a swim lane
|
The tailor makes alterations on the suit as part of the service
blueprint.
lh
ere
FIGURE 6.2
A flowchart for
selecting a supplier.
flowchart. A swim lane flowchart is used to show the
responsibilities of groups or individuals in either horizontal or vertical columns. It shows who or what is performing each step in the flowchart in the form of “swim
lanes” ina pool. In Figure 6.3, horizontal swim lanes are
drawn to demarcate
:
:
the various participants involved in
nee
;
the process of buying a suit from a retail store.
|
102) Part Two.
Process Design
FIGURE
Service to select and have altered a suit from a retail store.
6.3
Customer
Sales
Associate
Greet
customer
Request
preferences (size,
i
style, and price)
Salespe eos
takes customer
to racks
Do we
have a
adit oe
f
interest?
xe
No
Exit
Tailor
Line of Visibility
Tailor
Shop
For any type of flowchart a number of principles should be followed to create a flowchart that is easy for individuals unfamiliar with the transformation process to understand
and that facilitates process-flow analysis. These principles are consistent with process
thinking, which views the transformation process as a system with inputs, outputs, customers, suppliers, boundaries, and processing steps and flows. The principles are as follows:
1. Identify and select a relevant transformation process (or system) to study. This
can be the entire supply chain for a product or a service, the entire firm, or a part of the
firm, for example, the shipping department. Ideally, the selected transformation process is
thought to affect performance.
2. Identify an individual or a team of individuals to be responsible for developing the flowchart and for subsequent analyses. This individual or team should have
some familiarity with the transformation process and should have process ownership, that
is, authority for initiating and/or implementing changes to the process. When a selected
transformation process cuts across different functions, a cross-functional team should be
involved. When a selected transformation process cuts across the supply chain, interfirm
collaboration becomes critical.
3. Specify the boundaries of the transformation process. The boundaries denote
where the selected transformation process begins and ends, identify the customer(s) and
the supplier(s) of the transformation process, and determine how many processing steps or
activities are to be evaluated. In some cases, a function or department within an organization is the customer or supplier; in other cases, another firm is the customer or supplier.
Chapter 6
Process-Flow Analysis
103 ||
Move customer
to the tailor
Line of Visibility
4. Identify and sequence the operational steps or the activities necessary to complete the output for the customer(s). It is important in process flowcharting to depict
what is actually happening and not what one thinks is happening. Once the “as it is” flowchart has been created and the transformation process has been analyzed, creating a “to be”
flowchart may help show what the transformation process should look like when improvement changes have been implemented.
5. Identify the performance metrics for the operational steps or the activities
within the selected transformation process. These metrics should be tied to the performance of the overall transformation process. For example, if delivery performance is of
interest, it may be useful to track the processing times for each operational step or activity.
Alternatively, if quality performance is of interest, it may be useful to track the defect rate
for each operational step or activity.
6. Draw the flowchart, defining and using symbols in a consistent manner.
Figure 6.4 shows the common symbols in Microsoft Visio for creating a systems flowchart. These symbols were used in Figures 6.2 and 6.3 and are also consistent with ISO
9000 standards for flowcharting.
When other specific forms of flowcharts are created, the individual or team responsible
may choose to use other symbols. This is allowed as long as the symbols are used consistently and, more importantly, a symbol key is provided to help interpret the flowchart that
HN} |104| Part Two.
Process Design
FIGURE
6.4
Common flowcharting
symbols.
Symbol
Meaning
This symbol shows the “start” and the “end” of the flowchart, thereby
specifying the boundaries of the transformation process to study.
The words “START” and “END” should be written inside the
symbol for clarity.
Terminator
This symbol denotes an operational step or an activity to be
performed. A short description of the operational step or the
activity should be written inside the symbol for clarity.
Process
Decision/Evaluation
a
Flow
This symbol represents a decision, an evaluation, or an “IF-THEN”
condition that has multiple potential outcomes (i.e., branches of
arrows). The decision, evaluation, or condition should be properly
described in writing inside the symbol for clarity. Each branch of
arrow should be properly labeled to denote the meaning of the
outcome from the decision, evaluation, or condition.
This symbol denotes the direction of flow within the flowchart; the
flow could be that of materials, information, or person (e.g., customer).
is drawn. Figure 6.5, for example, is a flow-process chart of the picking operations at a
distribution center that provides produce, dairy, and meat items to grocery stores. In this
case, the interest is in tracking the flow of “materials” inside the distribution center. The
flow-process chart depicts every step in the process from initial input of the customer order
to picking groceries in various aisles, to consolidation, inspection, and shipment of the
order. An information processing flowchart can also be drawn to depict the same picking
operation, but here the interest is in tracking the flow of “information” for the purpose of
management and control of the distribution center.
6.4
LO6.4
PROCESS-FLOW ANALYSIS AS ASKING QUESTIONS
Analyze a
process by asking
a wide variety of
questions informed
by the process
flowchart.
Creating a flowchart of a transformation process is an important first step in process-flow
analysis. Once created, the flowchart can be analyzed to yield insights into how the transformation process can be improved, given a specific improvement goal. The improvement
goal, for example, can be to increase efficiency, reduce throughput time, improve quality,
or even boost worker morale.
A systematic approach should be followed to analyze the created flowchart and the
underlying transformation process. This approach is epitomized by asking questions about
the flowchart and, by extension, the underlying transformation process. Table 6.1 shows
typical questions about the performance of a system regarding flow, time, quality, quantity,
and cost.
When these questions are asked, opportunities to improve the underlying transformation process can be identified. For example, looking at the flow-process chart in Figure 6.5
and asking questions about the picking operations at the grocery distribution center led to
the realization that many activities (transportation, inspection, delays, and storage) do not
add value to the service provided and should be reduced or eliminated. Groceries in fact
spent a considerable amount of time waiting for the next operation or in transit and very
little time in value-added operations (only 57 minutes out of 526 minutes). As a result
Chapter 6
FIGURE 6.5
Flow-process chart for
the picking operations.
Subject Charted
Produce, Dairy, Meat
Operation
Picking
aie
Process-Flow Analysis
105 ||
PEON.
PROCESS
CHART
RGS
Present
Proposed
O
Descriptions
= ise}
Separated according to work areas
:
Q
wn
Taken to start points
w i=}
Wait for order picker
Picker separates them order by order
(Produce) picker fills order
Ne}
To Dairy aisle
i)o
On conveyor waiting for picker
(Dairy) picker fills order
To Meat aisle
iS
On conveyor waiting for picker
(Meat) picker fills order
=
To inspection
Inspected
Loaded onto carts route-by-route
Waits to be taken to the warehouse
i)
Np
uUpe
=nm
‘=
Ww
uUNTutro
YM
Of,
FF.
oyoy
oO,
u
Sree
oe
de
oe
aee
a)
ofr
Oo}
Ww
}
-i
OnVvOO
EA
SN
SON
RS
RS
IE
N
SYMBOL KEY
Operation (a task or work activity)
Inspection (an inspection of the product for quantity or quality)
Transportation (a movement of material from one point to another)
Storage (an inventory or storage of materials awaiting the next operation)
Delay (a delay in the sequence of operations)
of asking questions, a number of changes were implemented, including relocating aisles
(i.e., a process layout change), revising picking methods to reduce bottlenecks and labor
time (i.e., changes to work methods and jobs), and designing special carts to make the
loading of delivery vans easier and faster (i.e., an equipment change).
TABLE 6.1
Process-flow
questions about
performance
Question Category
Examples
1. Flow
+
Is the transformation process balanced or unbalanced?
+
+
2. Time
wW .
Quantity
+
Where is the bottleneck in the transformation process?
Are all operational steps or activities necessary?
How jumbled is the flow within the transformation process?
*
*
How long does it take to produce/deliver one unit of output?
Can the length of this time be reduced?
*
What is the time between successive units of output?
*
*
Where is there excessive setup time?
Where is there excessive waiting time?
*
*
+
L.
Quality
5. Cost
How many units theoretically can be produced/delivered ina
given period (e.g., a week)?
How easy is it to change this quantity?
How many units are actually produced/delivered in a specified
period (e.g., a week)?
+
+
+
What is the historical defect rate?
Which operational step or activity contributes to the defect rate?
Where do errors occur?
*
+
How much does it cost to produce/deliver one unit of output?
What are the cost buckets that make up the cost to produce/
deliver one unit of output?
*
Cansome cost buckets be reduced/eliminated?
Similarly, looking at the service blueprint for the service of helping a customer purchase a suit from a retail store and asking questions might lead to suggestions for improvement such as the following:
*
If sales associates can be trained to listen better to customer requests, will customers be
more likely to find a suit of interest?
¢
Can customers call ahead and ask to have some suggested suits waiting for examination,
reducing their search time?
*
Is the tailor available while the customer is trying on the suit to provide suggestions on
how the suit can be tailored to fit better?
*
Does the layout of the retail store make it easy for customers to search and find what
they want?
In summary, process-flow analysis begins with a good flowchart of the transformation
process used to convert inputs into outputs. This can be facilitated by creating a flowchart that shows materials flows, information flows, or service flows. Once a flowchart is
created, appropriate questions should be asked to highlight improvement opportunities in
flow, time, quantity, quality, and cost.
6.5
LO6.5
PROCESS ANALYTICS
Calculate
process-flow
capabilities using
analytics.
Once a process flowchart aimed at improving a transformation process has been created, some
basic measures of a transformation process can be described. Process analytics uses these
measures to yield insights into the structure and performance of a transformation process.
Let’s study the airport security process during check-in at a major airport. There is a
line of passengers waiting to clear security and a number of security scanners for examining passengers and their carry-on luggage. We can measure the total time it takes from
Chapter 6
Process-Flow Analysis
107 ||
entering the security line until passengers are cleared
to catch their flights. It turns out that the following
three observations are related: the average number of
passengers in the line, the average rate at which security can process passengers, and the average time it
takes passengers to get through the line. This relationship is called Little’s Law, named after the operations
researcher who discovered it.
Little’s Law shows that the average number of items
in a system (/) is the product of the average arrival rate
to the system (R) and the average time an item stays
in the system (7). This average time in the system is
throughput time, the time from when the processing
Average waiting time in line at airport security follows
begins until the product or service is completely
fin-
er
cee
Philippe Merle/AFP/Getty Images
ished. It includes both active processing time and any
waiting time that occurs during processing. In mathematical terms Little’s Law is stated as follows:
I=TXR
where
/= average number of things in the system (or “‘inventory”’)
T = average throughput time (processing time + waiting time)
R = average flow rate in the system
In the case of airport security, if the security screeners can process an average of five passengers per minute (R = 5) and it takes an average of 20 minutes to get through the security
line (J = 20), the average number of passengers in line (I) will be 100 (R x T = 100). An
assumption is that the process is in a steady state in which the average output rate equals
the average input rate to the process.
Little’s Law is very powerful and is widely used in practice. It applies to manufacturing and
service transformation processes. Little’s Law can be used in a variety of settings and situations.
Example
Suppose a factory can produce an average of 100 units of product per day. The throughput
time, including all processing and waiting time for the product, is an average of 10 days.
T = 10 days
R= 100 units per day
Then the average inventory (partly finished product) in the factory will be
/= 10x 100 = 1000 units
For another example, the amount of money in accounts receivable can be considered as
inventory, or the stock of money. Using Little’s Law, if there is $2 million in accounts receivable (/) and $20,000 per day is added to and subtracted from (flows through) accounts
receivable (R), the throughput time is
100 days (7 = //R = 2,000,000/20,000)
Therefore, accounts receivable has 100 days of outstanding receivables.
Little’s Law applies to any steady-state transformation process including manufacturing, people waiting in lines, invoice processing, transactions in a legal office, and even
accounts receivable processing. Little’s Law is useful when any two of the three variables
Fy]| 108) Part Two
Process Design
in the formula are known, then the third can be calculated. The examples above show how
this is done to calculate / and T. We can also calculate R if we know / and T (R = 1/7).
Next, we extend process analytics to include capacity, supply, and demand. Capacity is
the maximum rate of output from a transformation process or the maximum flow rate that
can be sustained over a period of time. In the airport security example, the average flow
rate was five passengers per minute, but the capacity of the security checkpoint may have
been greater, say, eight passengers per minute. With random arrivals (such as passengers
arriving to enter the line) it is necessary to have capacity that exceeds the average arrival
rate. If the arrival rate is greater than the capacity, the line will build up to an infinite
length due to the randomness of the arrivals. This occurs because there are periods when
the arrivals are less than the average and the full capacity cannot be used during those
times. Queuing (or waiting line) theory, which is covered in a technical chapter,’ explains
these phenomena in detail.
Most processes are composed of several activities that require certain resources. In the
airport screening example resources include the workers who check each passenger’s identification and boarding pass, operators who run the scanning equipment, and the equipment itself. In general, if there are n resources that process each transaction, then
Capacity = Minimum (capacity of resource ,,..., capacity of resource,)
Note that the capacity of the entire process cannot be greater than the capacity of the most
constraining (the smallest capacity) resource, which is called the bottleneck.
The amount of output a transformation process actually produces will depend on its
capacity as well as the supply and demand of the process. The flow rate is as follows:
Flow rate
= Minimum (supply, demand, capacity)
In the factory example above, assume that capacity is 200 units per day, demand is 75 units
per day, and supply is 100 units per day. The flow rate would be 75 units per day (the minimum of the three variables) assuming they can produce only what is demanded. If they
were able to increase demand to 150 units per day, the flow rate would be only 100 units
per day unless supply could also be increased.
6.6
ANALYZING PROCESS FLOWS AT PIZZA U.S.A.
To cement our understanding of the concepts of process analytics, let us look at a Pizza
U.S.A. example. Suppose that one of the pizza stores produces fresh pizza with seven different topping choices, including the most popular “everything dump” pizza. The store is staffed
by two employees: a pizza chef and an assistant. It has an oven that can bake up to four pizzas
at a time. The transformation process (sequence of steps) followed at the store is as follows:
Minutes
Who
Take the order
1
Assistant
Make the crust
Prepare and add ingredients
Bake the pizza
S
Chef
2
Chef
24
Cut pizza and box the order
1
Assistant
Take payment
1
Assistant
Oven
"Technical chapters are available from Instructor Resources on McGraw-Hill Connect.
Chapter 6 Process-Flow Analysis
1109} |
1. What is the capacity of this process?
Looking at the three resources, we have:
¢
The assistant takes 3 minutes per order (1 + | + 1) and thus can process 20 orders
per hour.
¢
The chef takes 5 minutes per order (3 + 2) and can process 12 orders per hour.
¢
The oven takes an average of 6 minutes per order (24 + 4, because the oven holds
4 pizzas at a time), or 10 orders per hour.
For simplicity, we assume that each order is for one
pizza and that pizzas can be added to the oven any time
during the cooking cycle. The minimum of the three
resource capacities is 10 orders per hour, and so the system can produce 10 orders per hour.
2. What is the bottleneck in this process?
The bottleneck in this case is the oven. The assistant
is busy only half the time, and the chef has | minute
of idle capacity out of every 6 minutes of average baking time. Reallocating jobs between the chef and the
assistant to balance the workload may make the chef
happy but will not increase the flow rate of the process.
If Pizza U.S.A. wants to make more pizzas, something
must be done to accelerate the flow of pizzas through
»
ail
The oven is one of the resources that determines the process
capacity.
BananaStock’ Gath lkaages
the oven. ? or another
oven
must
be added. . The
lesson
here is that the process cannot produce more than the
bottleneck can process.
3. What is the throughput time?
If we assume there is no waiting time in this system, we simply add the times of all the
steps to fill an order:
14+34+2+24+1+1=32
minutes
It takes 32 minutes to complete all the steps and make one pizza. Note that adding an oven
will increase the capacity and move the bottleneck to the chef, but it will not change the
throughput time. Changes would have to be made in the actual process of cooking, preparation, or other flow times to reduce throughput time.
4. What is the flow rate?
Assuming that demand and supply exceed capacity, the flow rate is determined by the
bottleneck capacity of 10 orders per hour. However, this is the maximum flow rate; the
actual flow rate could be much less. If either demand or supply is less than capacity, then
the smaller of the two will determine the flow rate. In the following question, we assume
demand is only 60 percent of capacity, for a flow rate of six pizzas per hour.
5. What does it cost to make a pizza if the average demand is 60 percent of capacity?
Assume the chef gets paid $15 per hour, the assistant gets paid $11 per hour, and overhead cost is 50 percent of direct labor cost. At 60 percent of capacity, the average flow
rate is six pizzas per hour. The cost per hour of operations is $15 + $11 = $26 for labor
plus 50 percent added for overhead = $39 per hour, or $39 + 6 = $6.50 per pizza. Assume
Hy | 110] Part Two.
Process Design
the cost of ingredients is $2.00 per pizza. Therefore, the total cost is $6.50 + $2.00 =
$8.50 per pizza.
6. How can the unit cost ofpizzas be reduced?
Three possibilities are:
*
*
Increase demand through pricing, advertising, or other means.
If demand increases to exceed capacity, increase the flow rate of the entire transformation process by means of automation or process improvements.
¢
Reduce the unit cost of labor, materials, or overhead.
As you can see, these three approaches are interconnected because increasing demand will
also require an increase in capacity at some point, and increasing the flow rate does no
good unless demand is increased to sell the additional product.
6.7
LO6.6
PROCESS REDESIGN
Explain the
principles of process
redesign.
Process redesign usually starts with identifying critical processes required to meet the
customers’ needs. Then the critical processes, many of which cut across organization
boundaries, are analyzed in detail using the methods described in this chapter. Changes are
often made to these processes as a result of the insight from process-flow analysis. These
changes might include eliminating some steps and combining others, or could be as
extreme as a complete reconfiguration of process steps. As a result, business processes are
redesigned and integrated to better serve the customer. The term business process reengineering (BPR) has also been used to label extensive process redesign activities. See the
Operations Leader box titled “Crédit Suisse: A Successful Process Redesign” for an
example of how BPR is being deployed to aid in improving processes.
In their famous book Reengineering the Corporation, Michael Hammer and James
Champy argue that most business processes are antiquated and need to be completely
redesigned. Many existing processes have been designed within the confines of individual
functions, such as marketing, operations, and finance and also do not make use of complete
information systems. As a result, these processes take far too long to provide customer
service and are inefficient and wasteful.
Consider a major insurance company that had just this problem. When the customer
called about an insurance problem, the call was taken by the call center. The problem was
entered into a computer and passed electronically to one of several departments: underwriting, policy service, accounting, or another department. The problem then waited in
the queue, often for several days, until a worker had time to investigate it. In some cases,
the customer’s problem had been routed to the wrong department and had to be routed to
another department, again spending several days in the queue. If the problem required more
than one department to answer the question, the process of waiting was repeated. Finally,
someone in customer service would get back to the customer after several weeks. In many
cases, the original question was not answered completely or was answered incorrectly.
This process was redesigned by completely reorganizing the entire insurance operation
around customer service representatives who would attempt to handle customer requests
on the phone, if possible, using detailed protocols and standard scripts. If additional work
was required, the customer service representative checked with other specialists and got
back to the customer with an answer. The customer service representatives had been crosstrained in all the various disciplines required and were supported by the other departments. Although this required more training of customer service representatives, it greatly
improved the speed and accuracy of the service while saving many millions of dollars. It
also provided a single point of contact and less hassle for the customer.
Chapter 6
Process-Flow Analysis
414] ie
Crédit Suisse: A Successful Process Redesign
Headquartered
in Zurich, Switzerland, Crédit Suisse is a
global financial
services
institution
with more
than two
and corporate banking customers. In the past, the steps
of closing a customer
account
involved
many
manual
million customers in 50 countries and a global workforce
of more than 50,000 employees. Process redesign has
steps and a variety of banking personnel. The original
accounts closing process, shown here, was prone to
been applied to many of its major service processes. The
“accounts closing” process is described here.
errors, slow to respond, and not very efficient.
Customers’
rising
expectations for faster
service,
¥___
with
increasing regulations,
a need for a
=
CrebirSursse™
along
created
: faster and less errorprone process. Daily,
» hundreds of accounts
are
DSGNSR1/Shutterstock
closed
by
retail
After the process was analyzed, an important insight
came to light. Most requests for account closing could be
handled in a standardized manner and therefore could
be automated. A new software application was developed to allow the relationship manager to initiate all the
activities that would close a customer account. Other
bank
personnel
a small number
did not need
of unusual
to be involved, except in
cases. As a result, the time
to close an account was reduced by 50 percent and the
error rate was reduced to just 0.01 percent.
send inquiries
send inquiries
Original process
speak to
Relationship
Manager
or write
fill form or
send e-mail
Administrator
;
Accounting
send
instructions
Clerk
Redesigned process
Relationship
speak to
Manager
enter data
or write
Source: Peter Kling and Claus Hagen, “The Fruits of Business Process Management: An Experience Report from a Swiss
Bank,” Business Process Management Journal 13, no. 4 (2007), pp. 477-87; www.credit-suisse.com, 2019.
& P|112] Part Two
Process Design
Process redesign is radical redesign when processes simply cannot be improved in
small steps and require a complete rethinking and rearrangement of process activities to
improve them in a major way, as was the case for the insurance company described above.
Often radical redesign is supported by new technology, in the form of either production
technology or information technology.
To pursue a successful radical redesign requires four principles:
1. Organize around outcomes, not tasks. The insurance company was originally
organized according to tasks, using the classic division of labor. When the company reorganized around the outcome, which was customer service, dramatic improvements were
made. A customer service representative handled all activities associated with the desired
outcome. Although it is not always possible to have one person do everything, jobs can be
broadened and handoffs between departments can be minimized.
2. Have the people who do the work process their own information. When bedside
or portable information system access is available, nurses can update patient electronic
medical records as they are dispensing medications to the patient. By doing so, nurses
avoid delaying the record update and also do not “hand off” the information for input by
someone else, thus reducing the likelihood of inadvertent errors. This principle can be
applied in many situations in which information is passed from one department to another.
3. Put the decision point where the work is performed, and build control into the
process. It is better to push decision making to the lowest possible level. This will eliminate layers of bureaucracy and speed up the decision-making process. In the insurance
example, the customer service representative had greater latitude to make decisions directly for the customer rather than referring decisions to other departments. To accomplish
this, however, information and controls must be built into the process itself.
4. Eliminate unnecessary steps in the process. Simplifying the processes frequently
means that unnecessary steps and paperwork are eliminated. Every step is examined by using
the flowcharting techniques discussed earlier, and only those that add value for the customer
should be retained. Process redesign can be used to streamline and implify work flows.
Process redesign is just one of many methods that can be used to improve operations.
It uses a process view of the organization as a way of improving process flows. As a result
of process redesign, processes will be simplified, process flows improved, and non-valueadded work eliminated.
6.8
KEY POINTS AND TERMS
This chapter has emphasized process-flow analysis by building on the ideas of systems,
measurement, flowcharting, analytics and process redesign. The key points are:
*
*
*
*
A prerequisite to process-flow analysis is definition of the system to be analyzed. Systems definition requires isolation of the system of interest from its environment by
defining a boundary, customers, outputs, inputs, suppliers, and process flows.
The process view is the idea that a business is a set of horizontal processes that are
interconnected with the objective of meeting customer needs.
Process flowcharting creates a pictorial description of a transformation process. The aim
is to create flowcharts, or visual diagrams of a transformation process, that are easy to
understand by people who may not be familiar with the underlying transformation process.
Process flowcharting can be applied to materials flow, information flow, and customer
flow. In manufacturing, a flow-process chart is created to show materials flow. In services,
a service blueprint is created to show how customers interact with service providers.
Chapter 6 Process-Flow Analysis 413 ||
¢
Process-flow analysis takes the flowchart and the measurements of a transformation
process and seeks answers to relevant questions. These questions help highlight opportunities that can be implemented to improve the transformation process.
¢
Process analytics is essential to process improvement. Little’s Law relates inventory to
throughput time and flow rate of a stable system. The bottleneck resource determines
the capacity of the entire process.
¢
Process redesign is used for changing how a process is carried out. It is often crossfunctional in nature and may require a complete overhaul of work methods, flows, and
information systems.
Key Terms
LEARNING
ENRICHMENT
(for self-study or
instructor assignments)
Process-flow analysis 98
Process thinking 98
System 98
System boundaries 98
Process view of a business 99
Process flowcharting 100
Process mapping 100
Flow-process charting 100
Service blueprinting 100
Systems flowchart 101
Process ownership 102
Process analytics 106
Little’s Law 107
Throughput time 107
Capacity 108
Bottleneck 108
Flow rate 108
Process redesign 110
Business process
reengineering 110
Radical redesign 112
How to Draw a Simple Process Map
Video
https://youtu.be/DXuPeDj4TsE
6.41
_ Little’s Law Worked Problem
https://youtu.be/h- 1Q-uuuQkQ
Video
6.07
Little’s Law and Lead Time
https://youtu.be/JUszeJViSjU
Video
S07
Business Process Reengineering—Definition
https://youtu.be/Wi-BmxkA7
YU
Video
S512
Disney World Queuing and Waiting Time
www.mhhe.com/disney
Video
9:26
SOLVED PROBLEMS
Problem
1. A ticket line for a Minnesota Vikings football game has an average of 100 fans waiting
to buy tickets and an average flow rate of 5 fans per minute. What is the average time
that a ticket buyer can expect to wait in line?
Solution
Using Little’s Law J = T x R, solve for T:
T=I+R=100+=5=20
Problem
A ticket buyer can expect to spend an average of 20 minutes in line.
2. Joe’s commercial laundry has contracts to wash bedsheets for hotels. Joe intakes each
batch of sheets, which takes | minute, and then the sheets are washed, taking 20 minutes,
and dried, taking 30 minutes. The batch of sheets is ironed, taking 10 minutes for one
ie ||114] Part Two
Process Design
employee to complete each batch, and there are two employees ironing sheets. Finally,
Joe packages the sheets and bills the customer, taking 2 minutes. Joe has five washing
machines and seven dryers that can process one batch of sheets each.
a. What is the capacity of the laundry system, and what is the bottleneck?
b. What is the average throughput time of a batch of sheets?
C If the flow rate is 10 batches per hour, what is the average number of batches of
sheets in the system (inventory)?
Solution
. The capacity of each resource is as follows:
¢
Joe takes 3 minutes for each batch and can thus handle 20 batches per hour.
¢
Ironing takes 10 minutes, and so each employee can handle 6 batches per hour
and the total capacity for two employees is 12 batches per hour.
¢
Washing machines take 20 minutes per batch or three loads per hour for each
machine, and there are five machines, for a total capacity of 15 batches per hour.
¢
Dryers take 30 minutes per batch or two loads per hour from each machine x
seven machines for a capacity of 14 batches per hour.
The most constraining (minimum capacity) resource is the ironing, and so the system
capacity is 12 batches per hour and ironing is the bottleneck.
. The average throughput time (assuming no waiting time) of the system for each
batch of sheets is:
T=1+4+20+
30+ 10+ 2 = 63 minutes
.
Problem
[=T xX R= (63 = 60) X 10 = 10.5 batches (note that the 63 minutes must be converted to hours using 60 minutes in an hour).
. A restaurant has 30 tables. When
the guests arrive, the manager
seats them, servers
serve them, and the cashier assists them when they pay the bill. The process is shown
with processing times above the process steps and waiting times that occur between
O perations below the steps. One manager, one cashier, and four servers are available.
Oper. time
Manager
1 minute
Server
2 minutes
Find table
Bring menu
&
order drinks
Time between
Server
3 minutes
})
Serve drinks
&
'
Server
4 minutes
Server
1 minute
Cashier
2 minutes
Serve food
Bring check
Pay cashier
|y
order food
10
20
30
5
Oper. (minutes)
. What is the capacity of the system and the bottleneck resource?
. What is the throughput time for each customer?
. If there are 20 arrivals per hour, what is the average number of tables filled?
ist)
lop
Solution
. The capacity of each resource is as follows:
¢ The manager takes | minute each and can handle 60 customers (or tables) per hour.
¢ The cashier takes 2 minutes each and can handle 30 customers per hour.
¢ Each server takes 10 minutes per table and can handle 6 tables per hour. There are
four servers, and so the total capacity for servers is 24 tables per hour.
¢ There are 30 tables available.
Chapter 6
Process-Flow Analysis
115 ||E
The resource with the minimum capacity is the servers, and so the system capacity is
24 tables per hour and the bottleneck is the servers.
b. The throughput time (including both processing time and waiting time) of the system
for each customer is
1424+34+44+14+2+4+54104
20+ 30+ 5 = 83 minutes
c. If there are 20 arrivals per hour, there will be an average of
I=TXR=
(83 + 60) X 20 = 27.7 tables being used
Discussion Questions
1 In the following operations, isolate a system for analysis
and define customers, services produced, suppliers, and
the primary process flows.
. Explain the differences between capacity, flow rate, and
demand.
. What kinds of problems are presented by the redesign
of existing processes that are not encountered in the
design of a new process?
a. A college
b. A fast-food restaurant
c. A library
. Explain how the process view of an organization is
likely to uncover the need for greater cross-functional
cooperation.
. Explain Little’s Law in your own words. How can it be
used, and what are its limitations?
. Provide a definition of a bottleneck. Why is it important
to find the bottleneck?
. Why is it important to define the system of interest
before embarking on improvement? Give three
reasons.
. Find examples of flowcharts and study their details.
Describe how the flowcharts might be used to make
improvements.
. Describe service blueprinting in your own words,
including examples of when it should be used.
Problems
ile In a company that processes insurance claims, the average flow rate is 10 claims per hour and the average
throughput time is 6 hours.
a. How many claims are in the system on average?
b. If the demand for claims to be processed is seven per
hour and the capacity is eight per hour, what is the
flow rate?
c. What assumptions have you made in your answers?
. Suppose a bank clears checks drawn on customers’
checking accounts by using the following process:
Receive
checks
Sort checks
by bank
Ship checks to
the check
clearinghouse
a. If the capacity for receiving checks is 1000 checks
per hour, for sorting checks is 800 checks per hour,
and for shipping checks is 1200 per hour, what is the
capacity of the system to process checks?
b. If the flow rate is an average of 600 checks per hour
and there are an average of 200 checks in the system, what is the average throughput time of checks?
c. What could be done to decrease the throughput
time?
. The Stylish Hair Salon has three stylists who provide
services to women. After checking in with the receptionist, which takes an average of | minute, the custom-
er’s hair is washed, dried, and styled, taking an average
of 25 minutes. The payment takes 3 minutes and is also
performed by the receptionist.
a. What is the capacity of the process, and what is the
bottleneck?
b. What is the average throughput time? If the average flow rate is five customers per hour, what is the
average number of customers in the system?
c. If the input to the system is random, what will happen as the flow rate approaches the capacity of the
system?
. Judy’s Cake Shop makes fresh cakes to customer
orders. After receiving the order by Judy’s assistant,
which takes 2 minutes, Judy then takes 8 minutes to
mix the ingredients for the cake and loads a cake
pan for baking. Then the cake is put into the oven for
30 minutes. The oven can hold three cakes at one time.
N) | 116 Part Two
Process Design
When the cake is taken out of the oven, it is cooled for
1 hour. The assistant then takes 2 minutes to pack the
cake for pickup and bills the customer, which takes
3 minutes.
a. What is the capacity of the process, and what is the
bottleneck?
b. What is the throughput time for a typical cake?
If on average five orders are taken per hour, how
What is the average cost of an order when operating at 10 orders per hour?
« What is the minimum cost per order that the system can achieve?
e. What assumptions have you made in these calculations that may not be reasonable?
6. A furniture factory makes two types of wooden tables,
«
large and small. See the flowchart below.
many cakes are there in the process (on average)?
5. The Swanky Hotel provides room service for its guests.
The process for room service begins with a room service manager who takes orders by phone at an average
of 2 minutes per order. The manager then sends the
2
:
:
ae
order to the kitchen, where it takes a cook an averag
of 16 minutes to prepare the food for each order. There
Setup
Run Time
Capacity,
Time,
per Piece,
Pieces
Minutes
Minutes
per Hour
15
10
Wood cutting
Make tourilegs
30
60
5
10
are four cooks in the kitchen. If the customer orders a
Make tops
60
12
8
beverage, the room service manager sends the order to
the bar at the same time the order is sent to the kitchen.
Binishahonvood
20
38
12
It takes 3 minutes for a bartender to fill the order, and
Assemble and ship
20
ig
14
80 percent of the orders require a beverage. When the
kitchen and bar orders are both ready, a waiter takes
them to the room and bills the guest. There are six
waiters to provide the service, and each order takes
20 minutes for the waiter to complete.
a. What is the capacity of the process, and what is the
bottleneck?
b. What is the throughput time of a typical order?
c. Assume that on Friday evenings an average of
10 room-service orders per hour are placed. How
many orders are in the system on average on Friday
nights?
d. Assume the following pay rates for the employees.
Waiters are paid $9 per hour (not including tips),
cooks are paid $15 per hour, the bartender is paid
$10 per hour, and the room service manager is paid
$18 per hour. Also, assume that 60 percent overhead
is added to direct labor and that the cost of food and
beverages averages $6 per order.
Small tables are made in batches of 100, and large
tables are made in batches of 50. A batch includes a
fixed setup time for the entire batch at each process
step and a run time for each piece in the batch. Both
large and small tables have the same processing
times. The capacities of each process step are given,
and apply to production of either type of table, as
shown in the flowchart.
a. What is the capacity of the system, and what is the
bottleneck?
b. What are the throughput times for batches of large
and small tables?
c. When producing at a rate of six small tables per
hour on average, how many tables will be in the
system?
7. Draw a flowchart of the following processes:
a. The procedure used to pay your bills
b. College registration
c. Checking out a book from the library
Cut wood
Finish
|
Assemble
furniture
|
and ship
Chapter 6 Process-Flow Analysis
8. Review your answers in problem 7 to be sure that you
are correctly using the flowchart symbols described in
this chapter.
. Use the key questions of what, who, where, when, and how
for problem 7 to suggest improvements in the processes.
. Using the correct symbols, draw a flowchart of the following processes:
a. Preparing yourself for a job interview
b. Going to the library to study and returning to your room
1417) |
ial, Use the what, who, where, when, and how questions to
make improvements in problem 10.
. Draw a service blueprint for the following:
a. Pizza delivery
b. Automobile repair
. Analyze the service blueprints in problem 12 for possible improvements. Use the flow, time, quantity, quality, and cost questions.
Ct
eAUs ome lasek
Lean Thinking and
Lean Systems
oe
LO7.2
Describe the five tenets of lean thinking and the seven forms of waste in a lean system.
LO7.4
Explain how setup time, lot size, layout, and maintenance are related to lean thinking.
LO7.8
Explain how to implement a lean system.
Lean is a way to reduce waste and non-value-added activities in organizational processes.
For example, most emergency departments in hospitals are not lean. The average patient
spends three hours in the emergency department even for a relatively minor medical visit.
Non-value-added activities for patients such as waiting for a blood test, then waiting for an
x-ray, then waiting for a nurse, and finally waiting for a doctor accumulate to more than
two hours of waiting time for a three-hour visit. Less than one hour of the visit adds value
for the patient. Lean concepts and principles can improve the flow of patients and take out
much of the non-value-added waiting times.
Lean thinking and lean systems are applied in a wide variety of industries and settings.
They are used to improve operations processes in manufacturing and services. Lean ideas
Chapter 7
Lean Thinking and Lean Systems
119] |
are also used to improve processes outside of operations, including software development
and maintenance, annual budgeting, and even for collecting on delinquent accounts! What
is lean and how do firms use lean ideas to improve their business?
In this chapter we introduce the lean concepts, principles, and techniques that organizations adopt for performance improvement. These concepts, principles, and techniques can
be deployed to reform not only manufacturing systems but also administrative systems,
service systems, and entire supply chains. We begin by looking at the evolution of lean
before presenting lean thinking as a set of five tenets.
7.1.
EVOLUTION OF LEAN
LO7.1 Describe the
origins and evolution
of lean thinking.
— After World War II the U.S. system of mass production was the envy of the world. Mass
—_production—the production of standardized discrete products in high volume—was the
norm. Materials were produced in large batches, and machines were made to run faster
to reduce unit costs. In some cases this resulted in sacrificing quality in the name of efficiency and creating narrow jobs that led to worker dissatisfaction, but still the world bought
products manufactured in the U.S.
In the 1960s a “Japanese miracle” started at the Toyota manufacturing company. After
visiting U.S. manufacturing companies, Toyota determined that it could not copy their system of mass production. Not only was demand for Toyota automobiles low at that time,
there was a severe lack of resources. Because of the lack of resources, Toyota developed
a strong aversion to waste. Scrap and rework were deemed wasteful, and so was inventory
that tied up storage space and valuable resources. Toyota realized that it needed to produce
automobiles in much smaller batches, with much lower inventory, using simple but highquality processes and involving workers as much as possible. This realization became the
foundation for what is known today as the Toyota Production System (TPS) and Just-InTime (JIT) manufacturing. This production system 1s now used around the world. See the
Operations Leader box titled “TPS at the Toyota Plant in Georgetown, Kentucky, USA.”
JIT manufacturing first came to the U.S. in 1981 at the Kawasaki motorcycle plant in
Nebraska, which used some of the TPS ideas. However, instead of transforming the entire
system, JIT manufacturing focused primarily on inventory reduction but ignored other
aspects of Toyota’s complete system. As a result, many U.S. companies that attempted to
copy the TPS ideas achieved only partial improvement.
In 1990 Womack, Jones, and Roos studied JIT auto-
mobile manufacturing in Japan, the U.S., and Europe and
popularized the term lean production in their famous
book The Machine That Changed the World: The Story
ofLean Production. Lean production was defined as systematically eliminating waste in all production processes
by providing exactly what the customer needs and no
more. They reported that the best plants using lean production had a big edge in automobile assembly performance anywhere in the world. Labor productivity in the
best plants exceeded that in the worst plants in all three
regions by a factor of 2 to 1, defects were reduced by
half, and inventory was reduced from two weeks’ worth
Lean thinking can be applied in hospital emergency rooms to
speed patient treatment.
Paul Bradbury/Getty Images
to only enough to maintain production for two hours.
The
ae best U.S.-owned plants, indeed, had labor produc-
tivity (vehicle assembly hours) and quality comparable
i1] | 120| Part Two
an
Process Design
SAH
Toyota Motor Manufacturing Kentucky, Inc. (TMMK), is
Toyota’s flagship manufacturing facility in the United
States, currently producing 550,000 vehicles annually.
Established
in 1986,
Georgetown,
TMMK
Kentucky,
occupies
is over 8.1
1,300
acres
million square
in
feet
in size, and employs 8,000 workers to build the Camry,
Avalon, and Lexus models.
At TMMK, employees are trained not only in required
job skills but also in problem-solving and continuous
improvement methods. Job tasks have been standardized to minimize waste and assure quality. Employees
can stop and are strongly encouraged to stop the production line when a quality problem is detected. Employees are, moreover, actively involved in suggesting ways
to improve their work and work environments,
with an
astounding 100,000 suggestions on average per year.
Toyota has done more than transfer the Toyota Production System (TPS) to TMMK. Realizing that the performance of TMMK depends on its suppliers, it has worked
aggressively to help its 350 U.S.-based suppliers to
implement the TPS. TMMK created the Toyota Supplier
Support Center in Erlanger, Kentucky, to provide consulting services to suppliers. Suppliers deliver to TMMK
Walter Cicchetti/123RF
frequently, allowing TMMK
average, enough
to hold inside its facility, on
inventory to last for just four hours of
production.
Source: A. Harris, “Automotive Special Report—Made in the
USA,” Manufacturing Engineer 86, no. 1 (2007), pp. 14-19
www.toyotaky.com, 2019.
\
to the best Japanese-owned plants in the U.S., while European plants lagged behind. This
showed that the best U.S. plants could adopt the TPS and compete with Japanese plants, but
average U.S. plants were still far behind, particularly behind plants located in Japan.
Today, the concepts, principles, and techniques that encompass lean production are
often referred to as lean thinking and are being deployed across a broad spectrum of
global firms, including 3M, Black & Decker, Deere & Company, Delta Airlines, Ford,
General Electric, Hewlett-Packard, IBM, Starbucks, Taco Bell, United Health Care, and
Wells Fargo. In virtually all instances, benefits such as increased inventory turnover (50 to
100 times per year), superior quality, greatly reduced customer waiting time, and substantial costs savings (15 to 20 percent) have been reported.
7.2
LO7.2
LEAN TENETS
Describe the
five tenets of lean
thinking and the
seven forms of waste
Lean thinking, as the name signals, is a way of thinking about processes at work in all
types of organizations. Lean thinking is built around five tenets that subsume specific concepts, principles, and techniques. The five tenets guide organizations in delivering value to
customers efficiently.
in alean system.
Create Value
The first tenet in lean thinking is to specify precisely what about a product or service creates value from the customer’s perspective. Value is defined by the customer and provided
in the product or service the customer needs at a place, time, and price the customer is
willing to pay. Value is not what the firm thinks but what the customer says, the “voice of
Chapter 7
TABLE
Th
a2 7
7.
Q
SANS NIUBLES
+
Source: Adapted from Taiichi
Ohno, Toyota Production
Productivity Press, 1988).
4
Producing
more
+
SAAS
than
me
the demand
Re
e e GereSr
of customers,
resulting
1124] |)
3
in unnecessary
inventory, handling, paperwork, and warehouse space.
of Waste
Syscen Beyond Waree scale
Production (New York:
Overproduction:
Lean Thinking and Lean Systems
Waiting time: Operators and machines waiting for parts or work to arrive from suppliers or other
operations; customers waiting in line.
-
Unnecessary transportation: Double or triple movement of Wiaterials due to ppor layouts, lack Of
coordination, and poor workplace organization.
*
*
Excess processing: Poor design or inadequate maintenance of processes, requiring additional
labor or machine time.
2
Too much inventory: Excess inventory due to large lot sizes, obsolete items, poor forecasts, or
improper production planning.
2
Unnecessary motion: Wasted movements of people or extra walking to get materials.
*
Defects: Use of material, labor, and UES for production of defects, sorting out bad parts, or
warranty costs with customers.
the customer.” Value is often a solution to a problem a customer is facing or improvements
in the quality of life that the customer is willing to pay for. Value, as such, is dynamic in
nature and changes over time as customer preferences change. Firms should design and
deliver product and service features that customers value and stop doing activities that are
not valued by customers (unless required for other reasons, for example, legal). This may
mean removing unvalued product features or reducing waiting times in service systems.
Waste, in lean thinking, is anything that does not contribute value to the product or
service being produced and delivered to the customer; waste adds costs that are greater
than the customer-perceived value. The Japanese term for waste is muda. In many manufacturing, administrative, and service processes only 5 to 10 percent of total throughput
time adds value for the customer. Firms want to eliminate obvious waste, but many forms
of waste are hidden. For example, the value-added time to produce a product may be only
three hours, but it takes a week to complete it. The muda or non-value-adding time might
include waiting for machines or labor to become available, dealing with backlogs, searching for materials, or correcting processing errors.
Table 7.1 defines the seven forms of waste originally identified by Tatichi Ohno, Toyota’s
former chief engineer, who is considered the father of the TPS. Womack and Jones in their
book, Lean Thinking, introduced an eighth form of waste: underutilization of workers. This
stems from not recognizing, developing, and utilizing the mental, creative, and physical
abilities of employees. Contributing to this waste are factors such as poor hiring and training
practices, high employee turnover, and an organization culture that does not respect people.
Value Stream
The second tenet in lean thinking is to identify, study, and improve the value stream of the
process for each product or service. The value stream identifies all the processing steps
and tasks undertaken to complete a product or deliver a service from beginning to end.
A typical value stream thus can include both value-added and non-value-added processing
steps and tasks. The goal in studying the value stream is to eliminate the non-value-adding
processing steps and tasks.
One technique supporting this tenet is value stream mapping, which creates a visual
representation of the value stream of a process, much like process flowcharting. Value
stream mapping requires direct observation of work and the flow of work within a process
so that opportunities for improvement can be identified. The Japanese refer to such direct
observation of work as gemba.' Using a gemba approach, the analyst walks through the
' A Japanese term meaning “the real place.”
( i 122) Part Two
FIGURE 7.1
Process Design
Health care value stream map.
Journal of Quality & Participation 25, no. 2, pp. 161-191.
Source: Adapted from Bushell, Mobley & Shelest, “Discovering Lean Thinking at Progressive Health Care,”
Doctor Visit
Information Flow
ea
Ey
*
ES
Patient Out
Patient In
40 minutes
15 minutes
Wait
20 minutes
45 minutes
MEU
5 minutes
8 minutes
16 minutes
11 minutes
Check-in
Visit Prep
Visit Doctor
Pharmacy,
Blood Test
Se
Time
|
Total
Throughput
Time
entire process from start to finish while observing and documenting each step where the
work is actually happening. The resulting value stream map shows the beginning and ending points of the process, the steps and tasks between those points, and relevant performance information about the process. Figure 7.1 is a simplified value stream map showing
how patients flow through a clinic. Improvements to the process come from studying the
value stream map and then asking and answering the question, “Is this step or task necessary in creating value for the customer?” Processing steps and tasks that are not necessary or are non-value-adding, such as the numerous waiting times in the figure, should be
reduced or removed to improve performance and ultimately enhance value in the product
or service provided to the customer. In Figure 7.1 the total throughput time is 160 minutes,
but 120 minutes of it is waiting time. Much of the waiting time (non-value-added) can, no
doubt, be eliminated by revising the process.
Ensure Flow
The third tenet in lean thinking is to ensure that flow within a process is simple, smooth,
and error-free, thereby avoiding waste. To appreciate this tenet, look at Figure 7.2, where
production is viewed as a stream and the water level as the inventory of raw materials,
work-in-progress, and finished goods. At the bottom of the stream are rocks, which represent problems related to quality, suppliers, delivery, machine breakdowns, and so forth.
The traditional approach is to hold inventory high enough to cover up the rocks (problems)
and thus keep the stream flowing. Lean thinking calls for the opposite: lowering the water
(inventory) level to expose the rocks (problems). When the rocks have been pulverized
(1.e., the problems have been solved), the water is lowered again to expose more rocks. This
sequence is iterated until all rocks are turned into pebbles and the stream (production system) flows smoothly and simply at the true market demand rate while only needing a low
Chapter 7
FIGURE 7.2
Lean Thinking and Lean Systems
123} |)
Simple, smooth, and nonwasteful flow: A stream analogy.
Prob
Problems
s
>
©
=
E
S
Original situation
(inventory covers problems)
Water level lowered
(problems are exposed)
Water flows smoothly
(once problems are solved)
level of inventory at any given time. Inventory in this analogy is a form of waste that hides
problems that contribute to other forms of waste besides inventory.
The idea of simple, smooth, and error-free flow means that production flows are sim-
ple and direct, and do not change from one production run or one customer to another.
To accomplish this, small batches will be needed to match market demand, rather than
building up inventory with large batches of production. Such predictable flows ensure
that the exact appropriate resources, including labor and equipment, can be devoted to
each production step. It also means that workers should understand the connection of
their own work to work performed upstream (before them in the production process)
as well as work that follows their own. These direct and unambiguous links in the process provide complete certainty about exactly who has performed what work within the
process. As a result, workers themselves are actively engaged in controlling the smooth
and error-free flow of production. This is done by simply limiting the inventory buildup
between operations, limiting defective products and keeping the flow of production moving. As mentioned earlier, problems are exposed—not hidden and covered by inventory.
The problems are fixed by the production workers themselves who keep the flow moving.
There are many techniques used to accomplish smooth production flow. They are covered
later in the chapter.
The goal of ensuring simple, smooth, error-free flow can be extended up the supply
chain to suppliers. They need to fill orders just-in-time to meet production schedules of the
customer. This means that suppliers may have to deliver goods to factories on a daily basis
or perhaps multiple times within a day. Ultimately, suppliers will also have to convert to
lean operations in their factories to produce products in small batches only when needed
and when pulled by the customer’s manufacturing plant.
In service operations smooth and error-free flow is achieved by reducing waiting time
for customers and providing exactly what the customer needs. Customers should flow from
one stage to the next in a service encounter. Value stream mapping is used to reduce or
eliminate waiting time. In a similar way to manufacturing, service employees should be
in charge of ensuring quality outcomes for customers and smooth flow as the service is
delivered in a timely manner.
Customer Pull
The fourth tenet in lean thinking is to produce only what is pulled by the customer. Complying with this tenet requires replacing the push system typical in traditional mass production with the pull system of lean production. A push system aims to produce goods or
Ny | 124,
Part Two
Process Design
FIGURE 7.3
Push vs. pull system
illustration.
Source: https://www.panview
.nl/lean-productie-theorie/
het-pull-principe-productieop-kantoor-het-leven
ensure delivery of services well in advance of demand, often according to a schedule or
plan created from potentially inaccurate forecasts. Large batches of materials are pushed
from one process or machine to the next regardless of whether the inventory is needed.
This allows machines and processes to be utilized at full capacity, but builds up excess
inventory. See Figure 7.3 for an illustration of push vs. pull systems.
A pull system, on the contrary, waits for the downstream process customer to signal
a need for a good or service before producing it to fulfill that need. The signal from the
customer is then sent visually upstream in the various stages of production—and even the
supply chain—to signal what and when production and delivery are needed. No upstream
process is authorized to produce a good or service until a downstream customer asks for
it, thus minimizing inventory throughout the production system. An example of a push
system is the “hub and spoke” system used by many of the major airlines. Flying pointto-point by some airlines is a pull system based on what the customer wants. No customer
wants to connect through a hub to get to their destination but the system is designed this
way for airline efficiency. However, by flying point-to-point Southwest Airlines developed
a pull system approach that revolutionized the airline industry.
Strive for
Perfection
The fifth tenet in lean thinking is to strive for perfection. Striving for perfection requires
continuous improvement of all processes as well as radical redesign when necessary. When
continuous improvement is undertaken, more value is provided by the firm in its quest for
ultimate perfection for the customer. The definition of perfection used here is an affordable good or service, delivered rapidly and on time, that meets the needs of the customer.
When customer needs change, the definition of value changes and so does the definition
of what constitutes perfection. There is, therefore, no end to the improvements that can be
sought and made.
In lean thinking, the process changes necessary in seeking perfection must be made
using scientific methods, including designing experiments and testing hypotheses. Employees are discouraged from changing a process based on intuition alone. Rather, lean thinking
promotes decision making based on scientifically derived evidence.
Quality is absolutely essential in a lean production system, because defective goods
or services are a major form of waste. Internally, wasted parts and labor add unnecessary costs, while externally, poor quality does not meet customer expectations of value.
But a lean system does not create quality. Quality must be an input into a lean system for
processes to create value for the customer, as needed to fulfill the first lean tenet and sub-
sequent tenets.
Chapter 7
Lean Thinking and Lean Systems
125) |)
Defective goods and services are clearly waste. More importantly, they can bring the
production process to a halt since workers often inspect their own work and do not pass
defects on to the next work center. In a lean system, quality problems rapidly gain attention as the production line or service process stops when problems occur. For example, in
fast-food restaurants, if the food is not at the required temperature, it cannot be delivered to
customers and may cause a backup at the drive-through window. Now customers must wait
longer for food, and the value of the food and service declines.
A lean production system is designed to expose errors and get them corrected at their
source rather than covering them up with inventory or other means. To cover a quality, or
maintenance, or worker training problem with inventory means to use excess inventory
to keep a production process moving when it should be stopped and fixed. The excess
inventory is waste and prevents quality problems from being quickly detected by the next
process. Because a lean system does not have excess inventory to cover up mistakes, striving for perfection is required. When the lean tenets and techniques are used consistently,
continuous improvement moves the process toward perfection.
One simple and relatively powerful technique used to strive for perfection is the 5 Whys
technique. This problem-solving technique systematically explores the cause-and-effect
relationships that underlie an observed problem (e.g., a product defect or late delivery).
By asking why at least five times, this technique is used to deliver insights into the root
cause of an observed problem so that proper corrective action can be taken to prevent the
root cause from re-creating the observed problem. Here is an example: A defective part
is produced. Why? The machine is out of tolerance. Why? The machine was not adjusted
properly. Why? The operator was not trained. Why? Standard procedures were not followed
by the supervisor for training a new operator. Why? The supervisor was too busy solving
other problems. Once the root cause 1s discovered the problem can be fixed, hopefully permanently for all operators and supervisors.
Another well-known technique supporting lean thinking is 5S. This is a technique for
organizing a workspace (e.g., production shop floor, office space, hospital station, tool
shop) to improve employee morale, environmental safety, and process efficiency. The
name of this technique comes from five Japanese terms, all of which when transliterated
and translated begin with the letter S. These terms are defined in Table 7.2. Underlying
the 5S technique is the belief that when a workspace is well organized, time will not be
wasted looking for “things” (e.g., a tool or paperwork); misplaced items also will be readily noticed. By having employees decide which items should be kept where as well as how
they should be stored, 5S can instill in employees a sense of ownership, help standardize
work, assure a safe work environment, and keep processes from becoming overly complex.
See the Operations Leader box titled “SS + Safety = 6S!”
TABLE 7.2
5S: Japanese
terms and English
translations
Term
Translation
Meaning
Seiri
To sort
Decide what things to keep and what things to discard so
that only the essential things remain.
Seiton
To straighten or set
in order
Arrange essential things in a manner that supports an
efficient flow of work.
Seiso
To shine, sweep,
or clean
Assure cleanliness by returning things to their storage
locations and removing things that do not belong.
Seiketsu
To standardize
Standardize work and adopt seiri-seiton-seiso throughout
so that all employees know what their responsibilities are.
Shitsuke
To sustain
Maintain seiri-seiton-seiso-seikutsu as a habit of work and
a way to operate.
LA
4
eR
OPERATIONS LEADER |
= el
!
IRA
woes
eval
Wigan
H
.
an | Psr
:a|i
|
Ee
5S + Safety = 6S!
The U.S. Environmental Protection Agency provides train-
restocked
ing materials to businesses for safe storage of hazardous
chemicals, calling the program 6S, for 5S + Safety. The
training uses standard 5S methods on removing unused
chemicals that are lingering in storage areas, decentralizing chemicals by moving them to where they are used
in work processes, and organizing and labeling storage
areas so that the chemicals are easy to find and their
absence is easily noticed. As a result, errors in using the
wrong chemical are minimized.
The photos show an example from one manufactur-
anyone should be able to find any item in 30 seconds
ing plant where,
before
chemicals
disorganized
were
5S activities were
and
conducted,
difficult
to locate.
Following 5S, the storage cabinet contains only chemicals
that
are
only a few
needed
nearby,
in quantities
that
days, and
labeled
shelves
be easily
can
as needed.
or less.”
The sixth
“A helpful rule of thumb
S for safety
is integrated
with
is that
standard
5S methods. To promote safe use and storage of chemicals, 6S training helps businesses develop appropriate visual warnings in areas near chemicals, and clear
the space around storage cabinets to avoid tripping
hazards.
Lean concepts like 5S, as seen in this example, can
be customized to the particular needs of an organization or situation. Safety is integrated here with 5S,
a logical connection when hazardous chemicals are
being used.
last
Source: Adapted, with photos, from www.epa.gov,
2020.
Before
Source: The Lean & Chemicals Toolkit/U.S. Environmental Protection Agency
73
ENSURE FLOW
As you can see, the five tenets of lean thinking have many techniques and concepts associated with them. We have already covered a fair amount of detail on all tenets except tenet 3
(ensure flow) and tenet 4 (customer pull). In the remainder of this chapter we focus on
these remaining two tenets to provide more detail and techniques on how they are actually
implemented in manufacturing and service settings.
The third tenet of lean is to ensure smooth and error-free flow by avoiding all of the
seven wastes and non-value-adding activities. Among others, this includes waste in overproduction, excess inventory, unnecessary transportation, and defects. This is a complex
tenet and requires several techniques, principles, and concepts. To ensure smooth flow
requires changes from traditional production in four ways: a stabilized master schedule,
reducing setup time and lot sizes, changing layout and maintenance, and cross-training
and engaging workers. These four changes and how they are related to ensure flow are
described next.
Chapter 7
Lean Thinking and Lean Systems
127] |)
Stabilize
One of the ways that firms can move toward achieving the lean tenet of simple, smooth,
Master
and error-free flow is to level the amount of work that is performed each day. For a service,
Schedule
this might mean scheduling a certain number of customers each day or using methods such
as advertising or pricing to even out the number of customers who want the service each
day. This practice is referred to as stabilizing the master schedule. Although this terminology is generally not used in services, the concept still applies. The process of produc— tion planning for manufacturing starts with a long-range production plan, which then is
broken down into annual, monthly, and daily plans.
Master scheduling for products is done to achieve a uniform load, the assignment of
approximately equal amounts of work to each machine or worker. Assume that one month
of production is scheduled in advance. Also assume that the monthly schedule calls for
10,000 units of product A, 5000 units of product B, and 5000 units of product C. If there
are 20 days of production in the month, the daily schedule will call for 1/20 of each model
produced in each day: SO0A, 250B, and 250C. Furthermore, the individual units will be
mixed as they go down the production line. The sequence will be /AABC/AABC/AABC/.
Note how two units of A are produced for every unit of B and C. Due to long changeover
(or setup) times in traditional manufacturing, large batches of final product are scheduled
to avoid changing the production process from one product to another. In traditional pro-
LO 7.3 Explainwhy
a stabilized master
schedule is required
for smooth flow.
duction with long changeover times, products A, B, and C might each be scheduled in
weekly batches sizes of 2500 units of A, 1250 units of B and 1250 units of C, assuming
one-fourth of monthly production is made before changing over.
Matching supply to demand is illustrated by the concept of takt time. 7akt is the
German word for the baton that an orchestra director uses to regulate the speed of the
music. In lean production systems takt time is the time between successive units of
production; this represents the speed of output. For example, a takt time of 2 minutes
means that one unit is completed every 2 minutes, or 30 units are produced in an hour
(60/2). In lean production systems the takt time of production should be set equal to
the average demand rate of the market to match production with demand and thus minimize inventories.
To establish the beat of the market, takt time can be computed by dividing the time
available for production (being sure to subtract for nonproductive time such as holidays
and lunch breaks) by the demand over the same period. For example, if market demand is
1000 units of a product per day and there are 7 hours of production time available in the
day (or 420 minutes), the takt time is then (420 + 1000) = 0.42 minute per unit. So, given
the available production time, production of one unit will have to be completed every
0.42 minute (about 25 seconds) to meet market demand. Producing at rates less than the
takt time will result in shortages in meeting the demand, and producing at rates greater
than takt time will result in building up inventory. The idea of takt time is to produce at a
constant rate that equals average demand of the market.
The objective of the lean production system is to produce the right quantity each day—
no more and no less. This minimizes finished-goods inventory since the production output
is closely matched to demand. This also helps reduce work-in-process and raw-materials
inventories, since stabilizing the master schedule provides nearly constant demands on all
work centers and outside suppliers. Contrast this with traditional mass-production processes in which lot sizes are large, and are not matched to market demand, resulting in
large inventories of finished goods, work in process, and raw materials.
Reducing
To achieve a uniform daily master schedule and smooth flow requires small lot sizes
Setup Time
and Lot Sizes
throughout each step of the production process. The lots produced are much smaller than
traditional manufacturing, which has long setup times between lots. Setup time is the
] | 128| Part Two
LO7.4
Process Design
Explain
how setup time,
lot size, layout,
and maintenance
are related to lean
thinking.
nonproductive time when machines are being adjusted before beginning work on a new
batch of parts. In a service setting, it might include readying a customer for service, for
example, guiding a customer from the waiting room to an exam chair, and adjusting the
chair position for a dental exam.
Reducing setup time, also referred to as changeover, is the key to small lot sizes. Shorter
setup times allow manufacturing and service processes to more closely match the rate of
demand. Then, the takt time can represent both the demand rate and the production rate.
Traditional manufacturing managers have focused on reducing production run times
per unit and gave too little attention to setup times. When long runs of thousands of
units are anticipated, run times naturally are more important than setup times. Since
setup time has received so little attention, phenomenal reductions are possible. For
example, at General Motors the time required to change a die in a large punch press was
reduced from six hours to 18 minutes. This allowed dramatic reductions in inventory
at this work center from $1 million to $100,000, and much smaller lot sizes. Hospitals
focus on surgical suite turnaround, or setup time between one patient departing and the
next patient’s surgery beginning. Some hospitals using lean concepts report as much as
30-minute reductions in turnaround, reducing this non-value added time, and essentially
increasing capacity for surgeries.
Single setups, sought in many companies, refer to setup times that are completed in
single digit minutes—less than 10 minutes per setup. This is often referred to as SMED
(single minute exchange of die). These low setup times can be achieved by designing a twostep setup process. First, external and internal setups are separated. Internal setup refers
to actions that require the machine to be stopped, whereas an external setup is preparation
that can be done while the machine is operating. The external setup is analogous to the ondeck batter in baseball; the player is warmed up and ready to move into position as soon
as the prior batter is finished. After internal and external setups are separated, as much of
the setup as possible is converted from internal to external. This is done, for example, by
using two sets of dies, one inside the machine and one outside; having quick change adjustments; and employing cleverly designed tools and fixtures. Once the machine is stopped,
it can be quickly readied for a new product since internal setup has been minimized. Much
improvement in setup time can be accomplished once people realize the importance of it
in stabilizing the master schedule and producing only
what is pulled by the next process.
Workers can practice setups to reduce the time. For
example, hospital personnel practice moving patients
from the emergency room to surgery so they can resolve
(before a real patient arrives) problems such as slow
elevators that might cause a delay. A good example of
quick changeover comes from the racetrack. Race cars
are quickly fueled, tires are changed, and the windshield
is washed in short pit stops, the racing equivalent of a
setup, by highly coordinated and trained pit crews using
specialized tools and equipment. Southwest Airlines is
renowned for its quick turnaround (setup) times with its
planes, utilizing the techniques discussed here. Much
of its success has been attributed to its ability to keep
Pit crews make quick changeovers of race cars in amazingly
planes productive in the air while minimizing time
short times.
Source: U.S. Air Force photo by Master Sgt Michael A. Kaplan
spent on the ground.
Chapter 7
FIGURE 7.4
Cellular manufacturing
in a U-shaped cell.
Equipment and
workstations are
arranged to facilitate a
continuous production
flow in small lots or
single units.
Lean Thinking and Lean Systems
1129] |)
INPUT
Components
and
Materials
OUTPUT
Finished
Goods
Changing
Layout and
Maintenance
Ensuring flow (tenet 3) has a natural effect on layout and equipment. The manufacturing
or service facility evolves toward a more streamlined flow because lot sizes are reduced,
inventory is held close to where it is used, and problems are resolved with standardized
solutions so that the problems do not arise again in the future. Moreover, since inventory is
typically kept low—only a few hours or days of supply—work spaces can be much smaller
because of the reduced storage space needed. One comparison of manufacturing plants
showed that lean plants need only one-third of the space of non-lean plants.
The effect of lean thinking on layout in a manufacturing plant is dramatic. In a nonlean layout suppliers deliver to stockrooms and parts are returned to the stockrooms after
certain stages of production are completed. In a lean production system all stockrooms
have been eliminated as stock is kept on the shop floor very close to where it is used.
This eliminates the wasted space of the stockroom and the wasted transportation moving
materials in and out of it. Work centers can eventually be organized into a group technology, or cellular manufacturing, layout. Cellular manufacturing is often arranged in a
U-shaped cell, and ensures flow from one machine or task to the next without interruption; see Figure 7.4. Parts (customers in service settings) flow smoothly between machines
and tasks within the cell. Most of the inventory buffers,
which are next to each machine, have been eliminated.
It is a natural consequence of lean thinking to evolve
toward cellular manufacturing.
With a lean production system, preventative maintenance of equipment is extremely important. This
is the regular checking of and caring for equipment,
much like oil changes and regular inspection of your
own car, so that it does not break down or fail. Since
In a lean system, stock is held on the shop floor.
SolStock/Getty Images
inventories have been cut to minimal levels as workers
strictly adhere to the daily plan for production, it 1s critical to avoid unplanned equipment failures. A lean system requires the right amounts of capacity, inventory,
workers, and everything operating as planned—each
and every day. Workers take responsibility for most of
their own equipment and workspace maintenance, and
b ||130] Part Two
Process Design
this gives them more control over their work and work environment. Repairs by maintenance personnel are performed between shifts so that it does not disrupt the regular daily
work schedule.
Cross-Training
and Engaging
Workers
LO7.5
Differentiate
how employees
are unique in lean
systems.
7.4
LO7.6
A final element in ensuring flow is cross-training and engaging workers who have a much
broader set of skills than their counterparts in traditional environments. Cross-training
of workers is therefore a critical human resources responsibility. Workers are trained and
experienced to operate several machines in a manufacturing setting or perform a variety
of tasks in a service setting. In manufacturing, the worker can shut down one work center
when there is no work to do there and move to another work center where parts are needed.
The worker should be trained for several different work areas to set up machines, do routine maintenance, and inspect the parts. The main benefit of cross-trained workers is that
they provide flexibility to the production system. Nurses who can be moved to different
hospital units and production workers who can be shifted to different production areas
place labor resources just where they are needed at any given time. This reduces the need
to have excess labor in specific work areas just in case capacity is needed on short notice.
Moving toward a flexible workforce may require changing the way workers are paid and
rewarded. Traditional pay systems are often based on seniority and level of job skills. New
pay systems in a lean system reward workers on the basis of the number of different jobs
they can perform. This will encourage workers to learn more skills and become more flexible. Labor unions often are organized along skill lines, and they do not tend to encourage
flexibility in the workforce. As a result, management needs to work closely with unions
to develop trusting relationships so they can work together to build the kind of workforce
needed for lean systems.
Lean is built on the idea of “respect for people.” It is respect for every person’s ideas,
desire to do a better job, and commitment to improve. This is an essential principle of the
Toyota Production System that engages employees to accomplish lean objectives. Managers must earn respect by trusting workers and staff to improve the system and encourage
them to participate in decision making. Respect also extends to customers and suppliers throughout the supply chain. Without an emphasis on “respect for people,” efforts to
implement lean systems will fail.
In lean systems, quality teams and suggestion systems are used to actively engage workers and engineers in problem-solving activities. Since inventory is not available to cover up
problems, an environment of participation, respect, and teamwork must be created to get all
employees to contribute individually and collaboratively toward production requirements
and problem solving. Lean production systems cannot be implemented without worker
understanding and cooperation. Management in all organization functions must ensure that
the workers understand their new roles and accept the lean approach.
CUSTOMER
Design a
kanban system
to achieve
customer pull.
PULL
It is not enough to arrange for smooth flow; the flow must be controlled. Lean thinking in
tenet 4 uses a pull system to control the flow of production. Production is only authorized
when the downstream work center requires it. A pull system is achieved in manufacturing
by kanban.
Kanban is the method of production authorization and materials movement in the lean
production system that supports the tenet of producing only what is pulled by the customer.
Kanban in the Japanese language means a marker (card, sign, plaque, or other device) used
to control the timing and movement of parts through a sequential process. The kanban
Chapter 7
Lean Thinking and Lean Systems
134]
system is a simple and visual “parts withdrawal system” involving cards and containers to
pull parts from one work center to the next just in time. In services, a kanban system might
be used to control inventory levels (by limiting items to fixed-quantity bins), or to pull
many types of work such as paperwork, or even the customer, through the process! Since
kanbans are mostly associated with manufacturing, we discuss how they are used to pull
inventory through a production system.
The purpose of the kanban system is to signal the need for more parts and ensure that
those parts are produced just in time to support subsequent work centers. Parts are kept
in small containers, and only a specific number of those containers are provided. When
all the containers are filled, production is stopped, and no more parts are produced at
that work center until the subsequent (receiving) work center provides an empty container.
Thus, work-in-process inventory is limited to available containers.
The final assembly schedule is used to pull parts from one work center to the next just
in time to support production needs, which are aligned with market demand. Only the final
assembly line receives a schedule from the dispatching office, and this schedule is nearly
the same from day to day. All machine operators and suppliers receive production orders
(kanban cards) from the subsequent (receiving) work centers. If production stops for a time
in the receiving work centers, the supplying work centers also stop soon because their parts
containers become full since their output is not being pulled by the receiving work centers.
The kanban system can be extended to suppliers so that the suppliers also respond (deliver)
only when parts are pulled by the factory.
To see how the kanban system works as a physical control system, assume that eight
containers are used between work centers A and B (A supplies B) and that each container
holds exactly 20 parts. The maximum inventory that can exist between these two work
centers is 160 units (8 x 20) since production at work center A will stop when all the containers are filled.
In the normal course of a day, the eight containers might be distributed as shown in
Figure 7.5. Three containers filled with parts are located at work center A in the output
area. One container is currently being filled by the machine at work center A. One full
FIGURE 7.5
Kanban system.
Production card
Production
kanban post
a
Withdrawal card
encarta
ates
Work
center B
In transit
Withdrawal
kanban post
i|| 132| Part Two
Process Design
container is being moved from A to B, two full containers are sitting in the input area of work center B, and one
container is being used at B. These eight containers are
needed since work center A also produces parts for other
P|
;
work centers, machines at A may break down, and move
?
times from A to B are not always exactly predictable.
Some companies control the movement of containers
by using two types of kanban cards: production cards
and withdrawal (move) cards. These cards are used to
authorize production and to identify the parts in any
container. Instead of using cards, production can also
be controlled by kanban squares that visually signal the
}t
5
;
H$
need for work (to fill the kanban square), or by visual
control of the empty containers.
Most importantly, the kanban system is visual in
nature. As empty containers accumulate, it is a clear
signal that the producing work center is falling behind.
When
all the containers are filled, production is stopped.
KANBAN SQUARE AT HONEYWELL. The dashed rectangle
signals the need for the production of a cabinet. Only one
The production lot size is exactly equal to one container
cabinet is placed on this square at a time. When the square is
of parts. All parts are neatly placed in containers of a
emptied by subsequent production, another cabinet is produced.
fixed size. All of these are visual indicators of the work
©Tulasi Ranganathan/Honeywell
that should be done or stopped.
The number of containers needed to operate a work center is a function of the demand
rate, container size, and the lead time for a container. This is illustrated by the following
formula:?
n=——_—
6.
se
where n = total number of containers
D = demand rate of the using work center
C = container size, in number of parts, usually less than 10 percent of daily demand
T = time for a container to complete an entire circuit: filled, wait, moved, used, and
returned to be filled again (also called lead time)
Example
Suppose demand at the receiving work center B is 2 parts per minute and a standard container holds 25 parts. It takes 100 minutes for a container to make a complete circuit from
work center A to work center B and back to A again, including all setup, run, move, and wait
times. The number of containers needed in this case is:
_ 2(100)_
25a
The maximum inventory in the production system, a useful measure of how lean the
system is, equal to the container size times the number of containers (200 units —= Sexe o)s
since the most inventory we can have is all containers filled:
Maximum inventory = nC = DT
* Safety stock can be added to the numerator to account for uncertainty in demand or time.
Chapter 7
Lean Thinking and Lean Systems
133} |]
Inventory can be decreased by reducing the size of the containers or the number of containers used. This is done by reducing the lead time, the time required to circulate a container.
When any of these times have been reduced, management can remove kanban cards from the
system and remove a corresponding number of containers. It is the responsibility of managers
and workers in a lean system to reduce inventory by means of a continuous cycle of improvement. Reducing lead time by reducing the fill, wait, move, use, or return times is the key.
A complete kanban system can link all work centers in a production facility. It can,
moreover, link the production facility to its suppliers. With such a kanban system, all material is pulled through the production system by the final assembly schedule, using a highly
visible shop-floor and supplier control system.
A pull system also works in a service environment where the flow is of customers and
information, not materials. When customers are pushed through a system, they tend to wait
in lines for services as bottlenecks occur and queues develop. A pull system signals customers to move when they can be served by the next process. Likewise, in offices where paperwork, or electronic information, is being processed from one employee to the next, a pull
system will limit the amount of work that can develop between employees. There is no need
to continue processing information when the next process does not pull it through. The
supplier should stop and work on something else, which, of course, requires cross-training.
At this point, the reader should recognize these characteristics of a pull system in either
manufacturing or service.
¢
Visibility of a queue of customers, information, or inventory (for manufacturing).
¢
Limits are defined for the length of the queue, or amount of inventory.
¢
Work stops when the queue is filled at that location and employees move to other tasks.
¢
The employees manage the flow through the system.
This completes our discussion of the techniques and concepts used to create a pull system.
7.5
CHANGING
LO 7.7 Compare
lean suppliers
to traditional
manufacturing
suppliers.
RELATIONSHIPS WITH SUPPLIERS
It’s not enough to institute a lean system internally in a firm; the company’s supplier relationships must also radically change. In a lean production system, suppliers are treated the
same way internal work centers are treated. Suppliers may receive kanban cards and special
containers, and they are expected to make frequent deliveries just in time for the processes
using those supplies. Suppliers are viewed as the external factory and as part of the production team, in line with modern supply chain management in which key suppliers are considered partners.
Several deliveries may be made each day if the supplier is located in the same vicinity; this is referred to
as co-location. Suppliers located at a distance may
have local warehouses where they receive bulk shipments and then break them down for frequent deliveries
to the customer.
Lean supplier partnerships may require frequent deliveries in
small lots.
Mario Tama/Getty Images
This is not desirable, however,
since
too much inventory builds up in the pipeline and lead
times from the factory to the warehouse can be long.
Local suppliers with short lead times are preferred.
Nevertheless, in some cases implementing lean production systems has simply shifted some of the inventory
from the customer to the suppliers when long distances
are involved or suppliers still have long setup times and
large batch sizes.
é||134| Part Two
Process Design
Suppliers are given specific delivery times rather than shipment dates. For example, a
supplier may be required to deliver parts that will last for two hours of production at 8 a.m.,
10 a.m., 12 noon, and 2 p.m. On each delivery, the supplier picks up empty containers
and associated withdrawal kanban cards, which lists the part number, the part name, and
the quantity required. Only that number of empty containers is filled for the next delivery.
Deliveries are made directly to the production line location where the supplies will be
used, bypassing incoming inspection. Ina retail setting, this may mean the supplier putting
inventory directly onto the grocery store shelf. This requires the receiving customer to have
complete confidence in the supplier’s quality. It also greatly reduces paperwork, lead time,
inventory, the number of receiving areas, and required storage space.
For cases in which it is too expensive to make several deliveries each day, several suppliers
may coordinate to make round-robin deliveries. For example, one supplier goes to three other
suppliers plus his or her own for the 8 a.m. delivery. Another supplier will make the 10 a.m.
run, and so on. This method can save on transportation expenses for small-lot deliveries.
Lean production systems tend to use fewer suppliers. This is done to establish long-term
relationships with the suppliers and to work together to ensure the quality of the items
needed. A complete reversal of thinking is needed here since we ordinarily assume that
fewer suppliers might price-gouge the customer, and that more sources are needed to keep
prices low. Supplier prices, however, can be managed with long-term contracts that include
negotiated price stability. This requires a totally different type of supplier—customer relationship from what was typical in the past.
Many companies develop an “integrated supplier program” to move toward a lean production system. The features of this type of program are as follows:
|. Early supplier selection. Suppliers are selected before the parts reach final design, and
so the design can be worked out completely with the suppliers.
2. Family-of-parts sourcing. A supplier takes responsibility for an entire family of parts,
allowing the supplier to establish economic volumes and reduce the number of separate
deliveries.
3. Long-term strategic relationship. An exclusive contract for the life of the part can be
given to a supplier in exchange for a specific price schedule. In other cases two or more
suppliers of the same part may be used to ensure reliable supply, but nevertheless, longterm strategic relationships are established.
4. Paperwork reduction in receiving and inspection. Reducing the work associated
with each order and delivery results in a direct savings to the customer and supplier.
Changing supplier relations is one of the most important aspects of creating a lean production system. As firms move toward leaner supply chains with fewer suppliers, they reduce
inventory across the entire supply chain. These supply chain partners then work to improve
flow across the entire supply chain.
Is it possible for a supply chain to be too lean? When inventory is significantly reduced,
a supply chain may be at increased risk of shutting down. Typically, inventory serves as a
buffer between supply chain firms so that if one firm slows or stops production, inventory
can keep the rest of the supply chain flowing. Production slowdowns can be due to natural (weather) or man-made (worker strikes) disruptions. Having fewer suppliers, a natural
characteristic of lean systems, can also increase the risk of supply chain disruption. Fewer
suppliers means fewer redundancies, or back-up arrangements, and maintaining production
flows across lean system suppliers becomes even more important. Being too lean naturally
increases some risks of supply chain disruption. There is a point at which a supply chain can
be too lean and upstream or downstream disruptions can grind the supply chain to a halt.
Chapter 7
7.6
IMPLEMENTATION
LO7.8 Explain how
to implement a lean
system.
Lean Thinking and Lean Systems
1435] |]
OF LEAN
Implementing lean thinking may sound simple but it is, in reality, very challenging. Management will be most effective if they clearly communicate a sustained commitment to lean thinking with all workers. Lean requires a change in philosophy and culture along with changes in
practices. New practices could include a reward system focused on creating flexible capacity,
engaging workers in problem solving, providing time and resources to root out waste, and
creating closer relationships with key suppliers. Table 7.3 provides a summary of the five lean
tenets and the practices associated with each one. Not all these tenets and practices must be
implemented in every situation. The approach should be customized to the situation.
Deploying lean thinking often starts and is maintained through kaizen events.’ A kaizen
event can take between two days and one week and is focused on creating significant
improvement in performance (quality, speed and cost) in one particular area of operations,
for example, the shipping area or an area that has experienced a recent increase in quality
problems. Kaizen is about fixing the process, not planning to fix it.
Kaizen events have helped Dr Pepper Snapple Group Inc. increase profits during industrywide declining sales in the U.S. soda market. Nearly all employees participated in teams
that used gemba (observation of the production process) to find waste and then eliminate
it. Improvements have included reducing average setup times between batches of different fountain syrup flavors from 32 to 13 minutes, reducing inventory holding space by
1.5 million square feet, and simplifying packaging to reduce printing costs. In five years,
kaizen events have totaled $270 million in savings.
To facilitate implementation of kaizen the following approach is suggested:
1. Establish a team of employees that will study the process that needs improvement.
These employees should come from different functions and levels of the organization to
represent all stakeholders involved with the process.
2. Have the team determine what the customer values. The customer can be internal (the
next process) or external to the organization. Only the customer can specify what is
valued in the good or service.
TABLE
7.3
Lean Tenets,
Techniques, and
Concepts
Lean Tenets
:
Techniques and Concepts
Create product or service value from
:
3
customer’s perspective
Understand customer-defined value
ae
Muda (waste)—work to eliminate
Identify, study, and improve the value stream
Value stream mapping
Gemba (observation)—work to improve flow
Ensure simple, smooth, and error-free flow
Stabilizing master schedule
Uniform load and Takt time
Cross-training workers
Reducing setup time and lot size
Cellular manufacturing
Preventative maintenance
Produce only what is pulled by customer
Kanban (signal)—visual control system
Supplier relationships/Co-location
Strive for perfection
5S—organize the work space
5 Whys—find root causes of problems
Kaizen (continuous improvement)
2 Kaizen is a Japanese word that means continuous improvement.
il|| 136] Part Two
Process Design
3. Construct a value stream map of all process steps and the associated times or value
that is added. Then analyze the value stream map and eliminate non-value-adding
activities by searching for the seven forms of waste. The 5S and 5 Whys techniques may be applied at this step to identify and reduce wasted time, space, effort
and resources.
4. Ensure flow to be smooth, steady, and error free to meet customer demand as it occurs.
Stabilize the master schedule to match market demand, produce in small lots, change
the layout, and cross-train workers.
5. Use customer demand to pull the flow of work through the process. Do not produce
until output is required by the customer. Let the customer signal when work from the
process is needed.
6. Implement the necessary changes identified by the team to achieve lasting improvement. Then repeat the cycle on another internal process or to the processes of upstream
suppliers and downstream customers.
Lean concepts, principles, and techniques are increasingly being deployed in diverse
settings. For example, the seven forms of waste can be observed in health care delivery, as shown in Table 7.4. In health care, overproduction may be seen in excessive or
unnecessary lab tests or surgery for a patient who may have been treated more easily and
inexpensively with physical therapy. Unnecessary transportation may involve moving a
patient among hospital departments for X-rays and other tests, when it may be possible to
complete such tests in the patient’s own room. Nurses often walk miles per day to complete their work, a form of unnecessary motion. Even government services can benefit
from lean thinking. Implementation of lean systems is also becoming common in many
administrative processes.
Lean tenets once were believed to be applicable only to mass production. This is
no longer true as the impact and benefits of lean are demonstrated again and again in
diverse manufacturing and service settings and across the supply chain. Lean thinking’s major impact in changing firm practices worldwide is being compared to the
Ford moving assembly line as one of the great innovations in operations and supply
chain management.
TABLE 7.4
Waste in Health Care
Overproduction
*
*
*
Multiple forms asking for the same information
Multiple copies of reports
Multiple lab tests
Waiting time
*
*
Waiting for lab test results
Unnecessary transportation
*
Patients being transported within and between hospitals
Excess processing
*
Re-entering patient information into system
Too much inventory
* . Overstocking of medication
* Overstocking of supplies
Unnecessary motion
*
Defects
Patient waiting
*
Walking to and from storage room
Searching for missing patient information and charts
*
Medical and surgical errors
*
Order-entry errors
Incomplete forms
Chapter 7
7.7
Lean Thinking and Lean Systems
137] |
KEY POINTS AND TERMS
Lean concepts, principles, and tenets can be deployed to reduce waste in manufacturing
and service firms. We have seen how the lean tenets create lean production systems with
non-value-added activities eliminated and waste minimized. Key points in the chapter
include the following:
¢
Lean thinking is a way of thinking about processes that includes five tenets: specify
customer value, improve the value stream, flow the product or service, pull from the
*
The five lean tenets seek to eliminate waste by utilizing the full capability of workers
customer, and strive for perfection.
and partners in continuous improvement efforts. Lean tools, or methods, are described
for each of the five tenets.
¢
In manufacturing,
smooth
flow is ensured
by a stable and level master schedule.
This requires consistent daily production within the master schedule and mixed
model assembly. Takt time matches the rate of output with the average demand rate
in the market.
¢
Reducing lot sizes, setup times, and lead times is the key to decreasing inventories in a
lean production system and ensures smooth flow. Service and administrative activities
should also work toward a fast changeover from one customer to the next and a reduced
lead time.
¢
The plant layout in a lean production system requires much less space and encourages
evolution toward cellular or group technology layouts.
¢
A lean system requires cross-trained workers who can perform multiple tasks. A flexible workforce will require changing the way workers are selected, trained, evaluated,
and rewarded.
Key Terms
¢
A kanban system is used to pull parts through the production system. A fixed number of
containers is provided for each part, thus limiting the amount of work-in-process inventory. The pull system can also be applied in service operations by providing only what
is needed when it is needed by the customer.
*
New supplier relationships must be established to make lean production successful.
Frequent deliveries and reliable quality are required. Often, long-term single-source
contracts will be negotiated with suppliers.
¢
Kaizen emphasizes continuous improvement. Kaizen events are used to implement lean
thinking improvements quickly in one week or less on a particular process.
¢
Lean concepts, principles, and techniques can be applied to design, manufacturing, distribution, services, and the supply chain.
Toyota Production System
(TPS) 119
Just-in-Time (JIT)
manufacturing 119
Lean production 119
Lean thinking 120
Waste (muda)
121
Value stream 121
Value stream
mapping 121
Gemba 121
Push 123
Pull 123
Perfection 124
5 Whys 125
3S) 125
Stabilizing the master
schedule
Internal setup 128
External setup 128
Cellular manufacturing 129
Preventative maintenance 129
Cross-training 130
Respect for people 130
127
Uniform load 127
Takt time 127
Reducing setup time
Single setups 128
Kanban
128
130
Reducing lead time 133
Supplier relationships 133
Co-location 133
Kaizen 135
s ||138| Part Two.
Process Design
ee
___s____
EE
LEARNING
ENRICHMENT
Introduction to Lean Thinking—Gemba Academy
https://youtu.be/a2551kYgIpI
Video
6:35
(for self-study or
Routing Out Waste in a Hospital
Video
https://youtu.be/jZLtbye--sg
8:46
tes)
In suitesOleh Cs enue
Video
9:13
Lean Manufacturing Tour—SS implementation
https://youtu.be/mqgHUwSakj8
The Toyota Production System
https://youtu.be/P-bDIY WuptM
.
Video
4:14
Push vs. Pull with Kanban Simulation
Video
https://youtu.be/a7
YvJBOn 161
8:48
SOLVED PROBLEMS
Problem
1. Kanban and takt time. A work center uses kanban containers that hold 300 parts. To
produce enough parts to fill the container, 90 minutes of setup plus run time are needed.
Moving the container to the next workstation, waiting time, processing time at the next
workstation, and return of the empty container take 140 minutes. There is an overall
demand rate of nine units per minute.
a. Calculate the number of containers needed for the system.
b. What is the maximum inventory in the system?
c. A quality team has discovered how to reduce setup time by 65 minutes. If these
changes are made, can the number of containers be reduced?
d. What is the takt time for this process?
Solution
a. T is the time required for a container to complete an entire circuit, in this case
90 minutes for setup and run time plus 140 minutes to move the container through
the rest of the circuit.
n= DT
+C=(9 x (90 + 140)) + 300 = 6.9 (round up to 7)
b. Since production will stop when all the containers are full, the maximum inventory
is when all containers are full, that is, nC:
nC = 7(300) = 2100
c.
n= DT+C=(9
X (25 + 140)) + 300 = 4.95 (round up to 5), so yes, the number of
containers can be reduced from 7 to 5.
d. Takt time = 1/9 minute = 60/9 seconds = 6.67 seconds. Since the process produces
9 units per minute, the takt time is 1/9 minute or 6.67 seconds per unit.
Problem
2. Kanban. Work center A produces parts that are then processed by work center
B. Kanban containers used by the work centers hold 100 parts. The overall rate of
demand is 4.5 parts per minute at work center B. The table below shows setup, run,
move, and wait times for parts at each of the work centers.
Chapter 7
Lean Thinking and Lean Systems
139 ||)
Work Center
Setup
Run time per unit
A
B
4
0.1
3
0.4
Move time
2
6
Wait time
10
20
a. What is the minimum number of containers needed between these two work centers?
b. Assume that two extra containers are available (at no extra cost). If these work cen-
ters use the two containers, what is the maximum parts per minute that could be
expected to flow through these two work centers? Could the work centers handle a
demand of 8.5 parts per minute?
Solution
a. Tis the time required for a container to go through both work centers and back to its
starting point. Therefore,
7= 4 + 3 = 7 minutes of setup time, 100 x (0.1 + 0.4) =
50 minutes of run time, 2 + 6 = 8 minutes of move time, and 10 + 20 = 30 minutes
of wait time. Thus,
n= DT
7= 7+ 50 + 8 + 30 = 95 minutes.
+C= (4.5 X (95)) + 100 = 4.275 (round up to 5 containers)
b. Assume two extra containers are available.
n= DT=+C
5$+2= (Dx (95)) = 100
D =7 X (100 + 95) = 7.37 parts per minute
Since 7.37 < 8.50, the work centers cannot handle a demand rate of 8.5 units per
minute if they have seven containers.
Discussion Questions
1. Visit a manufacturing facility in your area. What are
the major causes of inventory? Be sure to ask about lot
9. Are there manufacturing firms that should not use
lean? Why?
sizes and setup times. Would a lean production system
work in this facility? Why or why not?
2. Why did the concepts, principles, and techniques of lean
emerge and evolve in Japan, not in the Western countries?
10. Find an example from the Internet of the application of
lean thinking to a service operation. Describe how the
lean tenets are applied in this setting.
11. How can lean thinking be applied to accounting,
3. State the lean tenets in your own words.
4. Why is a stable master schedule desirable for a lean
production system? What is the effect if it is not stable?
5. How can lot sizes and inventories be reduced in a lean
production system? Discuss specific approaches.
finance, human resources, and marketing processes?
12. Identify some of the seven forms of waste in the following situations.
a. Restaurant
b. Doctor’s office
6. Describe typical supplier relations before and after
embracing lean tenets.
7. How do workers and managers in a lean production
system differ from their counterparts in traditional non-
13. Construct a value stream map for the following
processes.
a. Cafeteria
b. Grocery store
ea ents!
8. Discuss how lean thinking can lead to a reduction of
14. What does it mean to say that a supply chain can be too
lean? Give examples why this may be a problem.
costs (material, labor, overhead), other than inventory.
Be specific.
1) |140, Part Two
Process Design
Problems
1. Calculate the daily production quantities and sequences
from the following monthly requirements for product
models A, B, and C. Assume the month has 20 produc-
tion days.
6. A company is in the business of machining parts that go
through various work centers. Suppose work center A
feeds work center B with parts. The following times (in
minutes) are given for each work center.
a. 5000 A, 2500 B, and 3000 C
Work Center
b. 2000 A, 3000 B, and 6000 C
Suppose a lean work center is being operated with
a container size of 25 units and a demand rate of
100 units per hour. Also assume it takes 180 minutes
for a container to circulate.
a. How many containers are required to operate this
system?
b. What is the maximum inventory that can accumulate?
c. How many kanban cards are needed?
. Fora particular operation, the setup time is 10 minutes
and the run time is 50 minutes to produce a standard
lot of 40 parts. It takes three additional hours to circulate a container of parts after production is completed.
The demand rate is 20,000 parts per month. There are
160 production hours in a month.
a. How many standard containers are needed?
b. What is the takt time of this process?
Assume that a plant operates 2000 hours per year and
the demand rate for parts is 100,000 units per year. The
circulation time for each kanban container is 24 hours.
a. How many kanban containers are needed for a
container size of 100 parts?
b. What would be the effect of reducing the container
size to 60 parts?
c. What is the takt time for this process?
d. What takt time is needed for 80,000 units per year?
Suppose a work center has a setup plus run time of
30 minutes to make 50 parts. Also assume it takes
10 minutes to move a standard container of 50 parts to
the next work center and the demand rate is one part per
minute throughout the day.
a. Schedule this situation by drawing a picture of when
work center A should be producing and idle and
when movements of containers take place from A to
B, the using work center.
b. How many standard containers are needed for this
part to circulate from the picture in part a?
c.
Use the formula n = DT ~ C to calculate the number
of containers.
Setup time
Run time (per part)
Move time
A
B
3
O35
6
2
Om
8
A standard kanban container holds 50 parts that are
transferred from work center A to work center B. The
demand rate at work center B is four parts per minute.
a. How many kanban containers are needed for this
situation?
b. If move time is cut in half, what does this do to the
number of containers needed? How much will this
change reduce inventory?
. Suppose that a lean work center is being operated
with a lot size of 50 units. Assume that 200 parts are
demanded per hour and it takes three hours to circulate a container, including all setup, run, move, and
idle time.
a. Calculate the number of kanban containers required.
b. What is the maximum inventory that will accumulate?
c. What can be done to reduce the inventory level?
Suggest alternatives.
~
. A supplier provides parts to a manufacturing company that demands frequent deliveries. At the present
time it takes six hours to make a round trip between
the supplier’s warehouse and the customer, including
loading, travel, and unloading time. The lot size is
12 pallet loads on a truck, and the manufacturer uses
2 pallets per hour.
a. How many trucks are needed to ship the pallets to
the manufacturer?
b. What is likely to happen if the truck breaks down?
c. How can the supplier ensure that the customer does
not run out of parts even in the face of delivery
problems or other uncertainties?
d. What will happen to the supplier if the manufacturer
runs into trouble and shuts down for a period of
six hours?
PART
Quality
8.
Managing Quality
9.
Quality Control and Improvement
Quality is one of the four objectives of operations, along with cost, delivery, and
flexibility. To meet the quality objective, it is important to manage and control all
aspects of the quality system. Chapter 8 begins this part with a discussion of
managing quality, and Chapter 9 addresses quality control and improvement.
The main contribution of Part Three is a broad treatment of quality, which includes
management, planning, and policy concerns in addition to the more traditional
statistical topics. In practice, quality is primarily a management problem, and statistical
methods are used to first stabilize a system and to then achieve continuous
improvement of a stable system. @
Ce
AGP
seeaR
Managing Quality
ee
ee
eee
ee
ee
LO8.4
Apply the quality cycle to a product or service.
LO8.8
Compare and contrast ISO 9000 standards and the Baldrige Award criteria.
rt
Quality is one of the four key objectives of operations, along with cost, flexibility, and
delivery. While quality management is cross-functional in nature and involves the entire
organization, operations has a special responsibility to produce a quality product or service
for the customer. This requires the cooperation of the entire organization and careful attention to the management and control of quality.
Quality management has had many different meanings over the years. In the early 1900s,
managing quality meant inspection, which was the primary method used to ensure quality
Chapter 8 Managing Quality
1143] |)
The Ritz-Carlton Hotel Company, L.L.C.
The Ritz-Carlton Hotel Company, L.L.C., is a five-star
hotel management company that develops and oper-
'
sive
4
4%
4
|
ates 91 luxury hotels
and resorts
in 30
countries. The company targets primarily
industry
executives,
meeting
and
J
4
planners,
prestigious
travelers.
It
employs
40,000 people who
are
highly
trained
and
motivated
to
provide
quality
ser-
vice. The Ritz-Carlton
Hotel Company is the
Leonard Zhukovsky/Shutterstock
only hotel company
to win two Malcolm Baldrige awards.
Ritz-Carlton translates customer requirements into
employee requirements through its Gold Standards and
its strategic planning process. They have taken what
their customers want most and designed the simplest
ways to provide them. Ritz-Carlton’s data show that its
employees’ understanding of the Gold Standards is
directly correlated with guest satisfaction.
Employees
respond to a customer’s
requirements at
both the team and individual levels. They provide highly
personal, individual service. Customer
are captured and entered
likes and dislikes
into a guest history that pro-
vides information on the personal preferences of hundreds of thousands of repeat Ritz-Carlton guests. When
a customer stays at any of their hotels the information is
provided to the employees serving that customer.
If an employee detects a problem, the employee is
empowered to do whatever it takes to make the customer happy immediately, or the employee can call on
any other employee to assist. Such a system depends
on well-trained, perceptive, and motivated employees
along with a well-defined service delivery system. With
more than a million customer contacts on a busy day,
Ritz-Carlton understands that its customer and quality requirements must be driven by each _ individual
employee at the lowest level of the organization.
Results indicate that Ritz-Carlton hotels are doing an
exceptional job of translating customer requirements
into employee behavior and excellent systems. Ninetyseven percent of Ritz-Carlton’s customers report having a
“memorable experience” while staying in one oftheir hotels.
Source: www.ritzcarlton.com, 2020.
products (services were rarely inspected). In the 1940s, managing quality took on a statistical
connotation as statistical methods began to be used to control quality within the natural variation of the transformation process. Statistics pioneer Walter Shewhart developed statistical control charts to learn to maintain a production process within a state of statistical control and thus
reduce the amount of inspection required. In the 1960s, the meaning of quality management
was expanded to include activities across the entire organization as all functions contributed
to designing and producing quality. Quality was seen not as just an act of production; rather,
it was something the entire organization should strive to provide for the customer. Now, quality management is taking on a broader meaning, including continuous improvement, competitive advantage, and a customer focus together with quality along the entire supply chain.
The Operations Leader box shows how Ritz-Carlton Hotel, a two-time winner of the Malcolm
Baldrige National Quality Award, is implementing modern quality management principles.
8.1
QUALITY AS CUSTOMER
LO8.1.
Explain
quality, from
a customer
perspective.
REQUIREMENTS
Quality is defined as “meeting, or exceeding, customer requirements now and in the future.”
This means that the product or service is fit for the customer’s use. Fitness for use is related
to the benefits received by the customer. Benefits are based on the totality of features and
characteristics that determine the ability of a product or service to satisfy given needs. Only
the customer, not the producer, can determine if the product or service has the right benefits.
Py | 144] Part Three
Quality
Customer satisfaction is related to quality. It means that the quality
of the product or service meets the customers’ expectations. For example, the car functions as the customer expects, or a service is delivered in
a manner that meets customer expectations. The customer may receive
great benefits, but relative to expectations the customer is not satisfied.
Customer satisfaction is therefore a relative measure of quality.
Customer expectations can depend on many things. Advertising
can determine the customer’s expectations. Another factor is: Does the
customer know how to properly use the product or service? The product can be more difficult or easier to use than the customer expected.
RITZ-CARLTON. Any of the employees of
Also, the customer’s expectations may change over time as the cus-
the
Ritz Cabtonucen spendipite baal je
immediately correct a guest’s problem or
tomer becomes: familiar with different products
and services available.
y
handle a complaint. Its employees are the key
The expectations of customers of the Ritz-Carlton, for example, are
factor in Ritz-Carlton’s receipt of two Malcolm
certainly much different from those of a lower priced hotel.
Producers certainly attempt to meet customer requirements and
Sylvain Grandadam/age Fotostock
expectations through both design and production of the product or service. The producer specifies the quality attributes of the product or the service as carefully
as possible and then strives to meet those specifications while improving the production and
delivery process over time. Whether the resulting product or service meets the customer’s
requirements and expectations will be judged ultimately by the customer.
Baldrige National Quality awards.
8.2
PRODUCT QUALITY
LO8.2.
Characterize
product quality
based on four
dimensions.
For a manufactured good, the following dimensions of quality can be useful in understanding product quality:
*
Quality of design
*
Quality of conformance
¢
e
The “abilities”
Field service
‘
Quality of design is determined before a product is produced. This determination is
usually the responsibility of a cross-functional product design team, including members
from marketing, engineering, operations, and other functions.
Quality of design is determined by market research, the design concept, and product
specifications. Market research is aimed at assessing customer needs. Since there are different ways to meet those needs, a particular design concept must be developed. For example, the customer may need inexpensive and energy-efficient transportation—a need that
can be met by a large number of different automobiles, each representing a different design
concept. The design concept then is translated into a very detailed set of specifications for
the product, for example, a digital blueprint and bill of materials.
Quality of conformance means producing a product to meet the specifications. When
the product conforms to specifications, operations considers it a quality product regardless
of the quality of the design specifications. For example, inexpensive shoes will have high
quality of conformance if they are made according to specifications and low quality of
conformance if they do not meet specifications. Quality of design and quality of conformance thus represent two different uses of the term quality.
Another aspect of quality involves the so-called abilities: availability, reliability, and
maintainability. Each of these terms has a time dimension and thus extends the meaning
of quality over some time horizon. The addition of time to the definition of quality is, of
course, necessary to reflect continued satisfaction by the customer.
Chapter 8 Managing Quality
145 ||)
Availability defines the continuity of usability to the customer. A product is available
if it is in an operational state and not down for repairs or maintenance. Availability can be
measured quantitatively as follows:
Uptime
Availability =
Uptime + Downtime
Reliability refers to the length of time a product can be used before it fails. Reliability
is measured as mean time between failure (MTBF), which is the average time the product
functions from one failure to the next. The longer the MTBF, the more reliable the product.
For a cell phone, reliability is a measure of its expected lifespan until it fails.
Maintainability refers to the restoration of a product or service once it has failed. All customers consider maintenance or repairs a nuisance. Thus, a high degree of maintainability is
desired so that a product can be restored to use quickly. Maintainability can be measured by the
mean time to repair (MTTR) the product. For example, Caterpillar Inc. supports excellent maintainability by supplying spare parts for its equipment anywhere in the world within 48 hours.
Availability, then, is a combination of reliability and maintainability. If a product is
good in both reliability and maintainability, it will be high in availability. The above relationship for availability can be restated in terms of MTBF and MTTR:
MTBF
Availability =
MTBF + MTTR
For example, if a product has an MTBF of eight hours and an MTTR of two hours, its
availability will be 80 percent.
Field service, the last dimension of quality, represents maintenance, repair or replacement of the product after it has been sold. Field service is also called customer service,
sales service, or just service. Field service is related to variables such as promptness, competence, and integrity. The customer expects that problems will be corrected quickly, in a
satisfactory manner, and with a high degree of honesty and courtesy.
The four different dimensions of quality are summarized in Figure 8.1. As can be seen
there, quality is more than just good product design; it extends to quality control of production, quality over the life of the product, and quality of field service after the sale.
FIGURE
8.1
Different types of
Quality of market research
quality.
Quality of design
Quality of concept
Quality of specification
Technology
Quality of conformance
Employees
Management
Customer satisfaction
Reliability
Fitness for use
Availability
Maintainability
Logistical support
Promptness
Field service
Competence
Integrity
YEG)
partthree
Quality
8.3
SERVICE QUALITY
LO8.3.
Distinguish
service quality from
product quality
based on its distinct
measurement.
The definition and measurement of service quality is quite different from product quality.
Service quality has dimensions of the facilitating good, explicit service, and implicit service.
While the facilitating-good quality can be measured by using the dimensions of manufacturing,
explicit (visible) service and implicit (psychological) service require different measurements.
Product measurements can be largely objective while many service measures are perceptual or subjective. For example, product design quality can be measured by the product
features offered, such as the speed of acceleration of an automobile and its normal braking
distance. The conformance quality can be measured by the cost of scrap and rework in the
factory. Services can also have some objective quality features, such as the time to get your
meal at a drive-thru window, or if you can walk without pain after foot surgery. But subjective measures of how the customer perceives the service quality are also common.
The most popular approach to evaluating service quality is called SERVQUAL.'
SERVQUAL assesses customer perceptions about five dimensions of a service:
1. Tangibles. The appearance of the physical facilities, equipment, facilitating goods, and
personnel from the service firm. For example, if a restaurant is dirty, the food does not
look nice, and the employees are disheveled, the tangible quality will be low.
2. Dependability. The ability of the service firm to perform the promised service dependably and accurately without errors. For example, if a restaurant takes a reservation for
7:00 p.m. and the customer is not seated promptly or the waiters bring the wrong meal,
the dependability will be low.
3. Responsiveness. The ability of the service firm to provide service that is prompt and
helpful to the customer. In the restaurant, for example, the meal should be provided in a
timely fashion and customers’ questions about the menu answered.
4. Assurance. The knowledge and courtesy exhibited by employees of the service firm
and their ability to convey trust and confidence. In the restaurant example, does the
server know the menu and is he or she courteous in providing the service?
5. Empathy. The caring, individualized attention that the service firm provides to its customers. The server should show individualized concern for customers and their particular needs.
As can be seen, service quality dimensions are very different from product quality dimensions
and reflect the close interaction the employees have with the customer in service delivery.
SERVQUAL uses a questionnaire to measure these five dimensions. Service quality is
measured as the gap (or mathematical difference) between what the customer expects on each
dimension and what is provided. For example, if the customer does not expect a lot of empathy,
service quality can be perceived to be high even though not much empathy is provided. The
use of gaps as a measure of service quality has been debated vigorously. While some argue
that the perceived level of service provided should measure service quality, others claim that
the gap between what is provided and what is expected is a better measure of service quality.
8.4
QUALITY PLANNING, CONTROL, AND IMPROVEMENT
LO8.4.
Apply the
quality cycle toa
product or service.
The dimensions of product or service quality can be part of a process to manage quality.
The process of quality planning, control, and improvement requires continual interaction
between the customer, the operations function, and other parts of a business. Figure 8.2
"A. Parasuraman, V. A. Zeithaml, and L. L. Berry, “SERVQUAL: A Multiple-ltem Scale for Measuring
Consumer Perceptions of Service Quality,” Journal of Retailing, 64(1), 1988.
Chapter 8 Managing Quality
147) 1)
FIGURE 8.2
The quality cycle.
_ CUSTOMER.
Product
Quality needs
-MARKETING
Interprets customer
needs
;
Works with customer
to design product
Interpretation
of needs
Ga
ENGINEERING
OPERATIONS
Délines desi
f
sd
paca wa
ieee 2
;
4
Prepares specifications |J Specifications
Defines quality
characteristics
ne ete
——_—_—_——_———
By
QUALITY CONTROL
|
Plans
;
Ye
Produces the product
or service
ee a teose cee meta
|
and monitors
quality
eet etea Cross-functional Team — — — —
illustrates how these interactions occur through a quality cycle. The customer needs are
determined, usually through marketing as part of a cross-functional team. These needs
are either expressed directly by the customer or discovered through a process of market
research. Engineering, in conjunction with other functions, designs a product or service to
meet those needs. Quality function deployment is a useful technique for aligning the voice
of the customer (customer needs) with the engineering specifications.
Once the design concept and specifications have been completed, the quality of design
has been established. Operations, as part of the quality team, then produces the product or
service as specified. Operations must continually ensure that production occurs as specified by insisting on quality of conformance. This ordinarily is done through proper training, Supervision, machine maintenance, and operator inspections. The quality cycle must
be a never-ending process of gathering current customer needs, and then designing and
producing to meet those needs. In this way continuous improvement occurs.
Figure 8.3 is a description of the quality cycle for a mass transit system. In this case,
an agency, in place of marketing, interprets customer needs. A planner, working in greater
detail, then determines the design concept and the specifications for service. The operations
function delivers the service, and the quality cycle begins again as the public then restates its
needs or confirms that the present service is satisfactory. The quality cycle should exist in
every organization to ensure that all aspects of quality are planned, controlled, and continually
improved. Feedback from the customer is essential to produce quality products and services.
The implementation of planning, control, and improvement of quality through the quality cycle requires this sequence of steps:
1. Define quality attributes on the basis of customer needs.
2. Decide how to measure each attribute.
TMREE) Part three
Quality
FIGURE 8.3
The quality cycle ina
mass transit system.
County planning
Regional planning
Riders'
|
needs
State transportation agency
Operations office
Planner
;
Scheduler | |
Routes
Schedules
Budgets
Method
Facilities
Equipment
Evaluation | |
|
Inspections
Audits
Surveys
Hearings
. Set quality standards.
. Establish appropriate tests for each standard.
. Find and correct causes of poor quality.
. Continue
WwW
fH
Nn
to make improvements.
Planning for quality must always start with the product attributes. The quality team
determines which attributes are important to customer satisfaction and which are not.
For example, the manufacturer of Leggs panty hose has determined three important
quality attributes for its product: (1) a comfortable fit, (2) an attractive appearance, and
(3) a wear life that is considered reasonable by the customer. It has determined that fit
and appearance depend on the materials used, while wear life depends on materials and
stitch patterns.
A method must then be devised to test and measure quality for each of the product
attributes. For example, the manufacturer of L’eggs has developed a special cross-stretcher
that can test the strength of its product. L’eggs are also inspected visually for fabric defects,
seaming defects, and shade variations.
After deciding on the measurement techniques to use, the quality team sets
standards and tolerances that describe the amount of quality required on each attribute. Standards are set as desired targets. For example, a standard on L’eggs panty
hose is the amount of pressure that the garment must withstand on the cross-stretcher.
Tolerances are stated as (+ quantities), or minimum and maximum acceptable limits
around the standard.
After standards have been set, a testing process should be established. In the case of
L’eggs, this process is based on sampling procedures since it would be far too costly to test
and inspect each of the millions of pairs of panty hose produced each year.
It is not enough simply to inspect the products for defects. As the saying goes, “You
Cannot inspect quality into a product, you must build it in.” Upon discovering defects,
workers should find the underlying causes and correct them. Causes of poor quality can
include improper raw materials, lack of training, unclear procedures, a faulty machine, and
so on. When the causes of poor quality are found and corrected, the production system will
be in control and continuous improvement will be possible.
Chapter 8 Managing Quality
8.5
MISTAKE-PROOFING
LO8.5.
Explain how
mistake-proofing
and the supply chain
are integrated with
quality management
planning.
149 |
Planning, controlling, and improving quality can be augmented by focusing on preventing
errors from occurring in the first place. This requires designing products, services, and
internal procedures that are mistake-proof, working with suppliers to prevent errors, training employees before problems occur, and performing preventive machine maintenance.
Nevertheless, when errors do occur, they need to be corrected quickly and the system itself
changed to prevent errors of the same type from recurring.
The concept of mistake-proofing was developed in the 1960s by Shigeo Shingo, who
worked for Toyota Motors in Japan. It is called poka-yoke (pronounced poe-ka-yoke), which
means “mistake-proofing” in Japanese. The idea of poka-yoke is to design the product and process so that it is impossible for humans to make mistakes when producing, delivering, or using
a product or a service or for mistakes to be easily detected by humans when they do occur.
For example, your microwave will not start if the door is open, and your car will not start
unless your foot is on the brake pedal. These are examples of poka-yokes that prevent consumers from using products in unsafe or unintended ways. In the manufacturing process, parts
should be designed that cannot be assembled backward or left off the product by mistake.
Services should also be designed to avoid errors by both the producer and the customer. For
example, many rides in amusement parks have height requirements. To prevent a child who does
not meet the height requirement from getting on a ride, riders have to walk past a measuring post
at the entrance to the ride, with the height requirement clearly marked. A ride operator can quickly
gauge whether a child meets the height requirement before allowing the child to get on the ride.
If the error cannot be prevented from occurring, it should be made easy to detect. For
example, all of the warning lights on car dashboards try to get drivers to fix little problems
before they become catastrophes! For more examples of mistake-proofing, see Figure 8.4.
©John Grout’s Mistake-Proofing Center (www.mistakeproofing.com)
FIGURE 8.4
Examples of mistake
proofing (poka-yoke).
Courtesy of John
Grout’s MistakeProofing Center (www
.mistakeproofing.com)
Medical gas outlets are designed so
proper valves fit in only one outlet.
Scald protectors close water flow if the water
temperature becomes too hot.
PavelStock/Shutterstock
A car cannot start
in gear. It must
either be in park
or neutral.
These school buses have a wire loop on the
bumper that swings out to ensure no chiidren
are hidden under the hood.
| | 150] Part Three
8.6
Quality
ENSURING QUALITY IN THE SUPPLY CHAIN
Recall that product quality and service quality are defined by the customer and that the
cycle of quality planning, control, and improvement is driven by customer needs. Therefore, the customer input is important not only to the firm but also the entire supply chain in
driving quality performance.
For many products and services, more than 50 percent of the product or service inputs
(materials and other work) are purchased from suppliers. Indeed, with outsourcing,
100 percent of the product or service can be sourced from suppliers. For example, the iPhone
is completely outsourced by Apple, Inc., except for the design, which is done in-house.
In working with suppliers there are several principles that should be followed. First, the
supplier should be involved in the design of the product or service to maximize the prevention of design defects from the start (poka-yoke). Suppliers often can recommend new or
different materials or services that can improve quality or prevent defects from occurring
within production processes that suppliers have responsibility for and control over.
When a product or service is complex, it is important that suppliers maintain very high
levels of quality to ensure that the final product has the required quality. A concept called
rolled yield accounts for the cumulative defect rate observed by the final customer. For
example, suppose a product or service has 100 components or parts and each part has
a yield of 99 percent (1 percent defective). Then the overall yield of the final product is
obtained by multiplying the yields of all the individual parts. Since there are 100 parts with
the same yield (.99), in this case the rolled yield is .99 x .99 x .99 ... x .99 (100 times):
Rolled yield = (.99)!" = .366
As can be seen, the rolled yield is only 36.6 percent for the final product. Therefore, the
quality provided by the suppliers must be much higher to ensure a rolled yield of, say,
99 percent in the final product or service. For a 99 percent rolled yield, supplier quality
must be 99.99 percent for each of the 100 parts.
Supplier management requires more than simply selecting suppliers and measuring
quality compliance. Operations must also manage risk and ensure ongoing process control
by suppliers. Many product recalls in the U.S. are due to suppliers’ quality failures and not
by the selling firm itself. One case was Mattel’s toy recalls in 2007 due to lead paint contamination. These toys were all manufactured by a supplier in China. Yet, the selling firm
is often legally responsible for the costs of the recall. The financial and reputational losses
to Mattel were devastating from this recall of 19 million toys.
Supplier quality management thus requires a system not only to select suppliers but to manage them on an ongoing basis. Managing suppliers is more than a matter of ensuring compliance
to standards through incoming inspection. Oversight of the outsourced process is also required.
Certifying the supplier can help to accomplish this. Supplier certification means the supplier
has control over its processes and can pass an audit by the customer or an independent agency, at
the very least. The audit ensures that there is a quality system in place, including documented procedures, training, and ongoing statistical control of the process to prevent defects from occurring.
In selecting a supplier, the price and the product samples may look good, but is this
enough? A typical item sourced from an overseas supplier encountered the following issues:
*
The samples submitted to the buyer were “jewelry,” that is, carefully selected items that
were not being produced by a reliable production process. The samples were actually
carefully handmade and inspected to ensure their quality.
In the production process, quality was actually being inspected into the product with
only a 60 percent yield at final inspection. Only the good products were included
Chapter 8 Managing Quality
1151] |)
Boeing’s Supplier Rating System
~In 2017, Boeing, the world’s
largest
aerospace
company,
spent
$60
on
billion
supplier receives: Red (Unsatisfactory), Yellow (Improvement
Needed), Bronze (Satisfactory),
Silver (Very Good), and Gold
more
than 13,000 suppliers located
in 58 countries to support
the manufacture of complex,
advanced-technology
_ prod-
ucts ranging from commercial
and
(Exceptional).
The supplier ratings are
available at three levels of
funlovingvolvo/123RF
military aircraft to satel-
lites to weaponry systems to electronic and defense
systems. Suppliers play a critical role in helping Boeing
meet customer
product requirements
and delivery
~ mandates. With so many suppliers involved, how does
Boeing manage
to keep track of supplier performance,
especially those who may not be performing as well as
they should and are in need of help? The answer lies with
its enterprise-wide supplier performance rating system.
Boeing grades its suppliers in three performance
categories—quality performance, delivery performance,
and general performance. Supplier performance in each
category is tracked on a monthly basis and compared
to thresholds that then determine which of five colors a
visibility:
the
site
level,
the
business-group level, and the
company level. Consistent with the concept of 360-degree
feedback, suppliers get to see their performance ratings and learn from them. Suppliers can access detailed
reports at the incident level (e.g., an incident involving a
nonconforming shipment of supplier products).
Boeing uses its supplier rating performance system
not only to encourage and help its suppliers improve but
also to help various Boeing units select and partner with
capable suppliers. Exceptionally performing suppliers are
eligible for the Supplier of the Year award from Boeing.
Source: Kirsten Parks and Timothy Connor, “The Way
to Engage,” Quality Progress 44, no. 1 (April 201 1),
pp. 20-27; www.boeing.mediaroom.com, 2019.
in the sample. This means the production process was not capable of producing consistently good quality.
¢
Small lots were not feasible since production equipment had been set up for long runs,
with questionable quality discovered later, after production.
¢
The production cycle time was long and variable, resulting in large inventories.
This example indicates why certified suppliers are necessary to ensure that suppliers have
a good quality system in place for ongoing production of products or services. Without
this, suppliers may merely provide good samples at a low price, only to produce low quality later. Read the Operations Leader box about Boeing Company and how its supplier
rating system ensures quality from suppliers.
8.7
QUALITY, COST OF QUALITY, AND FINANCIAL PERFORMANCE
LO8.6.
Attribute
how cost of quality
is related to financial
performance.
Quality and financial performance are intimately related. First, we consider the relationship between quality and cost. A powerful idea in the area of quality is to calculate the
cost of quality, which includes prevention, appraisal, internal failure, and external failure
categories. All these, except prevention, are costs of not doing things right the first time.
When a cost is assigned to poor quality, it can be managed and controlled like any other
cost. Since managers speak the language of money, putting quality in cost terms provides a
powerful means of motivation, communication, and control.
I} |152] Part Three
Quality
Most companies have no idea how much they spend to manage quality. Those that have
found that the cost of quality is about 30 percent of sales, and typically ranges from 20 to
40 percent. Since these figures are two or three times greater than profit margins in many
companies, a reduction in the cost of quality can lead to a significant improvement in
profit. The best-managed companies have been able to reduce their costs of quality from
30 percent of sales to as little as 5 percent in a few years. This has been done while improving the quality of their products or services.
The cost of quality may be divided into two components: control costs and failure
costs. A full listing of these costs is given in Table 8.1.
Control costs are related to activities that remove defects from the production stream.
This can be done in two ways: by prevention and by appraisal. Prevention costs include
activities such as quality planning, new-product reviews, training, and engineering analysis.
TABLE 8.1
Prevention Costs
Costs of Quality
Quality planning
Source: Adapted from J. M.
Juran and A. B. Godfrey, eds.,
Juran’s Quality Handbook, Sth
ed. (New York: McGraw-Hill,
1999).
New product review
Costs of preparing an overall plan, numerous
specialized plans, quality manuals, procedures.
Review or prepare quality specifications for new
products, evaluation of new designs, preparation
of tests and experimental programs, evaluation
of vendors, marketing studies to determine
Training
Process planning
Quality data
Improvement projects
Control
Costs
LIRA
Sr cE nena
ee eee ACE
customers’ quality requirements.
Developing and conducting training programs.
Designing and developing process control devices.
Collecting data, data analysis, reporting.
Planned failure investigations aimed at chronic
quality problems.
Appraisal Costs
Incoming materials
inspection
Process inspection
Final goods inspection
Quality laboratories
The cost of determining quality of incoming
raw materials.
All tests, sampling procedures, and inspections
done while the product is being made.
All inspections or tests conducted on the finished
product in the plant or the field.
The cost of operating laboratories to inspect
materials at all stages of production.
Internal Failure Costs
Scrap
The cost of labor and materials for product that
cannot be used or sold.
Rework
The cost of redoing product that can be made
Downgrading
Failure
Retest
Costs
Downtime
to conform.
Product that must be sold at less than full value
due to quality problems.
Cost of inspection and tests after rework.
Idle facilities and people due to quality failures.
External Failure Costs
Warranty
Returned merchandise
Complaints
Allowances
The cost of refunds, repairing, or replacing
products on warranty.
Merchandise that is returned to the seller.
The cost of settling customer complaints due to
poor quality.
The cost of concessions made to customers due
to substandard quality.
Chapter 8 Managing Quality
153] |)
These activities mostly occur before production and are aimed at preventing defects before
they occur. The other category of control costs includes appraisal or inspection aimed at
eliminating defects after they occur but before the products or services reach the customer.
Failure costs are incurred either during the production process (internal) or after the
product is shipped (external). Internal failure costs include items such as scrap, rework,
quality downgrading, and machine downtime. External failure costs include warranty
charges, returned goods, and allowances. The potential loss of customers’ confidence in
the product or service is perhaps the greatest external cost.
The cost of quality can be a confusing term. Three of the four costs can be called the
cost of nonconformance or the costs of poor quality (appraisal, internal failure, or external failure). Prevention is the cost of achieving good quality. Figure 8.5 reveals how the
four cost of quality categories relate to one another. As prevention costs and appraisal
costs increase, internal failure and external failure decreases. A trade-off, therefore, exists
between the control costs categories (i.e., prevention and appraisal) and the failure costs
categories (i.e., internal failure and external failure). Figure 8.5 appears to suggest that
there is an optimal quality level where control and failure costs meet that is not 100 percent good quality. However, management stresses continuous improvement by finding
ways to shift the curve for prevention and appraisal costs to the right over time, allowing
quality to improve through more efficient prevention and appraisal activities.
Many companies find that by investing in prevention activities such as training, process
planning, and new-product review, they avoid costs that occur later in production (appraisal,
internal failure) or after production (external failure). Prevention is a tremendous leverage factor. Investing one dollar in prevention activities generates more than one dollar in appraisal,
internal failure, and external cost savings. Those savings flow directly to the bottom line,
which in many companies will more than double profits.
A construction company investigated its cost of poor quality and found a total of
72 instances of nonconformance on one highway construction project. Those instances
could be classified into the following types of preventable errors.
¢
Design problems that caused the reworking of a portion of the highway several times.
¢
Noncompliance by a cement contractor that required repair of the poured concrete.
¢
Subcontractor problems that resulted in failure to deliver by some subcontractors.
The company found that it could save considerable costs in future projects by seeking to
prevent these errors before they occurred.
FIGURE
8.5
Cost of quality
categories and
Cost of Quality Categories and Trade-offs
Cost/unit
trade-offs.
Internal failure + External
failure costs
Prevention &
appraisal costs
100%
100%
defective
good
P| | 154| Part Three
Quality
The cost of quality can be a powerful incentive for quality improvement when it is used
properly. It focuses management attention on waste due to excess failures or high control
costs. It also provides a quantitative basis for monitoring progress in reducing quality costs.
Quality improvement can also dramatically increase revenues through either a more
consistent product or new products or services that better meet customer needs. Improving
customer satisfaction can be a powerful driver of revenue and market share when customers receive a product or service that they really like.
Improving quality effects profitability and return on investment through both increasing
revenues and reducing costs. Quality has a powerful effect on the top line, margins, and
ultimately the bottom line.
8.8
QUALITY PIONEERS
LO8.7.
Recall the two
Many approaches to managing quality have been advocated.
Deming, Juran, Crosby,
key quality pioneers
Feigenbaum, Shewhart, and Ishikawa, to name only a few, are pioneers who have shaped
and their main ideas.
modern approaches to quality management. Of these, Deming and Juran are most noted
for teaching quality to the Japanese and restoring attention to quality by American and
European companies.
W. Edwards
W. Edwards Deming emphasized the role that management should take in quality improvement. Deming defined quality as continuous improvement of a stable system. This definition emphasizes two things. First, all systems (administrative, design, production, and sales)
must be stable in a statistical sense. This requires that measurements be taken of quality
attributes and monitored over time. If these measurements have a constant variance around a
constant average, the system is stable. The second aspect of Deming’s definition is continuous improvement of the various systems to reduce variation and better meet customer needs.
Deming expressed his philosophy of quality in his famous /4 points, which are listed in
Table 8.2. He stressed that top executives should manage for the long run and not sacrifice
quality for short-run profits. Deming argued that excessive attention to quarterly profit
reports and short-run objectives distracts top management from focusing on customer service and long-run quality improvement. He also argued, as others do, that management
should cease its dependence on mass inspection to achieve quality and stress prevention
of defects instead. Deming suggested that this should be accomplished by training of all
employees, good supervision, and use of statistical procedures.
Deming went on to exhort management to break down barriers between departments
and encourage people to work together to produce quality products and services. He
thought that many of the work standards, individual performance pay systems, and quotas
that companies use get in the way of cooperation among individuals and departments and
thus impede quality improvement.
Deming was a strong advocate of applying statistics to stabilize and improve processes.
Quality cannot be improved by trying harder. Workers and managers must have the proper
tools to identify causes of variation, to control variation, and to reduce variation in the product.
Deming and the other quality pioneers are advocates of the idea that most quality problems are caused by poor systems, not by the workers. They argue that quality problems
should not be blamed on the workers; rather, management must change the system to
improve quality. All levels of management must accept responsibility for quality.
Deming
Joseph Juran
Juran originated the idea of the quality trilogy: planning, control, and improvement
of quality. In the planning area he suggested that companies should identify the major
Chapter 8 Managing Quality
TABLE 8.2
Deming’s 14
Management
Principles
Requirements for a business whose management plans to remain competitive in providing
~ goods and services that will have a market.
Wes Create constancy of purpose toward
7.
Focus management and supervisors
on leadership of their employees to
8.
Drive out fear. Don’t blame employees
for “systems problems.” Encourage
improvement of products and services
Source: Adapted from W.
Edwards Deming, Out of the
Crisis (Cambridge, MA: MIT
Center for Advanced
Engineering Study, 1986).
with the aim of being competitive and
staying in business for long-run, rather
than short-run, profits.
Adopt the new philosophy by refusing
to allow commonly accepted levels of
mistakes, defects, delays, and errors.
Accept the need for change.
Cease dependence on mass inspection. Rely instead on building quality
into the product in the first place and
on Statistical means for controlling and
improving quality.
End the practice of awarding business
on the basis of price tag alone. Instead,
minimize total cost. Reduce the number of
help them do a better job.
effective two-way communications.
Eliminate management by control.
9. Break down barriers between departments. Encourage teamwork among
different areas such as research,
10.
processes.
;
Improve constantly, and forever,
systems of production to improve quality and productivity and thus constantly
reduce costs.
Institute training and retraining for all
employees.
©Roger Schroeder
design, manufacturing, and sales.
Eliminate programs, exhortations, and
slogans that ask for new levels of productivity without providing better methods.
11.
Eliminate arbitrary quotas, work
standards, and objectives that interfere
with quality. Instead, substitute leadership and continuous improvement of
work processes.
suppliers by eliminating those who cannot
provide evidence of statistical control of
Textbook author
Schroeder (left) learns
from Joseph Juran ata
conference.
1155] |)
12.
13.
14.
Remove barriers (poor systems and
poor management) that rob people of
pride in their work.
Encourage lifelong education and
self-improvement of all employees.
Put everyone to work on implementing these 14 points.
business
goals,
customers,
and
products
required. New products should be introduced only after they are carefully tested
and when they meet a verified customer
need. He also suggested that much of quality improvement requires careful planning
to ensure that the most important quality
problems are attacked first—“‘the vital few.”
Juran stressed control of quality through
use of the statistical methods covered in the
next chapter. He argued that management
should institute the procedures and methods
needed to ensure quality and then work to
keep the system continuously in control. Like Deming, Juran believed strongly in the statistical approach to quality as a way of achieving process control.
The third leg of the quality trilogy is improvement. Juran suggested both breakthrough
improvement and continuous improvement of processes. He argued that this could be done
once the system was brought under statistical control. Juran also believed that training and
involvement of all employees was necessary to ensure continuous quality improvement.
While the specific details of quality improvement may vary between Deming and Juran,
they have much in common. Some of their common ideas and those of other quality pioneers are shown in Table 8.3. As can be seen from the table, there is a common
running through quality thinking.
thread
WWREG) Part three
Quality
TABLE 8.3
Changing Quality
Assumptions
From
Reactive
To
Proactive
Inspection
Prevention
Meet the specifications
Product-oriented
Blame placing
Quality versus schedule
Cost or quality
Operations only
Predominantly blue-collar caused
Defects should be hidden
Quality department has quality problems
Continuous improvement
Subordinated to management team
General managers not evaluated on quality
Process-oriented
Problem solving
Quality and schedule
Cost and quality
Marketing, engineering, and operations
Predominantly white-collar caused
Defects should be highlighted
Purchasing, R&D, marketing, and operations
have quality problems
Part of management team
Quality performance part of generalmanager review
Quality costs more
Quality costs less
Quality is technical
Schedule first
Quality is managerial
Quality first
8.9
ISO 9000 STANDARDS
LO8.8.
Compare
9000 standards and
ISO 9000, established in 1987, is one of the major approaches that companies are using to
ensure quality today. ISO 9000 is one of several sets of standards developed by ISO (International Organization for Standardization). ISO, an international body consisting of mem-
the Baldrige Award
bers from 162 countries, is the largest developer of international standards. Besides ISO
criteria.
9000, other standards have been developed for environmental management
and contrast |SO
(ISO 14000),
social responsibility (ISO 26000), supply chain security (SO 28000), and risk management (ISO 31000).
ISO 9000 was originally oriented toward compliance, or what we have termed confor-
mance quality. Customer needs were not included in the original ISO 9000 standard—you
could make any product you liked, even if it did not sell, as long as the company had a
quality system to ensure that it could make what it said it could. The ISO 9000 standard has
been revised to include customer requirements, continuous improvement, and management
leadership to ensure that quality meets customer needs, not just conformance to specifications. The standard continues to be updated.
The ISO 9000 standards are meant to describe how a company should develop a system
and processes for ensuring quality, but the benefits go beyond quality. Firms adopting the
ISO 9000 standards, compared to those that have not adopted ISO 9000 standards, are more
likely to survive, to experience greater growth in sales and employment, to pay employees better wages, and to have fewer employment-related injuries. The ISO 9000 standards
ideally apply to large and small companies and to simple and complex products. ISO
9000 standards also apply to services and to software development. Read the Operations
Leader box on Physicians’ Clinic of Iowa and its efforts to become ISO certified.
The ISO 9000 standards specify that a company must have a quality system in place,
including procedures, policies, and training, to provide quality that consistently meets
customer requirements. A quality manual and careful record keeping are usually required
as part of the documentation. ISO 9000 requires that the company have process flowcharts, operator instructions, inspection and testing methods, job descriptions, organization charts, measures of customer satisfaction, and continuous improvement processes.
Chapter 8 Managing Quality
OPERATIONS LEADER
157] |]
c
Physicians’ Clinic of lowa Gets ISO Certified
Physicians’ Clinic of lowa (PCI), one of the largest private,
multispecialty physician groups
in lowa, provides medi-
» PHYSICIANS? CUNIC
cal and surgical care
in approximately 20
different
specialties
of Iowa, PC.
Q
Togetherin health.
7
and
_
subspecialties.
Founded in 1997 in
Cedar Rapids, lowa, PCI currently has more than 80
board-certified physicians, surgeons, and other providers.
On November
10, 2003, PCI became
ISO 9000 certi-
fied. Certification did not happen overnight. The journey
from when PCI leadership decided to seek certification
to becoming certified took approximately 2.5 years, while
= “McGraw-Hill Education/Mark Dierker
consuming about $108,000 in training and auditing. In the
first year following certification, PCI recouped more than
its initial investment with cost savings of over $200,000,
Since then, PCI has continued to seek recertification.
What
motivated
PCI to pursue
ISO
9000
t© support medical professionals in their practices while
improving their efficiencies and delivery of care. As a
certifica.
Consequence,
PCI
identified
how
to exert
better con-
tion was the recognition that it needed to operate as a_—*trOl over documents, records, and nonconformity, how
single organization, despite the diversity of specialties
tO audit and review key processes with an eye toward
and being in five different locations. A single organization would
allow PCI to have common
goals, processes,
improvement; and how to initiate corrective and preventive actions. PCI is convinced that it is not unique. The
health care sector, private or public, domestic or interna-
and procedures that would be compliant with mandatory
requirements by government agencies, as well as volun-
tional, can benefit from ISO 9000 certification.
tary requirements
Source: Robert Burney, James Levett, and Paula Dolan,
of accreditation organizations.
ISO 9000 offered PCI the answer to creating
prehensive,
patient-centric quality management
acomsystem
—“|SQ-Lating the Problem,” Quality Progress 42, no. 1
(January 2009), pp. 36-40; www.pcofiowa.com, 2020.
It is also expected that employees will be trained in the procedures and actually follow
them in practice. To ensure compliance, certified ISO 9000 registrars audit the organization and determine whether the system in use conforms to the formal system description
in their documentation. If no discrepancies are found, the registrar, who is external to the
company, certifies the company’s plant or facility. The product itself is not certified as
having high quality; only the process for making the product is certified. The ISO 9000
certification must be renewed periodically via return audits by a registrar.
ISO 9000 has had a major impact on worldwide quality practice. Many companies
are requiring ISO 9000 certification of suppliers as a condition for doing business. The
European Community has adopted ISO 9000 as a standard for selling in its markets and
compliance with [SO 9000 is required by some European customers. ISO 9000 certification has caught on around the world, and many countries and companies are also requiring
ISO 9000 certification of their suppliers.
ISO 9000 does not provide a complete quality system, because it does not address competitive strategy, information systems, and business results. A company can be making a
product that satisfies the customer for a shrinking market and going out of business and still
be ISO 9000 certified. Nevertheless, ISO 9000 is a good first step that addresses the fundamental processes needed to ensure a quality product and high levels of customer satisfaction.
TREE] Part three Quality
8.10
MALCOLM
BALDRIGE AWARD
The U.S. Congress established the Malcolm Baldrige National Quality Award in 1987 to
promote better quality management practices and improved quality results by American
industry. The criteria for the award have gained wide acceptance and are a de facto standard for “best quality practice” in the United States. Baldrige has requirements similar to those of ISO 9000, but strategy, information systems, and
business results are also required.
Each year the Baldrige Award is given to at most three organizations
in each of six categories: manufacturing, service, small business, health
care, education, and nonprofit. A few of the past winners are Milliken
& Co., Federal Express, Sunny Fresh Foods, Ritz-Carlton, 3M, IBM, the
University of Wisconsin at Stout, the Pearl River School District, and the
city of Irving, Texas. To win this award these organizations exhibited high
levels of quality management practice and performance excellence as
indicated by the Baldrige criteria. During the first twenty years, Baldrige
winners were mostly large manufacturing and service companies. Since
then, many of the winners have been nonprofits, health care, education,
and small businesses.
The Baldrige criteria recognize quality efforts that have senior management leadership, business results, employee involvement, control of internal
processes, strong customer satisfaction, and so on. The specific categories
of quality evaluated by the Baldrige examiners are listed in Table 8.4. These
seven Baldrige categories are judged from a self-evaluation report prepared
by each applicant for the award and a site visit by Baldrige examiners for
Malcolm Baldrige National Quality Award
applicants that pass an initial screening of the self-evaluation report.
Source: United States Department of Commerce
The.Baldrige criteria have 1000 points total, which are allocated among
the seven categories shown in Table 8.4. These criteria have evolved from
quality management to the more general “performance excellence” due to a
realization that the criteria are about achieving success of any enterprise.
The first category, leadership, is based on senior management commitment, vision,
active involvement by all managers in the business, and the extent to which quality values
have permeated the entire organization. It also includes societal responsibilities, ethical
behavior, and community involvement.
The second category addresses strategy. A company cannot provide quality products
and services unless it has a coherent business strategy that defines the target markets, competition, and how the company plans to compete (e.g., low cost or differentiation). Successful applicants have established high-level goals and strategic plans that are specific
and have been implemented.
Category three is customers. Winning companies collect customer data from a variety
of sources, including focus groups, market research surveys, and one-on-one contacts.
The information should be acted on and used to direct the company toward customer
needs and satisfied customers. Baldrige winners strive to delight their customers, not
just minimally satisfy them. They often exceed customer expectations and anticipate
customer needs.
Category four is measurement, analysis, and knowledge management. This category
includes decision making based on hard data, sometimes called “management by fact.”
The company’s database should be accessible to employees and should have comprehensive information on suppliers, internal processes, and customers. The information system
must be integrated with processes and used for decision making in the company.
Chapter 8 Managing Quality
TABLE 8.4
Seem
Se
Categories and Items
Framework,
1
;
Baldrige Execellence
2017-2018
Source: U.S. Department of
Commerce, 2018.
led
1.2.
5
j
?
ak.
85
85
Voice of the Customer
Customer Engagement
Measurement, Analysis, and Improvement of
Organizational Performance
4.2
Information and Knowledge Management
Workforce
90
85
Workforce Environment
Workforce Engagement
Operations
85
Work Processes
Operational Effectiveness
Results
450
Tat
7.2
7.3.
Product and Process Results
Customer Results
Workforce Results
7.4
Leadership and Governance Results
HAS
es
Strategy Development
Strategy Implementation
4.1
6.1
6.2.
7
:
ee
Measurement, Analysis, and Knowledge Management
Sal
5.2.
6
ieee
.
Customers
3.1
3.2
4
ah
Senior Leadership
Governance and Social Responsibilities
Strategy
Ja
2.2
3
GaTais
Leadership
:
2
ASia
1159] |
Financial and Market Results
TOTAL POINTS
1000
The fifth category is workforce, a very broad area. This category includes employee
involvement, continuous education and training, teamwork, and decision making by the
workers. There is also an evaluation of motivation, rewards, recognition, and leadership
development. Past Baldrige winners have been strong advocates of their workforce as the
basis for all improvement efforts.
Operations, the sixth category, includes process definition, documentation, statistical
process control, and the tools of process improvement. The best companies have in-depth
understanding of their processes, and have integrated processes across functions, departments, and the supply chain.
The seventh Baldrige category is results, and it accounts for the greatest number of points.
This category includes product and process outcomes, customer-focused outcomes, financial
and market outcomes, workforce outcomes, and leadership and governance outcomes. Standard quality measures such as percent defective product, customer returns, and on-time deliveries are considered, along with profitability, return on investment, and market share. Successful
companies can demonstrate improvement trends over time, not stellar results in one year only.
These seven categories represent a comprehensive framework for quality management and performance excellence in general. No standard approach is required to win
the Baldrige Award. Each organization is free to choose its own specific techniques and
approaches within the overall goals and criteria described above, and there is enormous
| | 160,
Part Three
Quality
variety among the past winners. This is a strength of the Baldrige Award: It allows flexibility
in defining performance excellence by individual companies and nonprofit organizations.
Many companies use the Baldrige criteria as a vehicle for internal assessment of their
quality and performance improvement systems. They train their own managers to make
Baldrige assessments of other divisions. The objective is not to win an award but to diagnose strengths and weaknesses of their management system.
8.11
LO8.9.
WHY SOME QUALITY IMPROVEMENT
Articulate
some key barriers
to successful quality
improvement efforts.
EFFORTS FAIL
Quality improvement is a proven strategy that has yielded significant financial benefits for
many companies. Yet quality efforts have also failed or yielded marginal results in other
companies. Some studies have indicated that only one-third of companies had obtained
significant results from their quality improvement programs, one-third had achieved moderate results, and one-third were dissatisfied with the results. Why has this happened, and
why have results been so inconsistent?
It may matter little which approach a company uses to improve quality that makes the
difference; it’s the implementation process. To improve quality, a company must change its
values, culture, and management philosophy, which is not easy.
One of the main reasons for quality implementation failure is the lack of leadership by
middle and top management. Management should do much more than pay lip service to
quality improvement efforts. They have the responsibility to institute a known system for
quality improvement such as ISO 9000 or Baldridge. They must provide the resources for
training, improvement specialists, and time for employees to participate in improvement.
They need to track progress themselves and offer rewards for improvement. It is only with
the full attention of management that quality will improve.
Some managers instinctively blame the employees when there is a quality failure. It
doesn’t occur to them to look at the system and rules that employees work within. Only managers can change the underlying system causes of the quality problem, not the employees.
Managers who believe in trade-offs can also take quality improvement efforts off track.
To them, consistent quality cannot be achieved without sacrificing the schedule or cost.
When they must decide to ship the product or fix a quality problem, these managers will
ship the product and fix the problem later. They also believe that it is too expensive to produce good quality. These managers don’t realize that consistent quality and prevention can
save money and drive better results in schedule, flexibility, and delivery.
Managers sometimes interfere with teamwork, which is often essential to achieying good quality. Either they do not really delegate decision making to the team or they
continue to reward individual performance over team performance. The reward system is
ingrained in the organization and is one of the most difficult things to change.
Finally, many quality efforts fail because suppliers are not certified for a functioning quality
system. Suppliers attempt to inspect quality into the product rather than develop a preventive
approach to quality system design. As a result, the customer cannot rely on consistent quality.
Thus, producing quality requires a systems approach to management, which must be
driven by customer needs. This approach conflicts with the philosophies and values of
some companies. Quality improvement therefore requires deep cultural change. Executives
have to lead by example to make the transformation.
The only way to institute successful quality improvement is through extensive education
of all employees and constant leadership at all levels of management. With this approach,
a quality system can be introduced into any service or manufacturing organization along
with its supply chain, and the financial, human, and market results will be iimpressive.
Chapter 8 Managing Quality
8.12
1161 ||)
KEY POINTS AND TERMS
This chapter concerns the management for quality. The key points include:
Key Terms
*
Quality is defined as meeting or exceeding customer requirements now and in the
future. The four dimensions of product quality are quality of design, quality of conformance, the “‘abilities,” and field service.
¢
Five dimensions define service quality: tangibles, dependabililty, responsiveness, assurance, and empathy. These measures can be obtained by surveying customers.
¢
There is a cycle of product or service quality—from understanding customer needs
through quality of design, production, and use by the customer. This cycle is controlled
by specifying quality attributes, determining how to measure each attribute, setting
quality standards, establishing a testing program, and finding and correcting causes of
poor quality. Continuous improvement of the system through prevention of defects is
the preferred approach.
¢
Supplier certification is a good way to ensure that suppliers have a quality system in
place to prevent defects from occurring.
*
Quality can both improve revenues and reduce costs. The cost of quality measures the
lack of conformance to customer requirements. Quality costs can be divided into control costs and failure costs. Control costs are due to prevention or appraisal. Failure
costs may be due to internal or external failures.
¢
Two quality pioneers, Deming and Juran, have taken somewhat different approaches
to quality but also have much in common. Deming argued that management needs
to change for quality to improve. He also advocated the aggressive use of statistical quality control techniques. Juran advocated the quality trilogy: planning, control,
and improvement.
¢
ISO 9000 process certification is based on a set of standards that address meeting
customer requirements and continuous improvement. It requires well-defined and
documented procedures along with trained operators who implement them to ensure a
quality process, a consistent quality product, and improvement.
¢
The Baldrige Award recognizes companies that achieve a total quality system as defined
and measured by the Baldrige criteria. The criteria have become the common definition
for excellence in quality and performance excellence for U.S. organizations.
*
Quality improvement efforts fail when management does not lead by example and does
not take a systems approach driven by customer needs.
Quality management 143
Fitness for use 143
Customer satisfaction 143
Quality of design 144
Quality of conformance 144
Availability 145
Reliability 145
Maintainability 145
Field service 145
Service quality 146
SERVQUAL 146
Tangibles 146
Dependability 146
Responsiveness 146
Assurance 146
Empathy 146
Quality cycle 147
Standards 148
Tolerances 148
Mistake-proofing 149
Poka-yoke 149
Supplier certification
Cost of quality 151
Control costs 152
Failure costs 152
Deming 154
Juran 154
Quality trilogy 154
ISO 9000 156
Baldrige Award 158
150
by]| 162, Part Three
Quality
LEARNING
ENRICHMENT
ISO 9001 and ISO 9000 Certification
https://youtu.be/IILXyEWOI18w
Video
2:14
(for self-study or
MidwayUSA Baldrige Award Winner
Video
5:26
https://youtu.be/DMRDpm7dGeg
UARletse)eFC TINS)
Video
How Health Care Uses Baldrige
https://youtu.be/vzSxdflfO9w
4:08
Cost of Quality: What Is It?
https://youtu.be/vkeoBe GEDyw
Video
2:48
Mistake Proofing (Poka-yoke)
http://asq.org/learn-about-quality/process-analysis-tools/
overview/mistake-proofing.html
Website
Discussion Questions
il How can quality be measured for the following?
b. Automobile repair
Which of the four manufacturing quality dimensions are
most likely to improve revenue? Which ones are related
more to cost reduction or to both revenue and cost?
c. Ballpoint pens
The following costs have been recorded:
a. Phone service
. What are the differences between quality of design and
quality of conformance?
x
Incoming materials inspection
$20,000
Training of personnel
40,000
. Product A has an MTBF of 30 hours and an MTTR
of
Warranty
45,000
5 hours. Product B has an MTBF of 40 hours and an
Process planning
15,000
MTTR of 2 hours.
Scrap
13,000
a. Which product has higher reliability?
Quality laboratory
30,000
b. Which product has better maintainability?
Rework
25,000
Allowances
10,000
Complaints
14,000
c. Which product has greater availability?
. Recall a bad service experience that you recently had.
What led to the service failure? Consider such dimensions as tangibles, reliability, responsiveness, assurance,
or empathy.
. Suppose you manufacture 10,000 wooden pencils per
day. What would be an appropriate quality planning and
control system for this product? Consider such factors as
product attributes, measures of quality, tests, and so forth.
. Can you name products and services that, in your
opinion, have relatively poor quality? Relatively high
quality? Are companies that provide better quality more
successful? How can you tell?
. How can supplier certification contribute to quality
products or services? Should all suppliers be certified?
Why or why not?
. What products have recently been recalled? Why? Check
the U.S. Consumer Product Safety Commission website.
What are the costs of prevention, appraisal, external
failure, and internal failure?
. Which of Deming’s 14 points do you agree with and
which ones do you disagree with?
. Contrast and compare the Deming and Juran approaches
to quality improvement. How different are these two
approaches?
. Why is ISO 9000 considered a first step or a basic
approach to quality?
. Critique the seven categories used by the Baldrige Award.
Are there some items that you think are missing?
. Compare the Baldrige-based approach to the use of
ISO 9000.
. What are some of the barriers to a successful quality
implementation effort?
Ger
ASP SIME
Quality Control and
— Improvement
Differentiate lean and Six Sigma.
In 1924, Walter A. Shewhart of the Bell Telephone Labs developed a statistical quality
control chart. Two others from the Bell Labs, H. F. Dodge and H. G. Romig, further devel-
oped the theory of statistical quality control in the 1930s. But little was done in industry
until World War II in the early 1940s. The war created demand for huge quantities of military goods from industry. The military required that industry adopt the new methods of
statistical quality control to help ensure that the goods it ordered would meet government
standards. As a result, statistical methods for control of quality were widely adopted by
industry. In later years, however, these methods were abandoned, only to be rediscovered
in the 1980s as a valid way to ensure quality products and services.
Service industries have been slower to adopt the methods of statistical quality control. As a result, there is a tremendous opportunity to use these methods in service firms.
Organizations in banking, health care services, airlines, and government offices find that
quality control tools are useful for controlling and improving quality.
ae
°
°
9°
9
o
°
°
°
°
o
°
8
o
o
°°o
o
°
-
=
°
(Oo
o
oa
°
o
°
°
°
°
>
9
°
2]
°
°
o
~/
°
on
x
°
°
P|| 164| Part Three
Quality
The American Society for Quality (ASQ) serves as a society for quality professionals
across all industries. While the initial emphasis was on statistical quality control methods,
the focus has broadened to include understanding customer needs, quality management
systems, and continuous improvement. ASQ helps quality professionals stay up-to-date on
tools, techniques, and most importantly, current thinking about the importance of quality
in industry. Read the Operations Leader box on Milliken & Company and how it uses statistical methods for quality control.
In applying quality control and improvement methods, we recognize that an organization consists of many interrelated processes that need to be controlled to produce quality products and services. It follows that quality control and continuous improvement are
highly cross-functional in nature and require the participation and support of the entire
organization.
All business students are likely to encounter quality control and improvement in their
career. Accounting will find that quality-related costs are reduced by 20 to 30 percent of
sales when quality in production processes is well-controlled. In human resources, implementation of quality programs requires in-depth workforce training on managing improvement projects, statistical training, and learning to inspect their own output. Marketing
will see a reduction in the number of defects produced, fewer customer complaints, and
increased sales. Finally, finance will see the results of these efforts on the bottom line.
9.1
LO9.1
DESIGN OF QUALITY CONTROL SYSTEMS
Describe the
steps in designing
a quality control
system.
The goal of quality control is to stabilize and maintain transformation processes (or more
generally “processes’”) to produce consistent output. All organizations consist of many
interrelated processes that need to be controlled to produce quality outputs. Continuous
improvement can occur only after a process is stabilized and under statistical control.
Therefore, we treat quality control in the first part of this chapter before discussing continuous improvement.
The design of quality control systems begins with process definition. Process definition
requires a clear understanding of the transformation process. This can be a manufacturing
process, a service delivery process, or an administrative process. Questions to answer in
defining the transformation process include: what are the outcomes of the process, what
are the process steps (or activities) to produce this outcome, and which process steps are
associated with product or service quality attributes?
Of course, a process is typically composed of many subprocesses, each having its own
intermediate product or service. A process can therefore be an individual machine, a group
of machines, or any of the many clerical and administrative processes that exist in the
organization. Each of these processes has its own internal customers and its own products or services that are produced. The internal customer is the next process (or processes)
downstream that receives the work output. For example, the customer of the design department is the machine shop that makes the parts. The customer of the machine shop is the
assembly department that uses the parts. When a large production system is broken down
into many smaller systems or processes, quality can be defined and controlled at each point
along the way.
After identifying the processes that need to be controlled, critical control points can
be chosen where inspection or measurement should take place. The types of measurement or tests required and the amount of inspection required at each of these points
should be determined. Finally, management should decide who will do the inspection,
the workforce itself or separate inspectors. Usually operator inspection by the worker
Chapter 9
Quality Control and Improvement
165 ||
OPERA
Implementing SPC at Milliken & Company
Milliken
South
pets,
&
Company,
Carolina,
and
is
a
headquartered
in Spartanburg,
global
in
fire retardant
leader
fabric. An
textiles,
innovator
car-
at heart,
Milliken
& Company
has long led the way
for “knowledge-based”
i investment,
| employing over
100
PhDs,
and
has
accumulated
more
than
5,000 patents worldwide. Today, Milliken & Company
has 39 manufacturing facilities located in the United
States, Europe, and China and employs approximately
7,000 associates globally.
Milliken & Company has a long history of deploying
statistical process control (SPC) to help monitor process
quality performance. In one example, SPC was used for
an out-of-control process. To solve this problem, several
people from different departments joined together. After
analysis by the team, the process was brought under
control and the process capability improved to where the
process could consistently meet specifications.
my
Source: Photo by Charlie Rahm, USDA Natural
Resources Conservation Service
Milliken & Company
National Quality Award
has won
the Malcolm
Baldrige
and many other quality awards.
The focus on quality helps Milliken & Company compete
successfully against lower-cost global suppliers.
Source: www.milliken.com,
2020.
at that point in the process is preferred because it places responsibility on those who
make the product or service. Once these decisions are made, it is possible to design a
complete system of quality control, which allows continuous improvement of a stable
system.
1. The first step in designing a quality control system is to identify the critical points in
each process where inspection and testing are needed. The guidelines for doing this are
as follows:
*
Ensure that incoming raw materials or purchased services meet specifications. Ideally, incoming inspection can be eliminated, or reduced to sampling, by certifying
the supplier. Supplier certification is normally granted to suppliers that have demonstrated they use statistical process control (SPC) and other methods to achieve
consistent quality performance. In this case, the products or services of the supplier
can be used with confidence by the customer.
¢
Inspect products or services during the production process. As a general rule, the
product or service should be inspected by operators before irreversible operations
take place or before a great deal of value is added to the product. In these cases, the
cost of inspection is less than the cost of adding more value to the product. A precise
determination of where in the process the product or service should be inspected
should be made from the process flowchart.
¢
The third critical inspection point is the finished product or service. In manufacturing, final products are frequently inspected or tested before shipping or
1] | 166 | Part
Three
Quality
before the product is placed in inventory. At an automobile assembly plant, for
example, a random sample of cars is taken directly off the assembly line and thoroughly inspected for appearance and function. The defects are noted and fed back
to assembly-line personnel so that they can correct the underlying causes. The
defects are also used to compute a quality score for comparison among assembly
plants.
It is usually far better to prevent defects from occurring than to inspect and correct
defects after production. Nevertheless, some measurement via sampling inspection is
necessary to maintain processes in a continuous state of statistical control and to facilitate improvement. Thus, inspection cannot be eliminated, but it can be reduced by a
vigorous process of prevention.
2. The second step in designing a quality control system is to decide on the type of measurement to be used at each inspection point. There are generally two options: measurement based on variables or attributes.
Variables measurement utilizes a continuous scale for product and service characteristics such as time, length, height, and weight. Examples of variables measurement
are the dimensions of parts, the viscosity of liquids, and the time it takes to answer a
customer service call.
Attribute measurement uses a discrete scale by counting the number of defective
units or the number of defects per unit. When the quality specifications are complex, it
usually is necessary to use attribute measurements. For example, a laptop may be classified as defective if it fails any of a number of functional tests or if the appearance of
the display is not satisfactory. In inspection of cloth, a defect can be defined as a flaw in
the material and the number of defects per 100 yards can be counted during inspection.
Determining the type of measurement to use also involves the specification of measuring equipment.
3. The third step in defining the quality control system is to decide on the amount of
inspection to use. Generally, a production process that is in statistical process control
minimizes the amount of inspection needed. Exceptions to this might be when process
variables are difficult to define or when the consequences of failure are very high. For
example, when human lives are at stake, both process control and inspection of every
production unit may be used.
4. The final step in designing a quality control system is deciding who will do the inspection. Usually, it is best to have workers inspect their own output and be responsible for
the quality of their work (called quality at the source). A prevention program, along
with worker responsibility for quality, will be less expensive than an extensive inspection program. In high-contact services there is no choice but to have quality at the
source, since the customer immediately perceives defects.
In some cases, the customer will be involved in inspecting the product or service.
Some business customers station inspectors at suppliers’ plants to examine and accept
or reject shipments before they are sent on to the customer. The government has inspectors in a variety of industries to ensure quality in the interest of public health and safety,
for example, in the food supply chain.
A well-designed quality control system requires a series of management judgments
and the participation of all functions. The control principles themselves are elementary, requiring performance standards, measurement, and feedback of results to correct
the process. The application of these principles in any specific situation is often complex. The guiding principle is to first control the system and then aim for continuous
improvement of the resulting stable system.
Chapter 9
9.2
Quality Control and Improvement
167 |||
PROCESS QUALITY CONTROL
LO9.2 Design a
process control
system using control
charts.
Statistical process control (SPC) utilizes inspection (or testing) of the product or service
while it is being produced. Periodic samples of the product or service are taken. When
inspection reveals that there is reason to believe that the product or service quality characteristics have changed, the production process is stopped. A search is carried out to determine an assignable cause of the change in the product or service. This could be a change
in the operator, machine, or material. When the cause is found and corrected, the process is
started again. A process can be brought into a state of statistical control and maintained in
that state through the use of quality control charts (also called control charts).
Process control is based on two key principles. The first is that random variability is present in any production process. No matter how perfectly a process is designed, there will be
some random variability in quality characteristics from one unit to the next. For example, a
machine filling cereal boxes will not deposit exactly the same weight in each box; the amount
filled will vary around some average figure. The aim of process control is to find the range
of natural random variation of the process and ensure that production stays within that range.
The second principle of process control is that production processes are not usually found
in a State of statistical control when SPC is not being used. Due to lax procedures, untrained
operators, improper machine maintenance, and so on, the variation being produced is usually much greater than necessary. The first job of process control managers is to seek out
these sources of unnecessary variation, the assignable cause variation, and bring the process into statistical control so that the remaining variation is due only to random causes.
Administrative processes in accounting, human resources, sales, marketing, and finance
in most organizations are also not usually under statistical control. These processes can
also be controlled using SPC. The same principles used to control production processes
can be used to control administrative processes.
SPC is carried out using quality control charts. In the control chart shown in Figure 9.1,
the y axis represents the quality variable or attribute characteristic that is being controlled
and the x axis represents time or a particular sample taken from the process. The center
line of the chart is the average of the quality characteristic being measured. The upper
control limit represents the maximum acceptable random variation, and the lower control
limit indicates the minimum acceptable random variation when the production process is
FIGURE 9.1
Quality control chart.
Average +
3 standard
deviations
Quality
measurement
Upper control limit (UCL)
Center line (CL)
average
Average —
3 standard
deviations
Lower control limit (LCL)
|
168| Part Three
Quality
in a state of control. Generally speaking, the upper and lower control limits are set at +
three standard deviations from the mean. If a normal probability distribution is assumed,
these control limits will include 99.74 percent of the random variation observed.
To understand this last point, take a look at the right side of Figure 9.1 showing a
normal probability distribution rotated on its side. It indicates that the distribution mean
(average) is located on the center line of the control chart and the tails of the distribution
are outside the control limits by just a small amount. Thus, 99.74 percent of the sample
observations that are taken and plotted on the graph in Figure 9.1 will fall inside the control limits, as long as the process is in a state of control.
A state of statistical control is defined as a process with a constant mean and variance
that is not changing over time. As long as these two measures of a process do not change,
the only variation that occurs is due to random causes and not due to a change in the underlying process. The purpose of a control chart is to determine when the process itself has, in
a very high degree of likelihood, shifted or changed and is therefore out of control.
In using a control chart to ensure steady-state operation, periodic samples are taken and
plotted on the control chart (see Figure 9.2). When the average of the variable or attribute
measurement falls within the control limits, the production process is allowed to continue
operating. From time to time, sample measurements signal that the process is no longer
in a state of statistical control. One such signal is when measurements fall outside of the
control limits. Other signals include measurements that, when plotted on the control chart,
reveal a trend upwards or downwards; an oscillation where measurements alternate in a
highly volatile up-down manner, or a pattern with many measurements being within one
standard deviation of the center line. When this occurs, the process is not in a state of control, and a search should be made to determine the assignable causes.
Assignable causes have also been termed special causes, those that cause points, for example, to fall outside of the control limits. Special causes are changes in materials, operator, or
machine that can be corrected and control restored. In contrast, common causes of variation
are those that randomly occur when the process is under statistical control. Common causes
cannot be removed without changing the design of the process itself. Through use of control
charts, the process can be maintained in a constant state of statistical control in which there is
only natural random variation (common causes) in the output from the transformation process.
Quality can be monitored using slightly different control charts for attributes or for variables. We cover each of these cases below.
FIGURE
9.2
Quality control chart
example.
Quality
Measurement
Sample
Chapter 9
9.3.
Quality Control and Improvement
169] |)
ATTRIBUTE CONTROL CHART
A quality characteristic can be measured on a discrete scale (e.g., an item is either good or
defective) rather than a continuous scale. An example of quality as an attribute is the percentage of defectives occurring in a sample. Other examples of attribute measurements are
the percentage of phone calls not answered within three rings, the percentage of customers
who are dissatisfied, and the percentage of parts from a supplier that are defective.
When the quality characteristic is an attribute, the appropriate type of control chart to
use 1s an attribute control chart. One particularly useful attribute control chart is the p chart
for the percentage of defectives in the sample, corresponding to the quality characteristic
of interest. For example, percentage defective is estimated by taking random samples of n
units each from a process at specified time intervals. For each sample, the observed percent
defective (p) is computed. These observed values of p are plotted on the p control chart,
one for each sample.
To determine the center line and control limits of the p chart, we take a large number of
samples of n units each. The p value is computed for each sample and then averaged over
all samples to yield a value p. This value of p is used as the center line since it represents
Attribute
Control
Chart:
A p Chart
Example
Suppose 200 records are taken from a data entry operation at two-hour intervals to monitor the data entry process. If we collect 200 records every two hours for 11 times, we
would have 11 samples, each with 200 records. The percentage of records in error for the
past 11 samples is found to be .5, 1.0, 1.5, 2.0, 1.5, 1.0, 1.5, .5, 1.0, 1.5, and 2.0 percent.
The average of these 11 sample percentages yields a p = 1.27 percent, which is the center line of the control chart. The upper and lower control limits are
UCE=VO127 3
One HST) = .0364
200
ECE =sO127
—8
CUES)
200
=O
iO
When the LCL is negative, it is rounded up to O because a negative percentage is impossible. The percent defective in each sample is then plotted on the p chart. Thus, we have
the following p chart:
3.64
Percent
defective
UCL
1.27 TALE
of TTYL
Sample number
Since all points are found to be in control, these 11 samples can be used to establish the
center line and control limits. New samples of 200 records can now be taken, plotted on
the p chart with data points moving toward the right over time, and interpreted to determine whether or not the process is still in control. Note, the formulas use the number of
items in each sample, 200, not the 11 samples taken. The 11 samples are immaterial to the
calculations except to determine the process average, Pp.
PURER) part three
Quality
the best available estimate of the true average percent defective of the process. We also use
the value of p to compute upper and lower control limits as follows:
——
penenp andy pees
——Clea
aeapee
n
n
In this case, the standard deviation of the process is the quantity under the square root
sign. We are adding and subtracting three standard deviations from the mean to get the
control limits. See the example of this computation for controlling computer data entry
operations.
After the p chart is constructed with its center line and upper and lower control limits, new
samples of the process are inspected, and the proportion defective in each sample is plotted on
the chart as an individual point. If the percentage falls within the control limits, no action is
taken. If the percentage falls outside the control limits (below the lower or above the upper),
the process is stopped and a search for an assignable cause (material, operator, or machine) is
made. After the assignable cause is found and corrected—or, in very rare cases, no assignable
cause is found—the process is restored to operating condition and production is resumed.
9.4
VARIABLES CONTROL CHART
Control charts are also used for measurements of variables. In this case, a measurement of
a continuous variable is made when each unit in the sample is inspected. As a result, two
values are computed from the sample: a measure of central tendency (usually the average) and a measure of variability (the range or standard deviation). With these values, two
control charts are developed: one for the central tendency and one for the variability of
the process. When the process is found to be out of control on either of these charts, it is
stopped and a search for an assignable cause is made.
When variable measurement is used, two control charts are needed because the normal
distribution is assumed and it has two parameters (mean and variance). Either the mean of
the distribution can change or the variance (as measured by range) can change. As a result,
we monitor both the average and the range of a process for control purposes.
Suppose that the average (x)and range (R) are computed each time a sample is taken. Then
a control chart for average and a chart for range will be used. The center line and control limits
for the average chart (often referred to as the x chart or x-bar chart) are computed as follows:
Clit
UCL =X+A,R
LED =7— ALR
where x (x-double bar) is the grand average of several past sample X averages and R is the
average of several past sample R values. Recall that the range (R) is simply the largest value
minus the smallest value in a sample. In the above formulas, A, is a constant that includes
three standard deviations from the average in terms of the range. Table 9.1 provides values
of A, for various sample sizes for the normal distribution. The center line and control limits
for the range chart are computed as follows:
CL=R
UC
01k
LGD
Chapter 9
TABLE 9.1
Control Chart
Constants
Source: Factors reproduced
from 1950 ASTM Manual on
Quality Control of Materials by
American Society for Testing
and Materials, Philadelphia.
Sample Size n
A.2
2
3
4
5
6
7
8
9
10
42
14
16
18
20
22
1.880
1.023
0.729
577
483°
419
poncréc
337
308
266
235
Pe
194
180
167
24
157
Quality Control and Improvement
D,;
D,
:
076
136
184
223
284
329
364
392
414
434
3.267
2.575
2.282
2.115
2.004
1.924
1.864
1.816
ATT
1.716
1.671
1.636
1.608
1.586
1.566
452
1.548
)
0
0
0
14741 |)
The constants D, and D, in Table 9.1 provide three standard deviation limits for the
range. The purpose of ees constants is to help calculate the upper and lower control
limits of the range chart as a function of sample size. A given sample size can be used in
Table 9.1 to look up the appropriate values of A,, D,, and D, for average and range charts.
Variables
Control Chart:
Thex &R
Charts
Example
The Midwest Bolt Company would like to control the quality of the bolts produced. Each
machine produces 100 bolts per hour and is controlled by a separate control chart. Every
hour, a random sample of six bolts is selected from the output of the machine and the diameter of each sample bolt is measured. From the six diameters, an average and range are
computed. For example, one sample produced the following six diameter measurements:
536, .507, .530, .525, .530, and .520. The average of these measurements is x = .525,
and the range is R = .029. We also know that the grand average of all past samples is
X = .513 and the grand average range is R=0.020. From these, the control chart parameters are computed as follows (see Table 9.1 for control chart constants with n = 6).
x Chart
Gl Sais
R Chart
Ciz=7020
UCL = .513 + .483(.020) = .523
UCL = 2.004(.020) = .040
LCL = .513 — .483(.020) = .503
EGE —10(020)—
Once each sample point is plotted on the two control charts, the process is found to be
out of control on average measurement and in control on range. (Note: x = .525 is outside
the upper control limit on the x chart, also the value of R = .029 is within its control limits).
We should therefore stop the process and look for an assignable cause that is making the
process produce bolts that are too large in diameter.
9.5
USING CONTROL CHARTS
There are two issues of concern in using control charts. First, the problem of sample size
must be faced. For an attribute control chart, samples should be fairly large, frequently in the
range of 50 to 300 observations. As a general rule, the sample must be large enough to allow
for the detection of at least one defective unit. For example, if the process being controlled
produces | percent defective units, a sample size of at least 100 should be used to detect one
defective unit on average. Control charts for variables require much smaller sample sizes,
usually in the range of 5 to 10 units because each unit provides much more information.
P| 172, Part Three
Quality
The second issue is how frequently to sample. This issue often is decided on the basis of the
rate of production and the cost of producing defects in relation to the cost of inspection. A highvolume production process should be sampled frequently since a large number of defective
units could be produced between samples. When the cost of producing defective units is high
in relation to the cost of inspection, the process should also be sampled frequently. An example
of a costly situation is one in which the entire production output must be screened when the
process is found to be out of control. Another example is when raw materials costs are high,
and produced units must be scrapped when inspection reveals a quality problem. In these cases,
samples should be taken frequently, provided that the cost of sampling is not too high.
Control charts are widely used in industry for both products and services. In manufacturing companies, control charts are frequently located on each machine to control the
quality output of that machine. Quality measurements are taken periodically and plotted
on the chart to ensure that the machine is still producing at its required tolerances and the
average and range haven’t changed.
In service industries control charts are used to control the time or the percentage of
defects from various processes—for example, the time it takes to answer a phone, the time
it takes to serve a customer, or the time it takes to collect accounts receivable. Service
industries also use control charts to monitor and control the percentage of dissatisfied customers or the percentage of late payments, for example.
9.6
LOY.3
PROCESS CAPABILITY
Define and
calculate process
capability.
Once a process has been brought under statistical control, process capability can be
assessed. Process capability is simply the ability of the process to meet or exceed the technical specifications obtained from customers. A process that is not capable of meeting
specifications means that defective outputs are being produced. Moreover, it is important
to remember that whether or not a process is in a state of statistical control and whether or
not a process is capable are two separate issues. Just because a process is in a state of statistical control does not mean that it is capable.
Whether or not a process is capable of meeting specifications can be determined by
computing the process capability index C,—the ratio of the specification (spec) width to
the process width:
Spec width
— Pp
Process width
If the process is centered within the specification range, as shown in both charts in
Figure 9.3, C, 2 1 will be a good indicator of the ability of the process to meet its specifications. Notice that the difference in the two charts, representing two processes, is the
process standard deviation, or the variability in measures produced by the process.
In practical use, the specification width is computed as the difference between the upper
specification limit (USL) and the lower specification limit (LSL). The process width is
computed by using six standard deviations of the measurement being monitored (60). The
standard deviation (o) refers to variation in measurement
in the individual
items being
produced. The logic for 66 is that most of the variation of a measurement is included within
+3 standard deviations of the mean, or a total of 6 standard deviations. Thus we have
Spec width
| USL—LSL
ja Process width _
60
If the process is centered in the specification range and ONS= |, the process is considered
to be minimally capable of meeting the specifications. A process with C, < 1 is not capable
Chapter 9
FIGURE 9.3
Quality Control and Improvement
173] |)
Process capability index examples.
Spec width
C p— Process width
USL-LSL
—
60
USL= 160
LSL= 100
o=
USL = 160
LSL = 100
=
5
10
Frequency
Frequency
100
160
Process width
Individual measure
Specification width
Individual measure
and must be improved by reducing the standard deviation or increasing the specification
width, if possible, to become capable.
For the normal distribution, if C,= 1 and the process is centered within the specifications
and under statistical control, 99.74 percent of the product or service produced will lie within
the specifications, corresponding to a defect rate of 2600 parts per million (ppm).' The figure
of 99.74 percent can be obtained by using the normal probability distribution tables from
Appendix A. If C, = 1.33, then 99.9967 percent of the product will lie within the specifications, corresponding to 33 ppm defective. Thus a slight increase in C,, causes a dramatic drop
in the defective rate of the process. Customers often specify Ce values from | to 1.5 or even
as high as 2.0, depending on their particular quality requirements. Table 9.2 shows why very
high process capability and extremely low rates of defects may be required in some cases.
One problem with the C,, measure is that it requires the process to be centered in the
specification range for an accurate measure of process capability. Because of this problem,
a more widely used measure (Cy) has been devised:
Cr
pk
USL—p
Min(
3
p- =)
eed
where tt = the process mean value and o = the process standard deviation.
TABLE 9.2
When
99.9 Percent Quality
Is Not Enough
If 99.9 percent quality standards were in effect, the following would happen:
*
Source: http://www.math.psu
+
+
114,500 mismatched
pairs of shoes will be shipped this
4
Mi
12 babies will be given to the wrong parents every day.
2 million documents will be lost by the IRS this year.
year.
-edu/tseng/class/99percent
+
2.5 million books will be shipped next year with the wrong cover.
Laslett
-
5.5 million cases of soft drinks produced next year will be flat.
'Note that 99.74 percent good product corresponds to (100 — 99.74) = .26 percent bad. The .26 percent
can be converted to 2600 ppm by multiplying .0026 by 1,000,000.
Py |174) Part Three
FIGURE 9.4
Quality
Computation of C,,.
USL = 160
LSL = 100
USL = 160
LSL = 100
o=
10
a
om)
1 = 100
g
pw =115
pa
=p)
ce ej 0
5
Cox =
Frequency
100
130
5)
a
=>»
Diam
be
160
100
115
130
Individual measure
Individual measure
(a)
(b)
Jt
160
This slightly more complicated measure of process capability overcomes the centering
problem by calculating the process capability for each half of the normal distribution and
then taking the minimum of the two calculations. The result is shown in Figure 9.4a, where
the value of Ci = (), while C, = |. This figure illustrates that the use of the C, index when the
process is not centered gives the wrong answer, since the process is not capable of meeting
the specifications, whereas C,, gives the correct answer with C,, = 0. A further example is
given in Figure 9.4b, where C,, = 1, even though the distribution is not centered. In this case,
the process is capable of meeting the specifications but could be improved by shifting the
mean closer to the center of the specification range. Because C,, more accurately reflects the
actual process capability, it is the measure commonly used by industry.
At this point, notice that Figures 9.3 and 9.4 are capability charts of the distribution
of individual measurements, not averages. In contrast, quality control charts are based on
plotting the sample averages of the individual measures to determine if a process is in
control. Thus, control charts and capability calculations serve two different purposes, one
to control the process and the other to determine capability. Capability charts use individual measures, not averages, because each individual item that is measured must be within
specifications.
9.7
LO9.4
CONTINUOUS
Apply
continuous
improvement
concepts using the
seven quality tools.
IMPROVEMENT
Continuous improvement is an ongoing effort to improve the process, product, or service. In the remainder of this chapter we deal with continuous improvement of processes
that are already under statistical control. Continuous improvement is needed if a process is
unable to meet customer specifications or other improvement is needed, perhaps to offer a
better product with fewer defects. If a process is not under statistical control, it must be
brought under control before continuous improvement begins. Assignable causes (machine,
operator, or material) that are often quite easily found are eliminated to bring the process
under control. When that has been done, then efforts can be made to eliminate common
Causes to improve the process further. Correcting common causes will require engineering
efforts to redesign the process, not just finding special causes that can be easily corrected.
Redesign can include specifying different input materials, buying a more capable machine,
Chapter 9
FIGURE 9.5
The seven quality
tools.
Quality Control and Improvement
Flowchart
1175] ||
Cause-and-Effect
Source: Gitlow et al., Tools and
Methods for Quality Improvement, 2d ed. (Burr Ridge, IL:
Irwin, 1994).
Variable
2
Process
Measure
instituting a new training program for operators, or making a process design change. Continuous improvement occurs only by correcting common causes.
Common causes are much more difficult to eliminate than special causes and thus
require the use of the seven quality tools shown in Figure 9.5. When finding common
causes to eliminate, teams of workers and engineers are often utilized. Table 9.3 summarizes the purpose of each of these tools for controlling and improving processes.
|
176, Part Three
Quality
TABLE 9.3
Purpose of the Seven
Quality Tools
Tool
Purpose
Flowcharts
Check sheets
Histograms
Pareto charts
Cause-and-effect diagrams
Scatter diagrams
Control charts
Understanding the process and identifying possible problem areas
Tabulating data on the problem area
Illustrating the frequency of occurrence of measures
Identifying the most important problems
Showing possible causes of the problem
Investigating relationships between two variables
Holding the gains from process improvement
The search for common causes and continuous improvement starts with flowcharting.
Flowcharts describe the flow of work and the relationships among steps in the process, and
reveal any unnecessary steps and waste that can be eliminated. A flowchart also identifies
possible quality problems that need to be investigated via further data collection and analysis.
Data collection is done using check sheets, which are a tabular list used to collect data
on the process. For example, a check sheet could contain critical process measurements
taken at periodic intervals during the day and tabulated by the time taken. It can also be
used to tabulate the frequency of certain defects or other quality-related events.
The next step in process improvement and problem solving is to display the data using
a histogram. A histogram is a frequency count using data from the check sheet to show
the form and shape of the distribution of the data. A histogram can indicate that some data
points are outliers, or there may be odd shapes to the distribution that indicate skewness or
possibly more than one mode or peak in the distribution.
A Pareto chart can be constructed to show the most important problems. In 1906, Vilfredo Pareto observed that a few items in any population constitute a significant percentage
of the entire group—the vital few. According to Pareto’s law, a few of the failure modes
account for most of the observed defects.
Consider the following example of a Pareto chart. Table 9.4 provides a count of possible reasons for hydraulic leaks found in assembling front-end tractor loaders in a factory.
As is noted in the table, the most common reason for a leak (defect) is loose connections,
followed by cracked connectors, and so on. These data are transferred to the Pareto chart
in Figure 9.6 by first calculating the percentage of all defects belonging to each cause, and
then arranging the causes from most to least frequent. Because it graphs the reasons for
leaks in decreasing order of occurrence, the Pareto chart readily shows the importance of
the various types of defects that have been found.
The Pareto chart shows which defects we should try to eliminate first. We see that we
should investigate loose connections first because they occur most frequently. Of course,
cracked connectors is a close second and should be investigated too, since these two causes
account for 78.6 percent of all defects. Pareto analysis is very helpful when one is first
studying a quality problem because it helps to focus problem-solving effort where it can
have the most impact.
TABLE 9.4
Defectives in
Front-End Loader
Hydraulics
Number Inspected (N) = 2347
Defective Items
Number of Defectives
Percent Defective
O-rings missing
16
3.9%
Improper torque
Loose connections
PAS
193
6.1
46.8
Fitting burrs
Cracked connectors
Total
Se
Say
131
412
11.4
31.8
100.0%
Chapter 9
FIGURE 9.6
Quality Control and Improvement
177)
400
Pareto diagram.
300
200
Percentage
Defectives
of
Number
100
Loose
Cracked
Fitting
Improper
connections
connectors
burrs
torque
O-rings
missing
The next step in the analysis is to take one of these failure modes, say, loose connections, and generate ideas for the causes of failure. This is done by using the cause-andeffect diagram, also called an Ishikawa diagram, after Dr. Kaoru Ishikawa, who first used
these diagrams in Japan.
A cause-and-effect (CE) diagram is shown in Figure 9.7 for the loose connections.
The problem itself, the effect, is placed on the right side of the diagram. The various potential causes of this problem are shown along the spine of the diagram and categorized, for
example, as materials, workers, inspection, and tools. The appearance of this diagram suggests a fishbone analogy. The bones of the fish are the probable causes of the problem, but
FIGURE
9.7
Cause-and-effect
diagram for loose
Large
connections.
Se
Inexperience
Material
Training
Type
Nuts
Size
Stripped
Fatigue
Method
Hose
Surface
defect
Wrong
size
Threads
Measurement
‘
Measuring tools
Errors
Judgment
Experience
Judgment
method
Torque
Adjustment
Inspector
Training
Air pressure
<=
am
UREA) Part three Quality
any cause can be listed. Each of the major causes is then broken down into more detailed
causes, giving rise to more bones on the fish. For example, the worker cause is split into
three possibilities: inexperience, fatigue, and training. Training in turn is divided into content and method.
CE diagrams frequently are constructed by quality improvement teams that include the
workers who produce the product or service. Through brainstorming, the team will identify a wide variety of possible causes for a problem. Then the team or an individual can
collect data to narrow down the potential causes before taking corrective action.
Additional data can be analyzed via a scatter diagram, which shows the relationship
between two variables. If a particular cause and effect are suspected to be related, the relationship will be apparent as a linear or curved pattern on the scatter diagram.
Using the seven quality tools makes it possible to reduce defects and thus improve quality, not just control it. For example, in the case of hydraulic leaks, the quality improvement
team may find that the loose connections are caused by the torque adjustment of the tools
and the type of threads used that would require redesign of the connection. After these
causes are corrected, the number of defects will be reduced. It is then possible to move to
the second most frequent problem, which is cracked connectors. Perhaps new materials are
needed in a revised design that are not so easily cracked. In this way, continuous improvement is achieved.
Once improvements have been made, the new process should be stabilized to hold
the gains by using a new control chart. The original control chart from before process
improvement will no longer be appropriate following the improvements. New charts
with new center lines and upper and lower limits can be created, based on data from the
improved process.
9.8
LO9.5
SIX SIGMA
Explain
Six Sigma and the
DMAIC process.
Motorola invented the term Six Sigma quality in the 1980s to reflect a desire for very high
levels of consistent quality in all its processes. At the time they realized that Japanese electronics competitors were capable of beating them on quality and cost. Six Sigma equates to
a defect rate of 3.4 parts per million (ppm), much better than they were capable of producing at the time. Getting there would require removing common causes from their
processes.
Six Sigma quality is related to the normal probability distribution, with sigma (o) denoting the standard deviation of the processes. Most quality control charts are created for processes that are at three sigma. Six Sigma refers to a stretch goal for process performance,
as well as the new DMAIC method for process improvement.
Six Sigma is a systematic method for process improvement that often uses the five steps
defined by the acronym DMAIC:
1. Define: The process is selected for improvement, and the project charter is specified.
2. Measure: Quality variables valued by the customer are measured, and goals are set for
improvement.
3. Analyze: The root causes of the current defect levels are identified, and alternatives are
considered for process changes. The analyze step uses some or all of the seven quality
tools.
4. Improve: The process is changed and checked for improvement.
5. Control: This step uses a control chart or measurements to ensure that the process
improvement is not lost over time.
Chapter 9
Quality Control and Improvement
1179 ||
The Six Sigma approach can be applied to processes in manufacturing, service, or administrative areas.
A simple application of Six Sigma occurred when a Motorola component supplier
had a quality problem. A Motorola quality professional with Six Sigma training used the
DMAIC approach to define, measure, and collect data on the problem. Using simple histograms and normal probability plots she found the data had a bimodal distribution. She
discovered that two machine operators were interpreting Motorola’s supplier requirements
differently. The supplier followed up with retraining the operators and validating the measurement technique.
Motorola has also applied Six Sigma in the finance and accounting department to
improve the cycle time to close books at the end of the month. It reduced this time from
a couple of weeks to a couple of days by removing both special and common causes. The
department also now measures how many errors were made in closing by tracking the
sigma level each month.”
Rather than improving one process at a time, management should make a strategic
choice of which processes to improve. Senior management should choose critical processes that are essential to implement the strategy of the firm. For example, top management may determine that the sales processes, the processes for hiring new employees, or a
particular manufacturing process should be selected for improvement.
Once a process is selected for improvement a cross-functional team is formed since
most processes cut across functional lines. A full-time trained process improvement specialist, usually called a “black-belt,’ is chosen to lead the improvement team. The team
then sets out to make improvements by using the DMAIC approach.
$ ¢
ls] OPERATIONS LEADER
iN
AA
sake AN
A ce
wna
a
ESA
k
.
|SS 858890008580 50 55 oo
ol
ci
Six Sigma Ranch & Winery
Six Sigma
Ranch & Winery brings the science of Six
Sigma methods to the craftsmanship of winemaking.
Owner Kaj Ahlmann, with a background in math and statistics, applied Six Sigma techniques in previous work at
General Electric. He became convinced the techniques
could be applied to winemaking as well. He was so confident that he named a winery after the Six Sigma process.
Six Sigma Ranch and Winery works toward accomplishing one goal: making wine of extraordinary quality at
an affordable price. Specifically, Six Sigma methods have
been used for:
+
Vineyard site selection
+
Selecting grapevine root stocks
+
Vineyard pruning
a#
SGrapeingeet
Wine
+”
Fermentation management
tempranillo wine in this critic’s experience.” Tempranillo
'$@ full-bodied red wine native to Spain.
oy
Six Sigma Ranch & Winery is located an hour north of
the well-known wine region of Napa Valley, California.
\
a
Pixtal/AGE Fotostock
Enthusiast
Magazine
critic said, “This is the best
A | Source: http://sixsigmaranch.com, 2019.
2www.supplychaindigital.com, 2019; www.isixsigma.com, 2019.
1 | 180 | Part Three
Quality
The team begins improvement by flowcharting the process and defining process defects,
using measures that are critical to the customer. Data are collected on these measures to
establish a current process baseline and goals for improvement. For example, if the process
is currently producing | defect in 100 opportunities (1 percent defective), a goal might be
to improve the process to 1 defect in 1000 opportunities, a 10 times (10X) improvement
factor. This kind of aggressive improvement approach is undertaken by Six Sigma teams
to ensure that significant change is achieved. Of course, the improvement goal must not
be set arbitrarily; rather, it is set on the basis of the economic benefit of improvement and
the time available for the team to accomplish its goal, with three to six months being the
typical project length.
Once the goal has been set, the team seeks the root causes of the current defect levels.
The team must be careful to go beyond symptoms and find the real causes. This is often
done by brainstorming and careful collection of data to analyze the situation. A variety of
tools such as CE diagrams, scatter diagrams, and Pareto charts are used at this stage.
After root causes are found, alternatives for improvement are considered and improvements are made. Then, further data are collected to ensure that improvements have
occurred, savings have been generated, and a control plan is put in place to ensure that the
changes are permanent. Quality control charts can be used at this point to maintain the new
process in a state of statistical control.
While the use of the Six Sigma approach is well established in manufacturing, Six
Sigma is also being deployed to improve service or administrative processes. The context
may be different but the DMAIC steps are equally applicable, as are the underlying principles. For an interesting application, read the Operations Leader box about Six Sigma
applied to winemaking.
Six Sigma has produced dramatic results in companies such as Motorola, General Electric, Citigroup, 3M, American Express, and Honeywell. These results are possible only
through aggressive senior management leadership, widespread training in Six Sigma, use
of full-time improvement specialists, and careful tracking of financial results. Six Sigma
is not merely a quality improvement approach but also a way to improve the net income
of the company. It is estimated that Six Sigma efforts have saved Fortune 500 companies in
the U.S. well over $400 billion in the past 20 years.*
9.9
LO9.6
LEAN AND SIX SIGMA
Differentiate
lean and Six Sigma.
Companies often combine lean and Six Sigma programs. While these approaches are complementary in nature, they also have several important differences in objectives, organization, methods, and type of projects. Table 9.5 compares differences between lean and Six
Sigma. These differences are based on typical implementations of lean and Six Sigma, but
definitions and usage vary widely in practice.
Objectives of Projects. Lean systems have the objective of eliminating waste defined as
non-value-adding activities. Six Sigma efforts, by contrast, are aimed at reducing defects
in the product or service. Thus, the objectives of these efforts are different, but at the same
time they can be overlapping. For example, lean can attack defects as one of the seven
wastes. Six Sigma can attack waste when it is causing a defect in the customers’ eyes, but
would not normally attack internal waste such as excessive inventory, wasted motions, or
*https://www.sixsigma.com/why-six-sigma/.
Chapter 9
TABLE 9.5
Comparison of Lean
and Six Sigma
Quality Control and Improvement
Differences
Lean
Six Sigma
Objectives
Reduce waste (non-valueadded activities); waste can
Reduce defects. Defects
184] |)
can include, in part, some
include, in part, defects
non-value-added activities
Team leadership
Part-time leaders (usually)
Use of champions
Champions notused
Full-time leaders (usually)
Vice president champions
Only selected employees
Four weeks for a black belt
- Organization:
Workforce involvement
Everyone is involved
Training
One week of training
Method used:
Steps followed
Five-step lean thinking
DMAIC steps
Data emphasis
Flowcharting ~
Less data-driven
Value stream mapping
Statistical emphasis
Any flowchart method
Use of pull systems
Yes, pull from the customer
Tracking financial impact
Not typically done
Type of projects:
Project complexity
Time for completion
Number of projects
Project selection
Simple projects
One week or less
Many small projects
Employee suggestions
Not part of Six Sigma
Closely tracked
Complex projects
Typically 3 to 6 months
Fewer large projects
Strategic project selection
unnecessary movement of materials—typical lean activities. While Six Sigma tends to be
used to reduce variance in processes, lean improves process flow.
Organization of Projects. Another difference is in the way these programs are organized.
Six Sigma typically relies on full-time black belts as project leaders and vice president
champions who select and oversee the projects. In contrast, lean systems rely on part-time
project leaders and a more informal hierarchy. Lean improvements often involve the entire
workforce, while Six Sigma projects are more selective in workforce involvement. Six
Sigma black belt training is extensive and usually requires four weeks of training plus successful completion of one or more projects. Lean training is more informal and typically
lasts only a week. Six Sigma programs are therefore different in how they are organized.
Methods Used. A third difference is in the methods employed. Lean thinking relies on a
process that starts with the customer need. While Six Sigma also starts with a customer
need, the highly structured DMAIC sequence of steps differs from those used in lean thinking. Also, lean systems do not stress the use of data or statistical analysis to the same extent
as Six Sigma. As noted in Table 9.5, value stream mapping is used in lean while Six Sigma
does not use a particular type of flowcharting. Another important difference is that only
lean uses the concept of pulling demand from the customer to flow the product or service.
Finally, lean does not formally track project cost savings or revenue improvements, while
Six Sigma insists on careful tracking by the finance function.
Types of Projects. The last area of difference is the type of projects undertaken. Lean projects are usually simple and are often based on employee suggestions, but Six Sigma is used
for complex and difficult process improvement projects. A typical Six Sigma project takes
three to six months and is aimed at a large impact, often $200,000 in savings or more. Lean
projects can last as little as one week using kaizen events and often have much less impact
from each project. Lean programs will attack many more small improvement projects than
Six Sigma and may not select projects for strategic importance.
|
1382| Part Three
Quality
Both Six Sigma and lean are aimed at improvement, but in different ways. Thus, an
organization already using lean can benefit from using Six Sigma to attack larger complex projects with a more formalized data-driven approach, using full-time project leaders
to reduce defects and variance. Conversely, an organization using Six Sigma can benefit
from the fast-hitting small-scale kaizen approach of lean, aimed at eliminating waste and
improving process flow. Similarities are that both approaches start with identifying a true
customer need that is not being met, and both are focused on process improvement.
Some companies are developing an integrated lean and Six Sigma approach. For
example, they might use the DMAIC methodology and full-time project leaders from Six
Sigma, and then incorporate the value stream mapping, pull systems, and waste reduction
focus of lean. This approach would attack waste, improve flow, and reduce defects (variance). For more detail view the video on Lean Six Sigma in the Learning Enrichment box
below.
9.10
KEY POINTS AND TERMS
This chapter introduces methods for controlling and improving quality. The major points
include the following:
*
Quality control is defined as the stabilization and maintenance of a process to produce
consistent output. Continuous improvement can occur once a stable process is achieved.
¢
Operations consists of a sequence of interconnected processes, each with its own internal customers. Critical points must be defined for inspection and measurement to control and improve these processes.
¢
Process control charts should be considered for critical points in the inputs (by the suppliers), as part of the process, and for the outputs. The critical control points are best
identified in a flowchart of the process.
¢
Using process quality control, periodic samples are taken from a production process. As
long as the sample measurements fall within the control limits, production is continued.
When the sample measurements fall outside the control limits, the process is stopped and a
search is made for an assignable cause—operator, machine, or material. With this procedure,
a production or service process is maintained in a continuous state of statistical control.
*
Itis preferable to use statistical process control instead of inspection whenever possible,
because SPC is prevention-oriented. SPC can be used for internal quality control and
achieving a certified-supplier status, which requires a stable production process.
*
A process is capable when it can consistently meet its specifications with high probability. This requires that the specification width (USL-LSL) is greater than the process
variation width (60) for a process that is centered.
*
*
*
Six Sigma is an organized and systematic approach to process improvement. It utilizes
the five DMAIC steps: define, measure, analyze, improve, and control. Careful analysis
using statistical tools is needed to identify the root causes of defects perceived by customers, analyze changes, and control the improved process.
Companies are now combining lean and Six Sigma process improvement approaches.
While these approaches both start with current customer needs, they are different in
their objectives, organization, methods, and types of projects. However, they are complementary in seeking process improvement and can be used in an integrated fashion.
There are seven quality tools that can be used to monitor quality, identify quality problems, and conduct root cause analysis when quality problems occur.
Chapter 9
¢
Key Terms
183 ||
Every function in the company can benefit from the application of quality control and
improvement. Other functions outside operations, and suppliers and customers outside
the company, will also benefit when quality control and improvement principles are
used.
Walter A. Shewhart 163
Internal customers 164
Critical control points 164
Operator inspection 165
Supplier certification 165
Variables measurement 166
Attribute measurement 166
Statistical process control 167
Assignable cause 167
Centemline sl6/ioe:
Upper control limit 167
Lower control limit 167
State of statistical control
Special cause 168
Common cause 168
Process capability 172
Continuous improvement
Seven quality tools 175
LEARNING
ENRICHMENT
Introduction to LEAN Six Sigma
https://youtu.be/DBKfGIP_NAg
(for self-study or
DMAIC Example
instructor assignments)
Quality Control and Improvement
168
174
Flowchart 176
Check sheet 176
Histogram 176
Pareto chart 176
Cause-and-effect diagram 177
Scatter diagram 178
Control chart 178
— Six Sigma 178
DMAIC 178
Video
3:14
Video
https://youtu.be/Pe4NK VEifx1
4:59
Honda Statistical Process Control
https://youtu.be/Sdj-8ZBY Ymo
Video
6:48
How to Calculate an X-bar Chart
https://youtu.be/RiKUZqW41UM
Video
3:47
Cu. Explained by Professor Cleary
https://youtu.be/ewHGNNzkjdU
Video
6:35
Statistical Process Control Slide Share
https://www.slideshare.net/anubijis/statisticalprocess-control- 16017031
Website
SOLVED PROBLEMS
Problem
1. p Control Chart A company that makes golf tees controls its production process by
periodically taking a sample of 100 tees from the production line. Each tee is inspected
for defective characteristics. Control limits are developed using three standard deviations from the mean. During the last 16 samples taken, the proportion of defective items
per sample was recorded as follows:
O1
02
O1
03
.02
O01
OO
02
| 184] Part Three
Quality
a. Determine the mean proportion defective, the UCL, and the LCL.
b. Draw a control chart and plot each of the sample measurements on it.
QO
Solution
. Does it appear that the process for making tees is in statistical control?
a. The mean proportion defective (center line) is
CL=
01 + .02 + .01 + .03 + .02 + .01 + .00 + .02 *)
.00
( + .01 + .03 + .02 + .03 + .02 + .01 + .00
16
= .015
UCL.=,.015, +3
Se:
100
501 S:40365
= .0515
LepS="015: —3
O1LSC983)
100
.015 — .0365
= —.0215, which
is negative
Therefore, the LCL = 0)
p Control Chart
Defective
Proportion
2
4
6
8
10
12
14
16
Sample Number
c. All the points are within the control limits. We can conclude that the process is in statistical control.
Chapter 9
Problem
2. x and R Control Charts
Quality Control and Improvement
185 ||)
A cereal manufacturer fills cereal boxes to an average weight
of 20 ounces and has an average range of 2 ounces when the filling process is in control.
A sample size of 10 boxes is used in evaluating the process.
a. What are the CL, UCL, and LCL for the x and R charts?
b. A sample with the following 10 measurements was just taken: 20, 21, 19, 18, 19, 21,
22, 20, 20, 19. Is the process still in control?
Solution
ah
x Chart
ir
R Chart
()
Clg= 2
UCL = 20 + .308(2)
UGE = Tig
@)
= 20.616
= 37594
LCL = 20 — .308(2)
MEL, =} 02232)
= 19'384
= 0.446
Note: Table 9.1 is used to get the control chart constants.
b. Process control charts need to be checked for both mean and range. The sample
mean is 199/10 = 19.9, and the sample range is 22 — 18 = 4. The range is out of
control, above the upper control limit, but the mean is in control. They should stop
the process and look for an assignable cause.
Problem
3. Process Capability (C,, and C,)
The operations manager of an insurance claims-
processing department wants to determine the claims-processing capability. Claims
usually take a minimum of four days to handle. The company has a commitment to
handle all claims within 10 days. On average, claims are processed in eight days and
processing has a standard deviation of one day.
a. Compute C, and C, for the claims-processing department. Based on these computations, should the claims department improve its process?
b. Using the same data, recompute Coe but use an average claims-processing time of
seven days instead of eight days.
c. Using the original data, recompute Coy but use a standard deviation of 2/3 of a day.
Which change made the most improvement—the change in mean in part b or the
change in standard deviation? Can you explain the results?
Solution
a.
C= Min{ = 8 84)
p
Blin 3k)
= Min{0.667, 1.333} = 0.667
The C, calculation seems to indicate that the process is capable of performing within
specification; however, since C_, is less than 1.0, the process needs improvement if it
is to become even minimally capable of meeting the customer service specifications.
b. If a reduced mean claim-processing time of seven days per claim is used,
Gi
:
Min
2a
SL"
TI
*3EP)
= Min{1.0, 1.0} = 1.0
REG
Part Three
Quality
c. If a reduced standard deviation of claim-processing time of .667 day per claim is used,
Beets
se
=
1
10-8 8-4 \
3(.667) 3(.667)
fs
= Min{1:0,2.0} = 1.0
Either change would result in the process being capable of meeting specifications.
Since the mean processing time was not centered within the specification limits to
begin with, shifting the mean processing time toward the center of the specification
limits has exactly the same effect as decreasing the variation in claims-processing
time. Ideally, the manager should strive for a reduction in both the mean and the
variation of claims-processing time to enhance process capability.
Discussion Questions
1. Why did statistical quality control ideas catch on in the
1940s?
b. To brainstorm the reasons why a product might have
failed.
. Suppose you make electronic calculators that contain
a chip purchased from a local vendor. How would you
decide how much inspection to perform on the chips
supplied to you?
c. To find an assignable cause.
. For the following situations, comment on whether
f. To achieve the smallest possible variance in the time
it takes to wait on tables in a restaurant.
d. To determine if a process range is under control.
e. To reduce the variability of failures found in the
field under actual use of the product.
inspection by variables or by attributes might be more
appropriate:
. A cause-and-effect diagram is used to identify the possible causes of defects. Draw a CE diagram for the following situations:
a. Filling packaged food containers to the proper
weight.
b. Inspecting for defects in cloth.
a. Your car doesn’t start in the morning.
c. Inspecting appliances for surface imperfections.
b. You received a low grade on your last exam.
d. Determining the sugar content of candy bars.
c. A student fails to graduate from college.
Workers should be given more control over the inspection of their own work. Discuss the pros and cons of
this proposition.
Why are most processes not in statistical control when
they are first sampled for control chart purposes?
. It has been said that Six Sigma is a metric, a process for
improvement, and a philosophy for managing a business. Explain these different perspectives.
Define the purpose of continuous improvement of
quality.
tte Use the DMAIC steps to describe and improve the process of ordering a book from an Internet retailer. From
the perspective of the retailer, what would be done in
each of the steps?
How is a Pareto chart used to improve quality?
WO
Which technique would be useful for each of the following situations?
How can lean and Six Sigma approaches work together
in making process improvements?
ile If an organization was using neither lean nor Six Sigma,
how would you decide which approach to use first?
a. To rank-order the causes of a quality problem.
Problems
Two Excel spreadsheets are provided on Connect
assistance in solving the chapter problems.
for
it. Golden Gopher Airline issues thousands of aircraft boarding passes to passengers each day. In some cases a boarding pass is spoiled for various reasons and discarded by
the airline agent before the final boarding pass is issued
to a customer. To control the process for issuing boarding
passes, the airline has sampled the process for 100 days
and determined the average proportion of defective passes
is .006 (6 in every 1000 passes are spoiled and discarded).
In the future, the airline plans to take a sample of 500
passes that are issued each day and calculate the proportion
of spoiled passes in that sample for control chart purposes.
a. What is the sample size (1) for this problem? Explain
the significance of the 100 days used to determine the
average proportion defective.
Chapter 9
b. Calculate the center line and the upper and lower control limits, using three standard deviations for control
purposes.
2. We have taken 12 samples of 400 book pages and found
the following proportions of defective pages: .01, .02, .02,
.00, .01, .03, .02, .01, .00, .04, .03, and .02. A page is con-
sidered defective when one or more errors are detected.
a. Calculate the control limits for a p control chart.
b. A new sample of 400 pages is taken, and 6 pages are
defective. Is the process still in control?
eXce ] 3. Each day 500 inventory control records are
cycle-counted for errors. These counts have been
made over a period of 20 days and have resulted in the following proportion of records found in error each day:
Quality Control and Improvement
187 ||e
ie The producer of electronic circuits in problem 4 has
reconsidered the method of quality control and has
decided to use process control by variables instead of
attributes. For variables control, a circuit voltage will be
measured using a sample of five circuits. The past average voltage for samples of 5 units has been 3.4 volts,
and the range has been 1.3 volts.
a. What would the upper and lower control limits be
for the resulting control charts (average and range)?
b. Five samples of voltage are taken with the following
results:
Sample
1
2
3
4
5
x
R
3.6
2.0
eS}
2.6
2.6
O:7
SS,
21
3.4
2S
.0025 .0075 .0050 .0150 .0125 .0100 .0050 .0025 .0175 .0200
0150 .0050 .0150 .0125 .0075 .0150 .0250 .0125 .0075 .0100
a.
Calculate the center line, upper control limit, and
lower control limit for a p control chart.
b. Plot the 20 points on the chart and determine which
ones are in control.
c. Is the process stable enough to begin using these
data for quality control purposes?
4. A process for producing electronic circuits has achieved
very high yield levels. An average of only 8 defective
parts per million is currently produced.
a. What are the upper and lower control limits for a
sample size of 100?
b. Recompute the upper and lower control limits for a
sample size of 10,000.
c. Which of these two sample sizes would you recommend? Explain.
5. Widgets are made in a two-shift operation. Management
is wondering if there is any difference in the proportion
of defectives produced by these two shifts.
a. How would you use the p control chart to determine if
there is a difference between the two shifts? Explain.
b. On the first shift, samples of 200 units have been used
and p = .06. Calculate CL, UCL, LCL for the first shift.
c. On the second shift, six samples of 200 units have
been taken with the proportion of defectives .04, .06,
.10, .02, .05, and .03. Using the samples from the
second shift, has the process mean shifted upward or
downward? Explain.
6. In acontrol chart application, we have found that the
grand average over all past samples of 6 units is x = 30
and R= 5.
a. Set up x and R control charts.
b. The following measurements are taken from a new
sample: 38, 35, 27, 30, 33, and 32. Is the process
still in control?
What action should be taken, if any?
c. Discuss the pros and cons of using the variables
control chart versus the control chart described in
problem 4. Which do you prefer?
A machining operation requires close tolerances on a
certain part for automobile engines. The current specification for this measurement is 3.0 cm + .001. The quality control procedure is to take a sample of 4 units and
measure each of the parts. On the basis of past samples
of size 4, X = 3.0 and R = .0020.
a. Construct average and range charts for this part.
b. On the basis of the following data, is the process in
control?
Sample
1
2
3
X
R
3.0005
0.0024
2.9904
0.0031
3.0010
0.0010
4
5
3.0015 3.0008
0.0040 0.0010
c. Is the process creating output outside of its
specifications?
. The Robin Hood Bank has noticed a decline in daily
deposits. The average daily deposit has been running at
$109 million with an average range of $15 million over
the past year. The deposits for the past six days have
been 110, 102, 96, 87, 115, and 106.
a. What are the CL, UCL, and LCL for the x and R
charts based on a sample size of six days?
b. Compute an average and range for the sample that
consists of the past six days. Do the figures for the
past six days suggest a change in the average or
range from the past year?
. A grocery store purchases fresh fish every day from its
supplier. It has ordered 100 pounds of fish each day, but
the weight actually received varies from day to day with
an average range of 6 pounds. Over the past five days
it received the following weights of fish: 106, 94, 102,
100, and 97 pounds.
RES
Part Three
Quality
a. Using this five-day sample, is the process of the fish
supplier in control in average and range?
a. Calculate the center line and control limits for the x
and R charts from these data.
b. How can the supplier more carefully control the process to provide 100 pounds of fish each day?
b. Plot each of the 20 samples on the x and R control
charts and determine which samples are out of
control.
eXce/
11. As cereal boxes are filled in a factory, they
are weighed for their contents by an automatic
scale. The target value is to put 10 ounces of cereal in
each box. Twenty samples of three boxes each have
been weighed for quality control purposes. The fill
weight for each box is shown below.
c. Do you think the process is stable enough to begin
to use these data as a basis for calculating ¥ and R
and to begin to take periodic samples of 3 for quality
control purposes?
A certain process has an upper specification limit of
220 and a lower specification limit of 160. The process
standard deviation is 6, and the mean is 170.
Observation
Sample
1
2
3
1
yD.
3
4
5
6
7
8
9
10
141
12
13
14
1S
16
17
18
19
20
10.01
9.87
10.08
10.17
10.21
10.16
10.14
9.86
10.18
9.91
10.08
9.71
10.14
10.16
Onis
10.16
10.20
9.87
9.84
10.06
9.90
10.20
9.89
10.01
1Oms
10.02
9.89
9.91
10.04
9.87
10.14
9.87
10.06
OM 7
9.94
9.81
10.10
9.93
9.91
10.19
10.03
JOANS
9.76
9.83
10.04
9.85
9.80
9.99
9.96
10.06
10.03
9.92
9.84
10.19
eo.92
9.87
10.03
10.06
9.99
10.01
a. Calculate C, and C,, for this process.
b. What could be done to improve the process capability C,, to 1.0?
lio} A certain process is under statistical control and
has a mean value of p: = 130 and a standard deviation of o = 8. The specifications for this process are
USE
S07 Sie —s00!
a. Calculate C, and Coe
b. Which of these indices is a better measure of process
capability? Why?
c. Assuming a normal distribution, what percent
of the output can be expected to fall outside the
specifications?
14. A customer has specified that they require a process
capability of C, = 1.5 for a certain product. Assume
that USL = 1100, LSL = 700, and the process is centered within the specification range.
a. What standard deviation should the process have?
b. What is the process mean value?
c. What can the company do if it is not capable of
meeting these requirements?
Ng
Capacity and
Piocersah
ANH ag
eta
PART
Scheduling
10.
Forecasting
11.
Capacity Planning
12.
Scheduling Operations
13.
Project Planning and Scheduling
Operations managers are responsible for providing sufficient capacity to meet their
firms’ needs. Forecasting is a prerequisite for any capacity or scheduling decisions.
The forecasting chapter is followed by long-, medium-, and short-range capacity
and scheduling decisions. @
;
Forecasting
nn
oh et oe a
i
i
LO10.2
Describ e the four common
LO10.6
Evaluate factors that impact forecasting method selection.
(L010.7_Describ
methods of qualitative forecasting.
i
Syren
pero oP Maticee
Ss are used to forecast.
Sa
ae
Mle See aes
LO10.8
Explain the benefits and costs of CPFR.
LO10.9
Solve advanced forecasting problems. Nese:
LO10.1
Describe
why forecasting is
important.
Nee Mes
;
Forecasting is the art and science of predicting future events. In organizations, managers are
usually most interested in predicting future demand. Before appropriate software was available,
forecasting was largely an art, but it has more recently developed into a science as well. Although
managerial judgment is still required for forecasting, managers today are aided by both simple
and sophisticated analytical tools and methods, as well as significant amounts of data from a
wide variety of sources. For example, airlines forecast future capacity needs by analyzing their
own historical demand data as well as data on the pricing and capacity of their competitors.
e
S
o
6
ts)
°
°
°
-
6
20
° °
o
°
o
0
Sie
°
ts
° °
4 o
©'6
~
°
.
©
°
o
°
?
~
oOo
°
°
°
°
Chapter10
Forecasting
1191] |]
aoe
OPERATIONS LEADER
|#
A Winning Forecast for Harrah’s Cherokee Casino Resort
oa
=
have helped the hotel achieve an annual occupancy rate
of 98.6 percent. How do they do it?
Cherokee
first determines
true customer
demand,
based on past demand and reservation requests that
were denied (because facilities were full). Customers
Steve Allen/Brand X Pictures/Jupiterimages
Harrah’s
Cherokee
of Cherokee,
North
Casino
Resort,
Carolina,
draws
located
in the town
about
four million
visitors each year. Its facilities include over 1100 hotel
rooms. The Eastern Band of Cherokee Indians contracts
with Harrah’s, the world’s largest gaming company, to
manage the casino and hotel.
Sophisticated segmentation of customers, analysis
of past demand patterns, and focused pricing strategies
provide their casino membership number when
requesting reservations, making it easy for Cherokee to track both accepted and denied reservation
requests.
Next, Cherokee develops its daily forecast based on
historical data. For example, the forecast for a particular
Saturday might be based on demand for recent Saturdays as well as similar dates in previous years. The forecast model accounts for day of week, seasonality, trends,
and other special-event factors.
Based on its forecast and knowledge about the
price each customer segment is willing to pay, Cherokee sets a room price for each customer segment on
each day.
Source: Metters et al., “The ‘Killer Application’ of Revenue
Management:
Harrah’s Cherokee Casino & Hotel,”
Interfaces, May—June 2008; and www.caesars.com/
harrahs-cherokee, 2020.
Many different methods of forecasting and their uses are described here. Selecting the
most appropriate forecasting analytics for a given situation must be done carefully for the particular use it is intended to serve. There is no universal forecasting method for all situations.
Forecasts are almost always wrong! It is rare for sales to exactly match the amount fore-
cast. A little variation from the forecast can often be absorbed by extra capacity, inventory,
or rescheduling of orders. But large variations can wreak havoc in the business. For example,
suppose 100,000 cases of a product are forecast to be sold in a particular year, but only
80,000 cases are actually sold. The extra 20,000 cases can end up in inventory, or workers
might be cut to reduce production levels. It is equally painful if the forecast is too low. Then
capacity is strained, overtime or extra workers may be added in a rush, or sales may be lost
due to stockouts. Consumers often see the results of under- or over-forecasting in retail stores
in the form of stockouts or clearance of leftover inventory. From these examples it is clear
that forecasting has a strong impact on operations and, indeed, all functions in the business.
In recognition of inherent forecasting error, all forecasts should have at least two numbers: one for the best estimate of demand (e.g., mean, median, or mode) and the other for
forecast error (standard deviation, absolute deviation, or range). To produce forecasts with
only an average is to ignore error, but this is a common occurrence in practice.
Given the challenge of getting the forecast “right,” there are three main ways for managers to accommodate forecast errors. One is to try to reduce the error through better forecasting. The second is to build more flexibility into operations and the supply chain. The third
is to reduce the lead time over which forecasts are required. This is because forecasting
error usually increases as the forecasting time horizon increases, so next week’s forecast
Py}| 192| Part Four
Capacity
and Scheduling
will usually have less error than next month’s forecast. Even good forecasts will have some
error, but the smallest possible error along with reasonable forecasting costs is the goal.
Forecasting occupies a central role in the firm, and throughout the supply chain, because
of its complexity and its impact on the business. Harrah’s Cherokee Casino Resort performs complex forecasting, as described in the Operations Leader box.
10.1
FORECASTING
FOR DECISION MAKING
Although there are many types of forecasting, this chapter focuses on forecasting demand
for output from the operations function. Demand and sales, however, are not always the
same thing. Past sales data may not reflect real customer demand when stockouts occur.
In this case, customer demand is greater than past sales and cannot be accurately forecast
from historical sales data. Sales data must be adjusted, if possible, for lost sales.
We should also clarify the difference between forecasting and planning. Forecasting deals
with what we think will happen in the future. Planning deals with what we think should happen in the future. Thus, through planning, we consciously attempt to alter future events, while
we use forecasting only to predict them. Good planning utilizes a forecast as an input. If the
forecast is not acceptable, sometimes a plan can be devised to change the course of events.
Forecasting is one input to all types of business planning and control, both inside and
outside the operations function. Marketing uses forecasts for planning products and services, promotion, and pricing. Finance uses forecasts as an input to financial planning.
Human resources requires forecasts to anticipate hiring decisions and personnel budgets.
Forecasting is an input for operations decisions on process design, capacity planning, and
inventory. Forecasting is done by firms all along the supply chain.
For process design purposes, forecasting is needed to decide on the type of process and
the degree of automation to be used. For example, a low forecast of future demand may
indicate that little automation is needed and the process should be kept as simple as possible. If, on the other hand, the forecast shows that demand is growing, management may
decide that it is a good time to investment in automation.
Capacity decisions utilize forecasts for different planning horizons. For planning facilities,
a long-range forecast several years into the future is needed. For medium-range capacity decisions, a more detailed forecast by product line or service will be needed to determine hiring
plans, subcontracting, and equipment decisions. Short-range capacity decisions, including the
assignment of available workers
and machines to jobs, require a
highly accurate forecast.
Inventory decisions resulting
in purchasing actions tend to be
short range in nature and deal
with specific products. The forecasts that lead to these decisions
must
meet
the same _ require-
ments as short-range scheduling
forecasts: They must have a high
degree of accuracy and individual product specificity.
There are different types of
decisions in operations with dif-
ferent forecasting requirements,
Operations, Marketing, Finance, and Human Resources
collaborate to both create and use forecasts.
mediaphotos/Getty Images
Chapter10
TABLE
10.1
Forecasting
193 ||
Forecasting Uses and Methods
Time
Horizon
Accuracy
Required
§Number of
Forecasts
Management
Level
Forecasting
Method
Uses of Forecasting for Operations Decisions
Long
~ Medium
Single or few
Top
Capacity planning facilities
Process design
_
Long
Medium
Single orfew
Top
Aggregate planning
Medium
High
Few
Middle
Causal and time series
Lower
Time series
Lower
Time series
Scheduling
Inventory management
Short
Highest
=
_ Short
Highest
== Many
Many
—
Qualitative or causal
|
Qualitative and causal
_ Uses of Forecasting in Marketing, Finance/Accounting, and Human Resources
Long-range marketing programs
Long
Medium
Single or few
Top
Qualitative
Pricing decisions
Short
High
Many
—
Middle
Time series
New product introduction
Medium
Medium
Single
Top
Qualitative and causal
Cost estimating
Short
High
Many
Lower
Time series
Capital budgeting
Medium
High
Few
Top
- Causal and time series
Labor planning
Medium
Medium
Few
Lower
Qualitative and time series
as shown in Table 10.1. The table also shows some of the uses of forecasts in marketing,
finance/accounting, and human resources. And it indicates the three types of forecasting
methods associated with these decisions: qualitative, time series, and causal.
In general terms, qualitative forecasting methods rely on managerial judgment; they
do not use specific quantitative models. Qualitative methods are useful when there is a lack
of data or when past data are not reliable predictors of the future.
There are two general types of quantitative forecasting analytics: time series and
causal forecasting. Quantitative methods utilize an analytical model to arrive at a forecast.
The basic assumption for all quantitative forecasting methods is that past data and data
patterns are reliable predictors of the future. Forecasting relies on predictive analytics, constructing a useful forecasting model from past demand and other relevant data.
In the remainder of this chapter, we refer to long, medium, and short time ranges. “Long
range” will mean two years or more into the future, a common horizon for the planning of
facilities and processes. “Medium range” is between six months and two years, the normal
time frame for aggregate planning decisions, budgeting, and other resource acquisition
and allocation decisions. “Short range” refers to less than six months, where the decisions
involve procurement of materials and scheduling of particular jobs and activities.
10.2
QUALITATIVE FORECASTING
LO10.2 Describe
the four common
methods of
qualitative
forecasting.
METHODS
Qualitative forecasting methods utilize managerial judgment, experience, and relevant
data, if available. Because judgment is used, two different managers using qualitative
methods may arrive at widely different forecasts.
Some people think that qualitative forecasts should be used only as a last resort. This
is not strictly true. Qualitative forecasts should be used when past data are not reliable
indicators of future demand, for example, when changes in styles or technologies alter
customer preferences. Qualitative forecasting must also be used for new product and new
service introductions for which historical demand data are not available. In these cases,
qualitative methods can be used to develop a forecast by life-cycle analogy or by the use
|
194| Part Four
Capacity
and Scheduling
TABLE
Qualitative Forecasting Methods
10.2
Accuracy
Description of Method
Uses
Short
Term
Medium
Term
Long Term
Relative
Cost
Forecast developed by a
panel of experts answer-
Capacity or facility planning. To
Fair to
very good
Fair to
very good
Fair to
very good
Medium
to high
ing aseries of questions
on successive rounds.
Anonymous responses
are fed back on each
round to all participants.
Three to six rounds
may be used to obtain
convergence.
assess when
technological
changes might
occur.
2. Market
Panels, questionnaires,
Total company
Very
Good
Fair
High
surveys
test markets, or surveys
sales, major
good
used to gather data on
market conditions.
product groups,
or individual
Fair to
good
Medium
POOL LO
fair
Low
Qualitative
Methods
1. Delphi
products.
3. Life-cycles
analogy
Prediction based on the
introduction, growth, and
maturity phases of similar products. Uses the
S-shaped sales growth
Long-range sales
for capacity or
facility planning.
Poor
Fair to
good
Forecast by a group or
an individual on the basis
Total sales
and individual
Poor to
fair
Poor to
fair
of experience, hunches,
products.
curve.
4. Informed
judgment
or facts about the situation. No rigorous method
is used.
=)
>
Source: Exhibit adapted from David M. Georgoff and Robert Murdick, “Manager’s Guide to Forecasting,” Harvard Business Review, January—February 1986, pp. 110-120.
of market research data. Note, a systematic approach to qualitative forecasting is possible
even though judgment and experience is used.
Table 10.2 describes four of the best-known qualitative methods. Qualitative methods
typically are used for medium- and long-range forecasts for planning process design or the
capacity of facilities.
Both informed judgment and the Delphi method use expert opinion to arrive at a forecast. When informed judgment is used, a panel will discuss the forecast and arrive at a
consensus. The danger of this method is that sometimes not all members of the panel are
heard, and one person can dominate the panel in terms of the final forecast.
The Delphi method was developed to correct this situation. It consists of several rounds
of anonymous data collection before reaching a forecast. In the first round, each member
of the panel anonymously provides a forecast. Then the forecast information from all panel
members 1s fed back to each panel member, again anonymously, along with any reasons or
comments about their forecasts. In the second and subsequent rounds members can review
the forecasts of the other panel members and then revise their forecasts if they find new
information. After three or four rounds of data collection, there is a tendency for the forecast to converge to a range of forecast values, and members no longer adjust their forecasts
on the basis of panel feedback. As a result, the Delphi panel arrives not only at a most
Chapter10
Forecasting
195 ||
likely forecast (e.g., mean, median, or mode) but also an estimate of forecast error (e.g.,
standard deviation, absolute deviation, or range).
Market surveys are commonly used to get information from potential customers about
willingness to buy a product or service. A variety of methods can be used, including customer responses via phone, mail, or Internet. Also, test markets are an effective way to
gauge customer demand.
The life-cycle analogy method is based on the idea that product or service demand has
well-defined life stages (introduction, growth, and maturity) that follow an S-shaped curve.
To gauge the shape of the curve, an analogy with a similar product or service is used. For
example, an estimate of demand for a new website is based on the actual growth curve of
similar websites.
Although we are not describing qualitative methods in detail, we note their usefulness
in certain situations. Next, we turn to the first type of quantitative model: time-series analytics, which are well suited to creating forecasts of a short-range nature.
10.3.
TIME SERIES ANALYTICS
Time series analytics make detailed analyses of past demand patterns to predict demand
in the future. One of the basic assumptions of all time series analytics is that demand
can be decomposed into components such as average level, trend, seasonality, cycle, and
random error. A sample of these components for a representative time series is shown in Figure 10.1. When
the components are added together (or in some cases
multiplied), they will equal the original time series data
pattern.
The /evel is the relatively constant average demand
during a time interval. The trend is an increase or
decrease in the average demand over time. Seasonality
is aregularly repeated pattern of increasing and decreasing demand. It can be yearly (such as peak demand at a
campus bookstore at the start of each semester), but can
also apply to shorter time frames, weekly or daily. For
example, a 24-hour call center experiences a seasonal
Snow boarding is an industry that exhibits several demand
patterns. It is primarily aseasonal industry, and over many
years
the industry has experienced a'growth trend: Random”
factors like snowfall cause abrupt variations in demand.
Adie Bush/Getty Images
FIGURE
10.1
demand pattern on a daily basis. Peak demand occurs
during the day, moderate demand in the evening, and
=~.
Jittle demand
during the night. Theaecycle . is increas;
;
ing or decreasing demand over long time periods, often
Decomposition of time-series data.
Demand
Demand
Time
F] | 196| Part Four
Capacity and Scheduling
many years. Cyclical changes in demand may be due to changes in the overall economy
or changes in product or service life cycles, among other reasons. Random error reflects
short-term fluctuations in demand that cannot be forecast.
The basic strategy used in time series forecasting is to identify the magnitude and form
of each component on the basis of available past data. These components, except the random error component, are then used to estimate future demand in the form of a forecast.
In discussions of time series forecasting, the following symbols and terminology are
used:
Observed
Ditiyibs
Data
Period
ed
2
as
Demands
Forecasts at Time t
De
Da,
| t-2
| t-1
D, |
| t
IF+1
F +2
(Fone
| t+1
| t+2
| t+3
|
(
Present Time
D, = demand during period ¢
F ,, = forecast demand for period t + |
e, = D, — F = forecast error in period t
A, = average computed through period ¢
We are at the end of period f, having just observed actual demand that defines the value of
D, and are making forecasts for future periods t+ 1, ¢+ 2, t+ 3, and so on.
10.4
MOVING AVERAGE
\
LO10.3 Use
forecast analytics
to calculate a
moving average
and exponential
smoothed average.
The simplest analytics method of time series forecasting is the moving average. For this
method, it is assumed that the time series has only a level component plus a random error
component. No seasonal pattern, trend, or cycle components are assumed to be present in
the demand data. More advanced versions of the moving average can, however, include
these additional components.
When the moving average is used, a given number of periods (NV) is selected for the
computations. Then the average demand, A,, for the past N periods at time ¢ is computed
as follows:
D.+D_.++-++D
—l
A, =—
:
N
—N+1
et
(10.1)
Since we are assuming that the time series is level (or horizontal), the best forecast for
period f + | is simply the average demand observed through period t. Thus, we have
Ee Sy
Each time F’,,is computed, the most recent period of demand is included in the calculation
of the average and the oldest period of demand is dropped. This procedure maintains N periods
of demand in the forecast and lets the average move along as new demand data are observed.
In Table 10.3, a three-period moving average is used for forecasting purposes. Notice how
the moving average is offset by one period to obtain the moving forecast. The forecast error
is also shown in the table as the difference between actual and forecast demand. Always use
the forecast for period ¢ (F,) in computing forecast errors, not the average for period t (A).
Chapter10
TABLE
10.3
eg Pe
ea
Moving-Average
Toceencts
Period
art
ip
D,
(demand)
1
2
10
18
tapes Pinel: Cinthia ek mae Atami
ie
,
(3-period moving
a
(3-period
average)
forecast)
ERE
1971 |e
eae
a
(BD) ie
(error)
,
3
=}Sacer
S10
S)
6
7
8
9
10
(4
WA
13
14
15
30
12
16
8
22
14
15
27
30
23
15
ea),
19.0
19.3
12.0
Psae}
14.7
70
18.7
24.0
26.7
227
4
Example
weuegeaur
Forecasting
15
:
.
26 Fret
|
80
-4.0
20.7
24.7
19.0
oes
12.0
15:3
14.7
17.0
18.7
24.0
26.7
HOS,
—12.7
—3.0
-11.3
10.0
—1.3
OG
10.0
ies
—1.0
-11.7
Using the numbers in Table 10.3 calculate a three-period moving average. The average
for period 3, A, is just the sum of demands from periods 3, 2, and 1 averaged over these
three periods:
A, =(29 + 18 + 10)/3 =19
The forecast for period 4 is equal to the moving average through period 3; therefore,
F, = 19. Once we know the actual demand in period 4, which is D, = 15, we can calculate
a forecast error as follows:
Ca D, i F,
The forecast error in period 4 is 15 — 19 = —4. Calculate the three-period moving average,
forecast, and error for the remaining periods, for practice.
The graph in Figure 10.2 shows the demand data from the example, the three-period
moving average, and a six-period moving average. The six-period moving averages are:
A, = 19.0, A, = 20.0, A, = 18.3, A, = 17.2, and so forth. It is a good idea to plot the data
and forecasts when making comparisons. Notice how the six-period moving average responds
FIGURE
10.2
Time-series data.
:
Demand
data
Three-period
moving average
J\ (bx
aa
/
oD
faeskA\
Demand
4
Time period
| | 198) Part Four
Capacity and Scheduling
more slowly to demand changes than the three-period moving average. As a general rule, the
greater the averaging periods, the slower the response to demand changes. A longer period
thus has the advantage of providing stability in the forecast but the disadvantage of responding
more slowly to real changes in the demand level. The forecasting analyst must select the appropriate trade-off between stability and response time when selecting the number of periods in N.
One way to make the moving average respond more rapidly to changes in demand is to
place relatively more weight on recent demands than on earlier ones. Any desired weights
can be specified so long as they add up to 1. This is called a weighted moving average,
which is computed as follows:
Pe
A= Wa
WD
te WDN~
t-N+1
(10.2)
with the condition
Example
If we have demands D, = 10, D, = 18, and D, = 29, and weights of W, = .5, W, = .3, and
W, = .2, the three-period weighted moving average is:
A. = (.5\(29) + (.3)(18) + (.2)(10) = 21.9
According to the formula, the weight W, is applied to the most recent demand in the third
period (29), W, to the second period (18), and W, to the first period (10). Notice how the
weighted moving average responds more rapidly than the ordinary moving average to the
increased demand of 29 in the third period.
One of the disadvantages of a weighted moving average is that the entire demand history for N periods must be used in the computation. Furthermore, the responsiveness of
a weighted moving average cannot be changed without changing each of the weights. To
overcome these difficulties, the method of exponential smoothing has been developed.
10.5
EXPONENTIAL SMOOTHING
The second predictive analytics method for forecasting is useful for reducing the amount of
past demand data that must be carried forward. Exponential smoothing is based on the simple idea that a new average can be computed from an old average along with the most recent
observed demand. Suppose, for example, we have an old average of 20 and we have just
observed a demand of 24. The new average will be between 20 and 24, depending on how
much weight we want to assign to the newest demand versus the weight on the old average.
To formalize this logic, we can write
A,zaD
H1—a)Aa,,
(10.3)
In this case, A__, is the old average (20), D, the newest demand (24), and a the proportion
of weight placed on the new demand (0 < a < 1).
To illustrate, suppose we use the values « = .1, D, = 24, and A,_, = 20. Then, from
Equation (10.3), we have A, = 20.4. If « = .5, we have A, = 22, and if « = .9, we have
A, = 23.6. Thus A, varies between the old average of 20 and the newest demand of 24,
depending on the value of « used.
Chapter10 Forecasting
1199 ||)
If we want A, to be very responsive to recent demand, we should choose a large value of
«. If we want A, to remain close to the old average, « should be small. Usually, « is given a
value between .1 and .3 to maintain reasonable stability.
In simple exponential smoothing, just as in the case of moving averages, we assume
that the time series is level with no cycles and that there are no seasonal or trend components. Then the exponentially smoothed forecast for the next period is simply the average
obtained through the current period, offset one period from the smoothed average. That is,
LRG
a
We can substitute the preceding relationship into Equation (10.3) to obtain the following equation:
F,,=aD,+(1-o)F,
(10.4)
Sometimes this alternative form of simple, or first-order, exponential smoothing is more
convenient to use than Equation (10.3) because it uses forecasts instead of averages.
Another way to view exponential smoothing is to rearrange the terms on the right-hand
side of Equation (10.4) to yield
Licata
om Sa)
This form indicates that the new forecast is the old forecast plus a proportion of the error
between the observed demand and the old forecast. The proportion of error in the new
forecast can be controlled by the choice of «.
Example
Suppose we forecasted for period 5, F, = 100, and observe the demand
for period 5,
, = 120. In this case we have an error of D, — F, = 20. If «= .1, then we add only 10 percent of this error to the old forecast to make the adjustment for the fact that demand has
exceeded the forecast. Therefore, the forecast for period 6 is:
F,, = 100 + (1)(20) = 102
Note that in using a smoothing constant of .1 we are not overreacting to the fact that
demand exceeded our forecast. However, if we want to react more quickly to demand
increases, we could increase the value of «. For example, what is the forecast for period 6
ifa=.5 or
a=.7? (Answers: For «= .5,F, = 110 and fora =.7,F, = 114.)
Students often ask why the name “exponential smoothing” has been given to this
method. It can be mathematically shown that the weights on each preceding demand data
point decrease exponentially, by a factor of (1 — «), until the demand from the first period
and the initial forecast F, is reached. Since the weights on the previous demands decrease
exponentially over time and all the weights add up to 1, exponential smoothing is just a
special form of the weighted moving average.
In Table 10.4, two exponentially smoothed forecasts are computed for « = .1 and
a = .3, using the same demand data as in Table 10.3. These data are shown graphically in
Figure 10.3. As can be seen, the « = .3 forecasts respond more rapidly to demand changes
but are less stable than « = .1. Which of these forecasts is better?
Before answering this question, we’ll look at a few rows in Table 10.4. In row | the initial
forecast for period 1, F, = 15, is given asa starting value. The demand for period | is only 10 units,
and so the forecast for period 2 will decrease. For « = .1 the new forecast F, will be 14.5,
and for « = .3 the new forecast will be 13.5. (Practice calculating these numbers yourself.)
|| | 200, Part Four
Capacity and Scheduling
TABLE 10.4
Exponential
Smoothing*
a= .1
Oye
s
D,
F,
D, —F;
F,
D,-F,
MAD,
TS (tracking
Period
(demand)
(forecast)
(error)
(forecast)
(error)
(a=:.3)
signal)
1
10
15
—5.0
15
—5.0
6.4
—.8
3
4
18
2g
15
14.5
14.85
16.26
35
14.15
—1.26
13:5
14.85
19.09
4.5
14.15
—4.09
5.8
8.3
FA
-.1
1.6
5
30
16.14
13.86
17.86
12.14
8.6
25
6
i
1
16
52
16.97
—5.52
—.97
21.50
18.65
—9.50
—2.65
8.8
7.0
1.4
1.4
8
8
16.87
—8.87
17.85
—9.85
Tee
—.1
g
10
11
12
13
14
5)
22
14
15
PAV
30
28
ih}
15.98
16.58
N6133
16.19
V2
18.54
18.99
6.02
-2.58
—1.33
10.81
278
4.46
—3.99
14.90
17.03
NEVA
15.78
VOSS:
22.40
22.58
7.6
6.2
4.7
6.7
LE,
Sy 7/
6.4
‘)
6
6
Zeal
SH
4.4
2.8
2(D, — F) Bias
36.01
71
—3.03
—1.12
ie22
10.85
0.60
—7.58
17.74
XID, — F| Absolute Deviation
95.05
103.38
13
*Assume F, = 15 as a starting point. Also assume MAD, = 7. See the text for definitions of MAD and tracking signal.
That is why we say that the forecast reacts more quickly to demand changes for higher values
of a but is less stable, since we do not know if the underlying long-term average has changed
or whether we are just seeing random fluctuation in the first period.
To answer the question of which is the best forecast, we need to look at forecast errors
over many periods. Two measures of forecast accuracy are computed in Table 10.4 for
15 periods. One measure is simply the arithmetic sum of all errors, which reflects the bias
in the forecasting method. Ideally, this sum should be zero, since the positive and negative
errors should cancel out over time. In Table 10.4, both methods have a positive bias, with
a =.1 producing more bias than « = .3.
The second measure of forecast error is the absolute deviation. In this case the absolute
values of the errors are summed, so that negative errors do not cancel positive errors. The
result is a measure of variance in the forecasting method. The total absolute deviation for
a= 1 is less than for 7 =.3,
Thus, we have the interesting result that the « = .1 forecast has more bias but less absolute deviation than the « = .3 forecast. In this case, there is no clear preference between the
two forecasting models; it depends on the manager’s evaluation of the importance of bias
FIGURE
10.3
40
Time-series data.
30
20
Demand
10
Time period
Chapter10
Forecasting
201 |[J
and deviation. However, if a forecast has both lower deviation and lower bias, it is clearly
preferred.
To determine a good value for a, forecasts should be computed for several values of a.
If one value of « produces a forecast with less bias and less deviation than the others, this
value is preferred. If no clear preference exists, trade-offs between bias and deviation must
be considered in choosing the preferred value of a.
The exponential smoothing method has the advantage that only one period of demand
and forecast data must be carried forward. Simple exponential smoothing cannot always
be used in practice because of trends or seasonal effects in the data. When these effects are
present, higher-order smoothing, trend-corrected smoothing, or seasonal smoothing may
be used. Some of these more advanced methods are presented in the chapter supplement.
10.6
FORECAST ACCURACY
LO10.4 Evaluate
forecast accuracy
using a variety of
thods.
Wwigadn
An estimate of forecast accuracy should be computed along with the forecast average.
This accuracy estimate might be used for several purposes:
1. To monitor erratic demand observations or outliers, which should be carefully evalui
ated and perhaps excluded from data analysis.
2. To determine when the forecasting method is no longer tracking actual demand and
needs to be reset.
3. To determine the parameter values (e.g., N and «) that provide the forecast with the best
accuracy.
There are four ways to measure long-run forecast accuracy over several periods. (Recall
that e, = D, — F, is the forecast error for period .)
A
Cumulative sum of
forecast errors
a
n
Choe »y 6
falitel
Mean square error
MSE = =!
Mean absolute deviation
of forecast errors
Soa
sy
n
n
Mean absolute
percentage errors
e,
»» D, oo
MAPE =
n
(expressed as
a percentage)
Note that 1 is the number of past periods used to compute the cumulative error measurements.
We have already referred to the value of CFE as the bias in the forecast. Ideally, the bias
will be zero, which occurs if positive errors are offset by negative errors. However, if the
forecast is always low, for example, the error will be positive in each period and the CFE
will be a large positive number, indicating a biased forecast. In this case the forecasting
method should be adjusted.
MSE and MAD measure the variance in the forecast error. The square root of MSE
is the well-known standard deviation o. MSE uses the square of each error term so that
)] | 202| Part Four
Capacity and Scheduling
positive and negative errors do not cancel each other out. MAD is computed from the absolute values of the error in each period instead of the squared errors. MAD is just the average error over n periods without regard to the positive or negative sign of the error in each
period. In practice, MAD is widely used because it is easy to understand and easy to use.
MAPE normalizes the error by computing a percentage error. This will make it possible
to compare forecast errors for different time series data. For example, if one time series has
low demand values and another has much higher demand values, MAPE will be an accu-
rate way of comparing the errors for these two time series.
When exponential smoothing is used, it is common to calculate the smoothed mean
absolute deviation period by period. MAD, is the mean absolute deviation in time period ¢,
which is defined as follows:
MAD =a|D,-F|+(1-a@)MAD__,
In this case, the current MAD, is simply a fraction a of the current absolute deviation
plus (1 — ) times the old MAD,_,. This is analogous to Equation (10.3), since the MAD,
is being smoothed in the same way as the forecast average. MAD. is an exponentially
weighted average of absolute error terms.
The current MAD, should be computed for each period along with the forecast average. The MAD, can then be used to detect an outlier in demand by comparing the absolute
value of the single period error (e,) with the MAD. If the absolute error is greater than
3.75 x MAD, we suspect that the demand in period f may be an outlier. The multiplier
3.75 is used here because this is comparable to determining whether an observed value lies
outside three standard deviations (6) for the normal distribution. In Table 10.4, MAD, was
computed for a = .3. As can be seen, none of the demand errors fall outside 3.75 x MAD,
and so no outliers are suspected in the demand data.
The second use of MAD, is to determine whether the forecast is tracking with the actual
time series values. To determine this, a tracking signal is computed, as follows:
Tracking signal = TS = eae
MAD,
The tracking signal is a ratio of bias (cumulative forecast error) in the numerator divided
by the most recent estimate of MAD.. If demand variations are random, control limits of
+6 on the tracking signal should ensure only a 3 percent probability that the limits will be
exceeded by chance.’ Thus, when the tracking signal exceeds +6, the forecasting method
should be adjusted to more nearly match observed demand. In Table 10.4, the tracking
signal does not exceed +6 in any period. Therefore, the forecast is considered to be tracking sufficiently close to actual demand.
As an example of these computations, refer to Table 10.4. In the last two columns of the
table we have computed smoothed MAD. and tracking signal. Starting with the arbitrary
assumption that MAD, = 7, we can compute MAD, using a = .3:
MAD, = .3|10 — 15| + .7(7) = 6.4
The tracking signal for period 1 is the cumulative error divided by MAD:
TS = —5/6.4 = —.8
As an exercise, compute MAD, and the tracking signal for period 2 and compare your
results to Table 10.4.
'The control limits and probability are based on the normal probability distribution and a value of a=.1.
Chapter10
Forecasting
203 ||]
In computerized forecasting systems, it is extremely
important to incorporate error controls of the type discussed above. This will ensure that the forecast system
does not run out of control. Instead, the user is notified
when outliers in demand are detected or when the tracking signal becomes too large.
10.7
ADVANCED TIME-SERIES
FORECASTING
A variation of exponential smoothing is adaptive
exponential smoothing. In one form, simple exponential smoothing is used but the smoothing coefficient is
varied at each forecast by +.05 to determine which of
the three forecasts has the lowest forecast error. The
resulting value of « is used for the next-period forecast.
Another type of adaptive smoothing is to continually
adjust on the basis of current forecast error. If there is a
large error, « will be large until the forecast comes back
on track. When the error is smaller, « will also be small
and a stable forecast will result. This method appears to
work quite well for inventory forecasting situations.
Table 10.5 summarizes four time series forecasting
At Kellogg USA, demand for cereal is translated into
forecasts, useful for inventory planning and meeting supply
chain goals.
methods. We have already discussed two of them, mov-
ing average and exponential smoothing. The remaining
two are described briefly below.
Rick Souders/Getty Images
TABLE 10.5
Time Series Forecasting Methods
Accuracy
Time Series
Methods
1. Moving
Description of
Method
Uses
Short
Term
Medium
Term
Long
Term
Relative
Cost
Poor
Very poor
Low
Forecast is based on arith-
Short- to medium-
Poor to
metic average or weighted
average of a given number
of past data points.
range planning for
inventories, production
levels, and scheduling.
good
2. Exponential
smoothing
Similartomoving average,
with exponentially more
weight placed on recent data.
Well adapted to large number
of items to be forecast.
Same as moving
average.
Fair to very
good
Poor to
good
Very poor
Low
3. Analytical
models
A linear or nonlinear model fitted to time series data, usually
by regression methods.
Limited, due to
expense, to a few
products.
Very good
Fair to
good
Very poor
Medium
4. Box-Jenkins
Autocorrelation methods are
used to identify underlying
time series and to fit the “best”
model. Requires about 60 past
Very
good to
excellent
Fair to
good
Very poor
Medium
to high
averages
Limited, due to
expense, to products
| requiring very accurate
short-range forecasts.
data points.
Source: Exhibit adapted from David M. Georgoff and Robert Murdick, “Manager’s Guide to Forecasting,” Harvard Business Review, January-February
1986, pp.
110-120.
My]| 204| Part Four
Capacity and Scheduling
A customized analytical model can be fitted to a time series with level, trend, and seasonal components. For example, a linear regression model or nonlinear methods can be
used. In some cases, the resulting model may provide a more accurate forecast than exponential smoothing. However, a custom-fitted model is more expensive because an individual with advanced
skills will be needed to develop it, and so the trade-off between
accuracy and model cost must be considered.
Another option, the sophisticated Box-Jenkins method, has been developed for time-series
forecasting. This technique permits more precise analysis of proposed models than is possible with the other methods. The Box-Jenkins method requires about 60 periods of past data.
For a special forecast involving a costly decision, the use of Box-Jenkins may be warranted.
In summary, time series methods are useful for short- or medium-range forecasts when
the demand pattern is expected to remain relatively stable. Time series forecasts are often
inputs to decisions concerning aggregate output planning, budgeting, resource allocation,
inventory, and scheduling.
10.8
LO10.5
CAUSAL FORECASTING ANALYTICS
Carry out
forecast analytics for
a causal model.
The second type of quantitative forecasting, causal forecasting analytics, develop a causeand-effect model between demand and other variables. For example, the demand for ice
cream may be related to population, the average summer temperature, and time. Data can
be collected on these variables and analysis conducted to determine the validity of the proposed model. One of the best-known causal methods is regression, which is usually taught
in statistics courses.
For regression methods, a model must be specified before the data are collected and
the analysis is conducted. The simplest case is the following single-variable linear model:
S=atbx
where
) = estimated demand
x = independent variable (hypothesized to cause »)
a = y intercept
b = slope
TABLE 10.6
Regression Example*
Data are collected for x and y in this model, and the parameters a and b are estimated. Then
estimates of future demand can be made from the above equation. Of course, more complicated models with multiple x’s can be developed.
We illustrate linear regression forecasting
Ta
ceEes
with.asimple.example.Suppose we are interyj
eit
ested in estimating the demand for bicycles
1
3.0
2.0
on the basis of local population. The demand
2
3.5
2.4
for bicycles over the past eight years (y,) and
:
4.1
2.8
:
44
fb
3.0
AS
(x,) are shown
the corresponding population in a small city
:
:
in Table
10.6. We first com-
6
57
36
pute the values of a and b for the regression
7
6.4
38
line using one
8
Pri)
_4.0
39
such as Excel Minitab, SPSS, JMP, or SAS.
24.8
The result is a = —1.34 and b = 2.01. The best
*The demand for bicycles, y, is expressed in Tene
The population, x,, is expressed in ten thousands of people.
of many
(least squares) equation
for bicycles
is thus y=
statistics packages,
for predicting demand
1.34 + 2.01x.
From
Chapter10
Forecasting
|205 |||
this equation we can see that the rate of increase in bicycles is 2.01 (thousands) for each
10,000-person increase in population. This rate of increase, or trend, allows us to project
bicycle demand in future years from population estimates, assuming that a linear equation
continues to be a good fit with population as a predictor variable.
Other forms of causal forecasting—econometric models, input-output models, and simulation models—are described in Table 10.7. In general, these models are more complex
and more costly to develop than regression models. However, in situations in which it is
necessary to model a segment of the economy in detail, an econometric or input-output
model may be appropriate.
Simulation models are especially useful when a supply chain or logistics system is
modeled for forecasting purposes. For example, suppose you want to estimate the demand
for TVs. A simulation model can be built representing the distribution pipeline from the
screen manufacturer, to the assembler, and finally to wholesale and retail distribution; all
imports, inventories, and exports from the supply chain would be included. Using this
model, a reasonable forecast for TVs several years into the future is obtained.
One of the most important features of causal models is that they are used to predict turning points in the demand function. Because of this ability causal models are usually more
accurate for medium- to long-range forecasts than time-series models that only predict
trends. Causal models therefore are more widely useful for facility and process planning in
operations.
TABLE 10.7
Causal Forecasting Methods
Accuracy
Short
Term
Medium
Term
Long
Term
Relative
Cost
Short- to medium-range
planning for aggregate
production or inventory involving a few
products. Useful where
strong causal relationships exist.
Good to
very good
Good to
very good
Good
Medium
Forecast of sales by
product classes for
_ short- to medium-range
planning.
Very good
to excellent
Very good
Good
High
Forecasts of companyor country-wide sales
by industrial sectors.
Not
available
Good to
very good
Good to
very good
Very
high
Good to
Good
High
Source: Exhibit adapted from David M. Georgoff and Robert Murdick, “Manager’s Guide to Forecasting,” Harvard Business Review, January-February
1986, pp. 110-120.
Causal Methods
Description of Method
Uses
1. Regression
This method relates
demand to other external
or internal variables that
tend to cause demand
changes. Uses least
squares to obtain a best
fit among the variables.
2. Econometric
model
A system of interdependent regression equations that describes some
sector of economic sales
or profit activity.
3. Input-output
model
A method of forecasting
that describes the flows
from one sector of the
economy to another. Predicts the inputs required
to produce required outputs in another sector.
4. Simulation
Simulation of the distribu-
model
tion system describing
the changes in sales and
| Forecasts of companywide sales by major
product groups.
Very good
very good
flows of product over
time. Reflects effects of
the distribution pipeline.
| | 206| Part Four
10.9
LO10.6
Capacity
and Scheduling
SELECTING A FORECASTING
Evaluate
factors that impact
forecasting method
selection.
METHOD
In this section, we present a set of factors to consider for selecting a forecasting analytics
method.
1. Use or decision characteristics. The forecasting method must be related to the use
or decisions required. The use, in turn, is closely related to characteristics such as accuracy
required, time horizon of the forecast, and number of items to be forecast. For example,
inventory, scheduling, and pricing decisions by a big box retailer require highly accurate
short-range forecasts for a large number of items. Time-series methods are ideally suited
to these requirements. By contrast, decisions by an auto manufacturer involving process,
facility planning, and marketing programs are long range in nature and require less accuracy. Qualitative or causal methods tend to be more appropriate for those decisions. In the
middle time range are aggregate planning, capital budgeting, and new product and new
service introduction decisions, which often utilize time series or causal methods.
2. Data availability. The choice of forecasting method is often constrained by available
data. An econometric model may require data that are simply not available in the short run;
therefore, another method must be selected. The Box-Jenkins time series method requires
about 60 data points (five years of monthly data). In some cases, data can be collected, but
then time and resources must also be considered relative to the importance of the forecast.
The quality of the data available is also a concern. Poor data lead to poor forecasts. Data
should be checked for outliers.
3. Data pattern. The pattern in the data will affect the type of forecasting method
selected. If the time series is level, as we have assumed in most of this chapter, a fairly
simple method can be used. However, if the data show trends or seasonal patterns, more
advanced methods will be needed. The pattern in the data will also determine whether a
time series method will suffice or whether causal models are needed. If the data pattern is
unstable over time, a qualitative method may be selected. One way to detect the pattern is
to plot the data on a graph as the first step in forecasting.
An issue concerning the selection of forecasting methods is the difference between fit and
prediction. When different models are tested, it is often thought that the model with the best fit
to historical data (least error) is also the best predictive model. This is not true. For example, suppose demand observations are obtained for the last eight time periods and we want to fit the best
time series model to these data. A polynomial model of degree seven can be made to fit exactly
through each of the past eight data points.* However, this
model is not necessarily the best predictor of the future.
The best predictive model is one that describes the
underlying time series but is not force-fitted to the data.
The correct way to fit models based on past data is to separate model fit and model prediction. First, the data set is
divided into two subsets, one for building the forecasting
model and the other hold-out set for testing the predictive
ability of the model. Several models based on reasonable assumptions about seasonality, trend, and cycle are
then fitted to the first data set. These models are used to
predict values for the second data set, and the one with
Retailers like Walmart use Google Analytics to understand
changing consumer preferences.
Mdnteorpeni\ la aa Ameo
Tenens
the lowest error on the second set is the best model. This
approach utilizes fit on the first data set and prediction on
.
;
the second as a basis for model selection.
The model would be Y = a, + a,t+a,t?+..-+ at’, where t = time.
Chapter10
Big Data
pede
Bisccullbe
ee big are
Be oe UCU
CHIC
Forecasting
|207 ||
Big data analytics are increasingly used in demand forecasting. Several surveys show that
a majority of supply chain executives consider big data analytics to be one of the most
important disruptive technologies in supply chain management today. Big data come from
sources both inside and outside the firm. Internal data useful for forecasting can include,
ce example, sales and loyalty programs. In addition to internal data, useful external data
might come from social media, or in the case of Walmart, from Google Analytics. Walmart
monitored preholiday food searches to guide forecasting for ingredients to stock in their
stores. When searches showed spikes in “tachos”—nachos with tater tots in place of
chips—Walmart increased its stock of tater tots and highlighted the dish in ads. Service
producers like Netflix also analyze big data, in their case to plan future shows that will
attract large audiences.
There are three types of resources needed to carry out big data analytics for forecasting.
First, firms need data, either collecting their own or acquiring it from other sources. Detailed
data as to the quantity and timing of past demand for particular products may be useful for
developing the types of quantitative models studied earlier in the chapter. Walmart, for example, collects data on more than one million customer transactions every hour. In addition,
less structured data such as call center recordings might be studied, coded, and analyzed for
insight into customer needs. This insight can be used to inform forecast models.
Second, firms need tools—hardware and software—to store and analyze the data. Cer-
tain data sets may require significant computing capacity, and firms may need to outsource
or acquire such capacity. And third, firms need experts with the capabilities to conduct
the analysis. Such skills are not currently widespread in most industries. But as more data
scientists, statisticians, and analysts are trained in methods including data mining and
machine learning, more rapid forecasting cycles will become common. See the Operations
Leader box on how Amazon is using big data to automate the forecasting processes.
customers.
Those data are now used to develop and
refine sophisticated forecasting analytics, enabling the
entire forecasting effort to be automated.
Under a program called “hands off the wheel” the
company is shifting tasks such as forecasting demand
for millions of products to algorithms. Initially, workers
Lisa Werner/Getty Images
Amazon.com,
Inc. is famous
for replacing
warehouse
workers with robots that carry out a variety of duties.
Using big data analytics, automation is also replacing the
humans who forecast demand and negotiate orders from
suppliers.
Since it was founded in 1994 in Seattle, Washington,
Amazon
has collected
data on the behaviors of online
could override the automated forecasts if they felt that
they should. More recently, Amazon is discouraging such
tinkering, and anyone overriding the automated forecast
has to justify their reason.
Amazon is automating interaction with suppliers as
well. An Amazon portal includes information on whether
it is interested in buying a supplier's product, the quan-
tity (based on the automated forecast), and the price they
will pay. Suppliers can simply check the portal for up-todate information.
Source: “Amazon’s clever machines are moving from
the warehouse to the headquarters,” by Spencer Soper,
Bloomberg, June 13, 2018, and www.amazon.com,
2020.
|NeaSHy || 208| Part Four
Capacity and Scheduling
Large or small data sets can be helpful for improving forecasts, in most instances. However, managers must balance the insights gained from forecast analytics with the time and
cost of obtaining them. They should prevent detailed data analysis from slowing their deci—a freeze and lack of
sion making, while similarly avoiding so-called “analysis paralysis”
action when facing unfamiliar and massive amounts of data.
10.10
LO10.8
COLLABORATIVE PLANNING, FORECASTING, AND
REPLENISHMENT
Collaborative planning, forecasting, and replenishment (CPFR) is sharing information
between business customers and suppliers in the supply chain during the planning and
forecasting process. CPFR is a valuable way in which information systems are critical to
daily operations and to supply chain performance. For example, a customer (e.g., retailer)
may have information on planned sales promotions that are not known to the supplier. In
this case a supplier’s forecast based on time series data alone would be inaccurate, but it
could be adjusted if the retailer information is shared upstream with suppliers.
Using CPFR, the customer and supplier exchange information on their respective forecasted demands. When there is a discrepancy in the forecasts, a discussion ensues to discover
the basis for the difference. After discussion, an agreed forecast is developed that becomes
the basis for replenishment planning. Note, this is a forecast and not an actual order from the
customer that would typically be placed at a later time. The collaborative forecast gives visibility into the replenishment planning processes beyond the usual ordering cycle.
Explain the
benefits and costs
of CPFR.
ig
yo
Ge
“:
:
OPERATIONS
LEADER
a
allel a
se
i7
Hi a
i
a
:
if
Havel
CPFR Creates Waves for West Marine
West Marine, the largest boating-supply retail chain in
the U.S. opened its first store in Palo Alto, California, in 1975.
Today, they have sales of more than $700 million from over
250 stores and an active website carrying products from
kayaks and anchors to life jackets and wetsuits.
During planning processes, retail stores have primary
responsibility for creating the demand forecasts for each
item. Retailers base their forecasts on data such as seasonal forecasts, planned promotions, and future product
assortment changes. These forecasts are then integrated
with planning at the company’s distribution centers and
its suppliers.
Thorsten Henn/Getty Images
West Marine gains supplier buy-in for using CPFR by
working closely with them to accommodate their materials planning schedules. Suppliers will know about special
After 10 years of using collaborative planning, forecasting, and replenishment (CPFR) with 200 of its major suppliers, West Marine has seen forecast accuracy rise 85
percent, and 96 percent of products are in-stock in retail
stores during their peak season.
promotions well in advance so that lead times (often international) can be managed.
Source: Larry Smith. “West Marine:
A CPFR Success Story,”
Supply-Chain Management Review, March 1, 2006; and
www.westmarine.com, 2020.
Chapter 10
Forecasting
209 || fs
CPFR works best in B2B relationships. Target stores, for example, share forecasted
demand with suppliers. As a result the suppliers gain visibility into expected demand shifts
and special sales promotions that Target is planning. CPFR helps coordinate the Target
supply chain. See the Operations Leader box about West Marine for another example of
the use of CPFR.
The important points to remember about CPFR are:
1. All parties must be willing to share sensitive information about demand, future sales
promotions, new products, lead times, and so forth. Assurances must be provided that
competitors will not have access to this proprietary information.
2. A long-term collaborative relationship that is mutually beneficial is needed. This will
require an atmosphere of trust and ongoing management attention.
3. Sufficient time and resources must be provided for CPFR to succeed. In other words,
there is a cost to receive the benefits of CPFR.
10.11
KEY POINTS AND TERMS
Demand forecasting analytics are crucial inputs to planning decisions within operations
and other parts of business. In this chapter, we have highlighted several important uses and
methods of forecasting. Some of the chapter’s main points are the following:
¢
Different decisions require different forecasting methods, including the following in
operations: process design, capacity planning, aggregate planning, scheduling, and
inventory management. Some of the decisions outside of operations that require forecasts are long-range marketing programs, pricing, new product and new service introduction, cost estimating, and capital budgeting. The available methods may be classified
as qualitative, time series, and causal methods.
¢
Four of the most important qualitative methods are Delphi, market surveys, life-cycle
analogy, and informed judgment. These methods are most useful when historical data
are not available or are not reliable in predicting the future. Qualitative methods are
used primarily for long- or medium-range forecasting involving process design, facilities planning, and marketing programs.
¢
Time series forecasting is used to decompose demand data into their underlying components and to project the historical pattern forward in time. The primary uses are shortto medium-term forecasting for inventory, scheduling, pricing, and costing decisions.
Some time series techniques are the moving average, exponential smoothing, mathematical models, and the Box-Jenkins method.
¢
Causal
forecasting
analytics include
regression, econometric
models,
input-output
models, and simulation models. These methods are used in an attempt to establish a
cause-and-effect relationship between demand and other variables. Causal methods can
help in predicting turning points in time-series data and are therefore most useful for
medium- to long-range forecasting.
*
Two measures of accuracy in forecasting are bias and deviation. Both should be
monitored routinely to control the accuracy of the forecasts obtained. For forecasting
applications, tracking signal and MAD are two methods used to determine if bias and
deviation, respectively, are well controlled.
¢
A forecasting method should be selected on the basis of five factors: user and system
sophistication, time and resources available, use or decision characteristics, data availability, and data pattern.
Hy]|210) Part Four
Capacity and Scheduling
*
*
Key Terms
Big data analytics are allowing firms to develop more sophisticated forecasts, using
both internal and external data. Big data analytics requires data, computing capabilities,
and expertise.
CPER is a method used to share and improve forecasts between customers and suppliers
along the supply chain and thereby reduce forecasting errors.
Qualitative forecasting
methods 193
Quantitative forecasting
analytics 193
Time series analytics 195
Eeveli95
Trend 195
Seasonality 195
Cycle 195
Random error 195
Moving average 196
Forecast error 197
Weighted moving average 198
Exponential smoothing 198
Simple exponential
smoothing 199
Bias 200
Absolute deviation 200
Forecast accuracy 201
Adaptive exponential
smoothing 203
Causal forecasting
analytics 204
Fit 206
Prediction 206
Big data analytics 207
CPFR 208
LEARNING
ENRICHMENT
Time Series Forecasting in Excel
https://youtu.be/BveOFQhSUXU
Video
6:21
(for self-study or
Delphi Technique
Video
(ast elon assiooments
https://youtu.be/bHwohMjG9OA
3:02
How Forecasting Helps Your Business
https://www.tradegecko.com/blog/
what-is-demand-forecasting-and-how-can-it-help-your-business
Website
Top-Rated Forecasting Software
Website
https://www.capterra.com/sales-forecasting-software/
Big Data Analytics at Walmart
https://blog.walmart.com/innovation/20170807/
5-ways-walmart-uses-big-data-to-help-customers
Website
More about CPFR
https://youtu.be/cBM2¢1j9q0s
Video
22
SOLVED PROBLEMS
Problem
1. Moving Average, Weighted Moving Average, and Exponential Smoothing The weekly
demand for chicken wings at a local restaurant during the past six weeks has been
Week
jaa Bp
Demand
| 650 | 521
kag
3
4
alah 3
563 | 735 | 514| 596
a. Forecast the demand for week 7 using a five-period moving average.
b. Forecast the demand for week 7 using a three-period weighted moving average. Use
the following weights: W, = .5, W, = .3, W, =.2.
Chapter10 Forecasting
211] 1)
c. Forecast the demand for week 7 using exponential smoothing. Use an a value of .1
and assume the forecast for week 6 was 600 units.
d. What assumptions are made in each of the above forecasts?
Solution
a.
D,+D,+D,+D,+D,
Fi =A,=
n
_ 596 +514 +735 +563 + 521
5
= 585.8
bk,
= Al=(W, x De) + (W3X D2) FE
xD)
= (5 X 596) + (.3 x 514) + (.2 X 735)
— 99)
c. Fj =A,=[(a)
xD.) + [Ud -@)
x F;]
= [(.1) x 596] + [U1 — .1) x 600]
= 599.6
d. We assumed the following: Future demand will be like past demand. No trend, seasonality, or cyclical effects are present. In the weighted-moving-average model, the
more recent demand is considered more important than older demand. In the exponential smoothing model, an « value of .1 puts very little weight on current demand
(10 percent), while most of the weight is put on past demand (90 percent).
Problem
. Exponential Smoothing, Exponentially Smoothed MAD, and Tracking Signal The
XYZ Company was flooded by a thunderstorm and lost part of its forecasting data.
Positions in the table that are marked [a], [b], [c], [d], [e], and [f] must be recalculated
from the remaining data.
tO
D,
(demand)
F(a = .3)
(forecast)
e,=D,=F,
(error)
(p= 5)
(MAD,)
Tracking
Signal
O
1
2
120
140
100.0
106.0
20.0
34.0
10.0
[a]
1953
18
[b]
3
160
[c]
[d]
[e]
[f]
Period
Solution
a. MAD, = (ax|D,-F |)+[(-a@) x MAD,__,]
MAD, = (ax|D, - F,|) + [C1 - a) x MAD,|
(.3 x {120 — 100]) + [(1 — .3)x 10.0]
=130
Wea omeel
‘MAD,
tat (D, -—F,)+(D,
- F,)
;
MAD,
_ 20.0 + 34.0
re
195
=2.8
i|| 212] Part Four
Capacity and Scheduling
c.F t+1 =e yee | lear Jee
F,=(axD,)+[d-a@)xF,]
(.3 x 140) + [C1 — .3) x 106.0]
= 16.2
fab
Fm
= 160— 116.2
= 43.8
e. MAD, = (a x|D,- F |)+ — a) x MAD__,]
MAD, = (a x|D, — F,|) + (1. — a) x MAD,]
= (.3 x |160 — 116.2|) + [(1 — .3) x 19.3]
= 26.7
aitig
eee
‘MAD,
tg 2 Piz FAO, -F)4+O,—F)
:
MAD
_ 20.0 + 34.0 + 43.8
26.7
esi Sad:
Discussion Questions
~~
How should the choice of « be made for exponential
smoothing?
lhe Is there a difference between forecasting demand and
forecasting sales? Can demand be forecast from historical sales data?
. Describe the difference between
fit and prediction for
forecasting models.
2. What is the distinction between forecasting and planning?
. Qualitative forecasting methods should be used only as
a last resort. Agree or disagree? Comment.
. Describe the uses of qualitative, time-series, and causal
forecasts.
. Qualitative forecasts and causal forecasts are not particularly useful as inputs to inventory and scheduling
decisions. Why is this statement true?
. What type of time series components would you expect
for the following products and services?
10. In the Stokely Appliance Company, marketing makes
a forecast based on market surveys. Meanwhile, operations makes a forecast based on past data, trends, and
seasonal components. The operations forecast usually
turns out to be 20 percent less than the forecast of the
marketing department. How should forecasting in this
company be done?
LT
What types of big data sources might be useful for
forecasting
a. Monthly sales of natural gas (for heating and
cooking).
a. Pajamas sold on a retail website?
b. Weekly sales of milk in a supermarket.
Explain how CPFR can be used to reduce forecasting
error.
c. Daily demand in a call center.
. What are the advantages of exponential smoothing over
the moving average and the weighted moving average?
b. A new urgent care clinic?
Under what circumstances might CPFR be useful, and
when is it not useful?
Chapter10 Forecasting
213 ||
Problems
Four Excel spreadsheets are provided on Connect for assistance in solving the chapter problems and supplement.
b. Graph these forecasts and the original data using
Excel. What does the graph show?
1. Daily demand for marigold flowers at a large garden
c. Which of the above forecasts is best? Why?
store is shown below. Compute:
a. A three-period moving average for each period.
4. The Handy-Dandy Department Store had forecast sales
of $110,000 for the previous week. The actual sales were
b. A five-period moving average for each period.
$130,000.
a.
Period
Demand
1
85
5,
92
Ss
71
What is the forecast for this week, using exponential
smoothing and « = .1?
b. If sales this week turn out to be $120,000, what is the
forecast for next week?
5. The Yummy Ice Cream Company uses the exponential
4
97
smoothing method. Last week the forecast was 100,000
5
93
gallons of ice cream, and 90,000 gallons was actually
6
82
sold.
7
89
a. Using « = .1, prepare a forecast for next week.
2. The number of daily calls for the repair of Speedy copy
machines has been recorded as follows:
October
Calls
:
Re
3
106
4
5
165
125
6
111
7
178
8
97
b. Calculate the forecast using « = .2 and a = .3. Which
value of a gives the best forecast, assuming actual
demand is 95,000 gallons?
6. Using the data in problem 2, prepare exponentially
smoothed forecasts for the following cases:
a. a=.land
F,=90
b. a= .3 and F, = 90
7. Compute the errors of bias and absolute deviation for the
forecasts in problem 6. Which of the forecasting models
is the best?
eXcel
8. At the ABC Floral Shop, an argument
developed between two of the owners, Maya
and Henry, over the accuracy of forecasting
a. Prepare three-period moving-average forecasts for the
data. What is the error on each day?
b. Prepare three-period weighted-moving-average
fore-
methods. Maya argued that exponential smoothing with
a = .1 would be the best method. Henry argued that the
shop would get a better forecast with « = .3.
casts using weights of W, = .5, W, = .3, W, = .2.
c. Which of the two forecasts is better?
a. Using F, = 100 and the data from problem 3, which of
the two managers is right?
3. The ABC Floral Shop sold the following number of geraniums during the last two weeks:
b. Graph the two forecasts and the original data using
Excel. What does the graph reveal?
c. Maybe forecast accuracy could be improved. Try addi-
Day
1
Demand
200
Day
8
Demand
154
2)
3
4
134
147
165
9
10
11
182
197
132
183
12
163
146
14
169
5
6
125
5
13
157
Develop a spreadsheet for the following.
a.
Calculate
forecasts
moving average.
using
a three-
and
five-period
tional values of « = .2, .4, and .5 to see if better accuracy is achieved.
9. Only a portion of the following table for exponential
smoothing has been completed. Complete the missing
entries using «= .1.
|
5
Period
D,
Le
O
1
BOON
90
2
280
S
309
e,
MAD,
20
racking
Signal
| 214] Part Four
Capacity and Scheduling
Develop an Excel spreadsheet to answer the following
questions.
10. A candy store has sold the following number of pounds
of candy for the past three days. Assuming « = .4, complete the following table.
Period
O
1
2
3
D,
A,
enF
ace!
MAD;
16
exes
20
26
14
11.
For the first seven days of data compare the absolute
deviation for forecasts using aw = .1 and a = .3. Start
with A, = 33. Which method is best?
Tracking
Signal
. Use the second week of data to make the same com-
1
parison. Use A, = 32. Which method is best now?
C.
Le
A grocery
store sells the following num-
a.
er of frozen turkeys during the week prior to
Thanksgiving:
Develop a spreadsheet using the first seven days of
demand to determine the best exponential smoothing
model for values of « = .2, a = .3, and a = .4. Select
the model with the smallest absolute deviation for
seven periods.
Turkeys Sold
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
What does this example illustrate?
13. The Easyfit tire store had demand for tires shown
below. Assume F, = 198.
Develop another spreadsheet using the second seven
days to compare the best exponential smoothing
model found in part a with a three-period movingaverage model. Compare the predictions on the basis
of the total absolute deviation.
80
53
65
43
85
101
What principles does this problem illustrate?
a. Prepare a forecast of sales for each day, starting with
F, = 85 and a= .2.
Day
b. Compute the MAD and the tracking signal in each
period. Use MAD, = 0.
c. On the basis of the criteria given in the text, are the
MAD and tracking signal within tolerances?
A
d. Recompute parts a and b using a = .1, .3, and .4.
Which value of « provides the best forecast?
OBWDNDN
NO
=
The famous Widget Company uses simple exponential
smoothing to forecast the demand for its best-selling
widgets. The company is considering whether it
should use « = .1 or « = .3 for forecasting purposes.
Use the following data for daily sales to arrive at a
recommendation:
Day
Demand
Day
Demand
1
2
3
4
S)
6
7:
35
47
46
SS,
26
3S
24
8
9
10
11
2
13
14
69
24
26
36
43
46
29
Demand
Day
Demand
200
209
Zils
180
190
(Qs
200
8
9
10
At
12
13
14
208
186
193
197
188
191
196
14. The ABC Floral Shop from problem 3 is considering
fitting various forecasting models on the first seven
days of demand and using the second seven days as a
hold-out sample for comparing the prediction accuracy
of the models. They have decided to use « = .25, but
aren't sure what starting value of forecast, F,, to use.
a.
Try values of F, = 160, F, = 170, and F, = 180 to
determine the best exponential model for the first
seven days using the minimum total absolute deviation as the criterion. You may modify the spreadsheet from problem 8 for the calculations.
Compare the best model from part a to a three-period
moving average model on the second set of data.
Which one has the smallest sum of absolute errors?
C.
What principles does this problem illustrate?
Chapter10
Forecasting
215] ||
Supplement
Advanced Methods
LO10.9 Solve
advanced forecasting
This supplement describes three additional methods for time series forecasting that have
tend and seasonal components. These methods are extensions of the techniques described
probleris
in the chapter.
:
When the time series model has a trend component, an exponential smoothing model
can be developed that is based on updating two variables in each time period: an average
level and a trend. The average level is computed as an expanded version of the first-order
equation to include trend, as follows:
A= aD,+ (1—a)(Aia+
Tu)
This average is then, in turn, used to update the estimate of trend by taking the difference in
averages and smoothing this difference with the old trend. The updated trend is thus
Tei AveaAge ich Gl iB) dius,
In this case, the smoothing constant B, which can be the same as or different from the con-
stant w used for level, is used for trend. The model requires initial estimates of A, and T, to
get started. These estimates can be based either on judgment or on past data.
Using the above values, we can compute forecasts for the future. The procedure is now
slightly different from the first-order case, because a constant trend is assumed in the time
series. The forecast for period t + K in the future is therefore
F pi Apt Rol). Kiso 2e3y <3:
One unit of trend is added for each period into the future. An example using these formulas
is shown in Table S10.1.
Time series that have both trend and seasonal components can be forecast. In this case,
three variables—average, trend, and a seasonal factor—are updated for each time period.
The average is computed for period ¢ as follows:
D
A,= (e) Feet) ae
tae.)
tL
In this case, the demand is adjusted by the seasonal ratio and smoothed with the old average and old trend. The trend for period t is
EECA
a)le
The seasonal ratio for period t is
TABLE $10.1
penal egont
ee
ig ri
Smoothing
|
ae
D, (demand)
A, (average)
1
85
85
T, (trend)
15
F, (forecast)
85
_
D,-F, (error)
2
105
100.5
3
112
15.05
115.2
100
15.01
5.00
115.55
-3.55
)
4
132
130.4
15.03
130.21
G79
5
145
145.4
iis:03
145.43
—.43
*Assume A, = 70, T, = 15, a =.1, B=.1.
|]|216
Part Four
Capacity and Scheduling
TABLE $10.2
Seasonal Exponential
Smoothing Method*
t
1
2
3
4
D,
A,
Tes
R,
le
Dia
(demand)
(average)
(trend)
(seasonal
ratio)
(forecast)
(error)
66
106
78
135
80.5
90.1
995
110.1
10.1
10.0
ys)
10.0
.804
1195
TE)
1.201
64
108.7
80.4
130.7
2.0
—2.7
—2.4
4.3
*Assume A, = 70, Ty = 10, L = 2, R_, = .8, R=
1.2, a
ep
=o y=
23
In this case, we are assuming that the seasonal cycle is L periods. There are L seasonal
ratios, one for each period. If the demand is monthly and the seasonal cycle repeats on
an annual basis, L = 12. Each month, one of the seasonal ratios will be updated to a new
value, along with the trend and average.
The model requires initial estimates of Ap, Tp PLN CW (Cane
ORa rc These initial esti-
mates can be based on judgment or data if available.
According to the updated values, the forecast for future periods in period f is
Laie
=
(A, ai KT, )(R,_7
x)
An example of this method is shown in Table $10.2.
If there is no trend, this method can be used with seasonal factors alone. In this case, the
above trend equation and 7, values are simply dropped.
One of the techniques used frequently in time series forecasting is classical decomposition. This involves decomposing a time series into level, trend, seasonal, and possibly
cyclic components. Decomposition is illustrated by an example with three years of quarterly data from a store that sells children’s toys. It is assumed that the seasonal pattern is
quarterly in nature, and there may be trend and level components in the data as well. Since
only three years of data are available, no cycle component will be estimated.
The quarterly data on the sales of toys are shown in Table $10.3. The biggest sales of
toys, by far, are in the fourth quarter. Visual inspection of the data indicates an upward
trend—but how can this trend be disentangled from the seasonality of the data? This is
done by first computing a four-quarter moving average. Decomposition requires the same
number of periods in the moving average as the seasonality of the data (i.e., 4 periods for
quarterly seasonality and 12 periods for monthly seasonality). This is done to average out
the high periods and the low periods of demand over the seasonal cycle. The four-period
moving average is shown in the third column of Table $10.3. These moving averages are
centered between periods, because a four-period average should represent a point with two
periods on each side. From column 3, the upward trend is clear, because the seasonality has
been removed from the data.
To calculate seasonal ratios, we need an average for each period. This is done in column
4 of Table $10.3 by constructing a two-period moving average of column 3. These new averages are then once again centered on the periods of data instead of between periods. Column
4 then represents the best average of the data for estimating the trend. It is also used to calculate seasonal ratios directly by dividing sales by column 4 to produce the seasonal ratios in
column 5. Interpretation of these ratios is as follows: Demand in the third quarter is 95.8 percent of the annual average, demand in the fourth quarter is 170.9 percent of the yearly average, and so on. To obtain a best estimate of the seasonal ratios, we simply average the ratios
for corresponding quarters. This calculation is shown on the bottom of Table $10.3. Note that
the seasonal ratios are quite stable in this example, but we have a minimal amount of data to
work with. Ordinarily, at least four years of data should be used to establish seasonal ratios.
Chapter10
TABLE $10.3
Forecasting
2171 |)
Two-Period
Classical
Decomposition
Method
Quarter
Sales*
1
30
3
55
5
Moving Average
SSS
G'S
58
Moving Average
\
Seasonal Ratio
& 57.4
= \
.958
4
100
+585
1.709
5
6
v
35
46
Ste)
60
65
67
59)5)
62.5
66.0
588
.736
894
8
120
69.75
68.4
1.754
9
43
72.75
Vales
.603
10
57
78.25
755
WSS)
a
12
ZA
142
;
Seasonal Ratios Quarter
Average
1
2
3
4
.588
136
958
894
1.709
1.754
.603
.596
55
746
.926
Ae7S2
*Sales are in thousands of dollars.
The original sales data and the de-seasonalized moving average, from column 4, are
plotted in Figure $10.1. The moving average indicates an upward trend line. Actually, the
trend might be slightly nonlinear, but let us assume for this example a linear trend line.
Then a regression line can be fitted to the eight moving-average points shown on the graph.
The result is
Y@)=47.8 + 2.631
where y(t) = sales and ¢ = time.
A trend line could also be fitted to the original sales data, but it is customary in classical
decomposition to use moving averages before fitting the trend line. This seems to give a
slightly more stable forecast.
FIGURE $10.1
160
SS
Eee
Seasonal toy sales.
140
— —-—-
Moving average
Trend (regression)
&
120
3
> 100
~
1
&
80
=]
oO
60
40
20
0
1
2
3
4
1
2
3
Quarter
4
1
2
3
4
} | 218] Part Four
Capacity and Scheduling
TABLE $10.4
Quarter
Seasonal Forecast
Calculations
es
Predicted Average
820
x
Seasonal Factor
S;
Forecast
14
84.6
.746
63.1
5
87.2
.926
80.7
16
89.9
eZ.
SIS):7/
596
48.8
To forecast sales for the coming year, the following method is used. First, use the trendline equation to predict the average for quarters 13, 14, 15, and 16 by inserting these values
of time into the above regression equation. This yields column 2 in Table $10.4. Then multiply the seasonal ratio for each quarter by the predicted average. The result is a forecast for
each quarter of the next year, as shown in Table $10.4.
Supplement Problems
1. Ace Hardware sells spare parts for lawn mowers. The
following data were collected for one week in May when
replacement lawn-mower blades were in high demand:
a. Simulate a forecast of the demand using trendadjusted exponential smoothing. Use values of
Ay = 90, Ty = 25, anda = 6 =.2.
b. Plot the data and the forecast on a graph.
Day
Demand
1
2
10
12
3. The SureGrip Tire Company produces tires of various
sizes and shapes. The demand for tires tends to follow a
3
13
quarterly seasonal pattern with a trend. For a particular
4
5
6
y,
15
Aad
20
2a
type of tire the company’s current estimates are as follows: Ay = 10,000, T) = 1,000, Ry = -8, R_, = 1.2,
R_,=1.5,andR_,=.75.
a. Simulate
a forecast
for the week,
c. Does this appear to be a good model for the data?
*
a. The
company
has just observed
the first quarter
starting with
of demand D, = 6000 and would like to update its
model given in the chapter supplement.
b. Compute the MAD and tracking signal for the data.
Use MAD, = 0.
OD
ae.
b. When demand is observed for the second quarter, it is
D, = 15,000. How much error is there in the forecast?
c. Are the MAD and tracking signal within tolerances?
c. Update the forecasts again for the coming year, using
F, = 10, T, = 2, a = .2, and B = .4. Use the trend
d. Simulate a forecast using simple smoothing for the
week, starting with F, = 10 and o = .2. Plot the forecast and the demand on a graph. Note how the forecast
lags behind demand.
j aN
eXce
forecast
for each of the next
Siig eueratan
eXce
four quarters
using
aus TeRELES
4. Management believes there is a seasonal
pattern in the above data for the Donut-Hole
Shop (see problem 2), with the first two days of a week
The daily deiandfonghtcoloie dams
representing one level; the third and fourth days repre-
fomihe Dome Haleshdie she taee ial
senting a second level; and the fifth, sixth, and seventh
days a third level. Thus, three seasonal factors have
been suggested: Ry = .9, R_, = 1.3, and R_, =.8.
a. Simulate a forecast of demand for days 1 to 7 using
A) = 85, T, = 0, and a = $ =y ="1:
b. Comment on the appropriateness of the forecasts
for a two-week period.
Day
Donna
an
Day
:
Demand
=
Z
Be
.
99
3y
120
110
10
11
110
90
5
6
7
75
60
50
12
13
14
80
65
50
Sais
:
5. Management of the ABC Floral Shop believes that its
sales are seasonal in nature with a monthly seasonal pattern and no trend. The demand data and seasonal ratios
for the past three years are given as follows.
Chapter10 Forecasting
Year 1
Year 2
Year 3
Seasonal
Month
Demand
Demand
Demand
Ratio
Jan.
$12,400
Feb.
Mar.
Apr.
May
23,000
15,800
20,500
25,100
June
16,200
July
Aug.
Sept.
Oct.
Nov.
Dec.
12,000
10,300
11,800
14,000
10,700
7,600
$11,800
DAA gl
16,500
21,000
24,300
15,800
11,500
10,100
11,000
14,300
10,900
7,200
$13,600
21,800
14,900
19,400
26,000
16,500
12,400
10,800
12,500
13,800
10,600
8,100
0.8
1.8
0.9
1.6
2.0
1.0
0.7
0.6
0.7
ile2
0.9
0.6
. Calculate a forecast for Year 3 using A, =
/219] |
15,000,
a = y = .3, and the seasonal ratios shown above. For
each period, calculate the forecast and the updated
seasonal ratio.
. Plot the original data and the forecast on a graph.
. Calculate the tracking signals for the past year using
MAD, = 0. Are they within tolerances?
. Using the classical decomposition method described
in the chapter supplement, calculate the seasonal
ratios from the data and determine the trend and average levels. Use these ratios and estimates of trend and
level to make a forecast for the next year.
Capacity Planning
ah
amt
hh &@ ® o
LO11.2
Illustrate with an example a facilities strategy that considers: amount, size, timing, location, and type.
LO11.4
Identify the demand and supply options that are available for S&OP.
LO11.6
TT 20
,
In this chapter, we discuss capacity decisions related to carrying out the production of
goods and services. Firms make capacity planning decisions that are long range, medium
range, and short range in nature. Long-range decisions are concerned with facilities that
typically extend one or more years into the future. The first part of this chapter describes
facilities decisions and a strategic approach to making them. In this chapter we also deal
with medium-range aggregate planning, which extends from six to eighteen months into
the future. The next chapter discusses short-range capacity decisions of less than six
months regarding the scheduling of available resources to meet demand.
Facilities, aggregate planning, and scheduling form a hierarchy of capacity decisions
about the planning of operations extending from long, to medium, to short range in nature.
First, facility planning decisions are long term in nature, made to obtain physical capacity
that must be planned, developed, and constructed before its intended use. Then, aggregate
2
eo
°
a
°
Chapter11 Capacity Planning EEXM!
FIGURE 11.1
PCy
ons.
decisi
eeLcanacitya
————————
ies.
[aFacilit
e
:
[
“Aggregate
planning
———T
_
\_ ||
;
7)
}
Facilities
decisions
:
Aggregate
planning
Scheduling
Scheduling
Months
Planning Horizon
planning determines the workforce level and production output level for the medium term
within the facility capacity available. Finally, scheduling consists of short-term decisions
that are constrained by aggregate planning and allocates the available capacity by assigning it to specific activities.
This hierarchy of capacity decisions is shown in Figure 11.1. Notice that the decisions
proceed from the top down and that there are feedback loops from the bottom up. Thus,
scheduling decisions often indicate a need for revised aggregate planning, and aggregate
planning also may uncover needs to update facility decisions.
Capacity planning requires involvement of all functional areas. Long-range planning is
closely tied with budgeting and the finance function, while medium-range capacity decisions require input from marketing and human resources. Even accounting and information
systems closely interact with capacity during planning phases and on an ongoing basis. We
discuss this cross-functional involvement in more detail later in the chapter.
11.1
CAPACITY DEFINED
LO11.1 Define
capacity and
utilization.
We define capacity (sometimes referred to as peak capacity) as the maximum output that
can be produced over a specific period of time, such as a day, week, or year. Capacity can
be measured in terms of output such as number of units produced, tons produced, or num-
ber of customers served over a specified period. It also can be measured by physical asset
availability, such as the number of hotel rooms available, or labor availability, for example,
the labor hours available for consulting or accounting services.
Estimating capacity depends on reasonable assumptions about facilities, equipment,
and workforce availability for one, two, or three shifts as well as the operating days per
week or per year. If we assume two eight-hour shifts are available for five days per week all
year, the capacity of a facility is 16 x 5 = 80 hours per week and 80 x 52 = 4160 hours per
year. However, if the facility is staffed for only one shift, these capacity estimates must be
halved. Facility capacity is not available unless there is a workforce in place to operate it.
Utilization is the relationship between actual output and capacity and is defined by the
following formula:
Utilization =
Actual output
Capacity
x 100%
c||222| Part Four
Capacity and Scheduling
ih
ul|
OPERATIONS LEADER
|
Delta Airlines
It is not your imagination. Air travelers are feeling pinched
75 percent
these days.
Over the last decade, the proportion of airline seats
occupied by passengers is way up. Airlines today fill, on
average, more than 85 percent of their seats, and the
most popular flight times and routes are completely full.
This contrasts with average seat utilization rates of about
j
20 years ago.
Industry leader
10 years
Delta
Air Lines, flying
passengers
year,
serves
each
ago
and
a spacious
as
65 percent
180
a useful
million
example.
Headquartered in Atlanta, GA, they fly to 304 destinations in 52 countries, including flights by their alliance
partners. Delta owns a fleet of more than 800 planes
which take off about 3000 times every day. With an
86.2 percent passenger load, their capacity utilization
rate continues a slow upward climb that is similar to the
industry average.
In the airline industry, passenger capacity is measured
in terms of the number of seats that can be sold. Those
seats have gotten smaller and closer together in recent
years, as any flyer can attest! Each unit of sellable capacity is valuable, and airlines will sell as many as they can.
So, if you are feeling crowded on your next flight, that
is capacity management
sattapapan tratong/123RF
in action.
Source: www.delta.com, 2020.
The utilization of capacity is a useful measure for estimating how busy a facility is
or the proportion of total capacity being used. It is almost never reasonable to plan for
100 percent utilization since spare (slack) capacity is needed for planned and unplanned
events. Planned events may include required maintenance or equipment replacement, and
unplanned events could be a late delivery from a supplier or unexpected demand.
Utilization rates vary widely by industry and firm. Continuous flow processes may have
utilization near 100 percent. Facilities with assembly-line processes may set planned utilization at 80 percent to allow for flexibility to meet unexpected demand. Batch and job
shop processes generally have even lower utilization. Emergency services such as police,
fire, and emergency medical care often have fairly low utilization, in part so that they can
meet the demands placed on them during catastrophic events. The Operations Leader box
describes capacity measurement and utilization at Delta Airlines.
It is possible in the short term for some firms to operate above 100 percent utilization.
Overtime or an increased work-flow rate can be used in the short term to meet highly variable or seasonal demand. Mail and package delivery services often use these means to
increase work output before major holidays. However, firms cannot sustain this rapid rate
of work for more than a short period. Worker burnout, delayed equipment maintenance,
and increased costs make it undesirable to operate at very high utilization over the medium
or long term for most firms.
In addition to the theoretical peak capacity, there is an effective capacity that is
obtained by subtracting downtime for maintenance, shift breaks, schedule changes, absenteeism, and other activities that decrease the capacity available. The effective capacity is
the amount of capacity that can be used in planning for actual output over a period of time.
To estimate effective capacity for the previously described two-shift facility, we must subtract hours for planned and unforeseen events when capacity is not being utilized.
Chapter11 Capacity Planning RES)
11.2
FACILITIES
LO11.2 Illustrate
with an example a
facilities strategy that
considers amount,
size, timing, location,
and type.
DECISIONS
Facilities decisions, the longest-term capacity planning decisions, are of great importance
to a firm. These decisions place physical constraints on the amount of goods or services
that can be produced, and they often require significant capital investment. Therefore,
facilities decisions involve input from all organizational functions and often are made at
the highest corporate level, including top management and the board of directors.
Firms must decide whether to expand existing facilities or build new ones. As we discuss
the facilities strategy below, we will see there are trade-offs that must be considered. Expanding current facilities may provide location conveniences for current employees but may not
be the best location in the long-run. Alternatively, new facilities can be located near a larger
potential workforce but require duplication of activities such as maintenance and training.
When construction is required, the lead time for many facilities decisions ranges from
one to five years. The one-year time frame generally involves buildings and equipment that
can be constructed quickly or leased. The five-year time frame involves large and complex
facilities such as oil refineries, paper mills, steel mills, and electricity generating plants.
In facilities decisions, there are five crucial questions:
1 . How much capacity is needed?
2 . How large should each facility be?
3. When is the capacity needed?
4 . Where should the facilities be located?
5. What types of facilities/capacity are needed?
These questions can be separated conceptually but are often intertwined in practice. As a
result, facilities decisions are exceedingly complex and difficult to analyze.
Facilities strategy is one of the critical parts of an operations strategy. Since major facilities decisions affect competitive success, they need to be considered as part of the total
operations strategy, not simply as a series of incremental capital budgeting decisions.
A facilities strategy considers the amount of capacity, the size of facilities, the timing of capacity changes, facilities locations, and the types of facilities needed for the long
run. It must be coordinated with other functional areas due to the necessary investments
(finance), market sizes that determine the amount of capacity needed (marketing), work-
force issues related to staffing new facilities (human resources), estimating costs in new
facilities (accounting), and technology decisions regarding equipment investments (information systems and engineering). The facilities strategy needs to be considered in an integrated fashion with these functional areas and will be affected by the following factors:
1. Predicted demand. The expected demand for future years is a key factor in adding or
reducing capacity.
2. Cost of facilities. Cost is driven by the amount of capacity added or subtracted at one
time, the timing, and the location of capacity.
3. Likely behavior of competitors. An expected slow competitive response may lead
the firm to add capacity to grab the market before competitors become strong. In contrast, an expected fast competitive response may cause the firm to be more cautious in
expanding capacity.
4. Business strategy. The business strategy may dictate that a company put more emphasis on cost, service, or flexibility in facilities choices. For example, a business strategy
to provide the best service can lead to facilities with some excess capacity or several
market locations for fast service. Other business strategies can lead to cost minimization
or attempts to maximize future flexibility.
P] | 224, Part Four
Capacity and Scheduling
5. Global considerations. As markets and supply chains continue to become more global
in nature, facilities often are located globally. This involves not merely chasing cheap
labor but locating facilities for the best strategic advantage, sometimes to access new
markets or to obtain desired expertise in the workforce.
Amount of
Capacity
One part of a facilities strategy is the amount of capacity needed. This is determined both
by forecasted demand and by a strategic decision by the firm about how much capacity
to provide in relation to expected demand. This can best be described by the notion of a
capacity cushion, which is defined as follows:
Capacity cushion = 100% — Utilization
The capacity cushion is the difference between the output that a firm could achieve and the
actual output that it produces to meet demand. Since capacity utilization reflects the output
required to satisfy demand, a positive cushion means that there is more capacity available
than is required to satisfy demand. Zero cushion means that the average demand equals the
capacity available.
The decision regarding a planned amount of cushion is strategic in nature. Three strategies can be adopted with respect to the amount of capacity cushion:
1. Large cushion. In this strategy, a large positive capacity cushion, with capacity in
excess of average demand, is planned. The firm intentionally has more capacity than the
average demand forecast. This type of strategy is appropriate when there is an expanding market or when the cost of building and operating capacity is inexpensive relative to
the cost of running out of capacity. Electric utilities adopt this approach, since blackouts
and brownouts are generally not acceptable. Firms in growing markets may adopt a
positive capacity cushion to enable them to capture market share ahead of their competitors. Also, a large cushion can help a firm meet unpredictable customer demand, for
example, for new technologies that very quickly become popular.
2. Moderate cushion. In this strategy, the firm is more conservative with respect to capacity. Capacity is built to meet the average forecasted demand comfortably, with enough
excess Capacity to satisfy unexpected changes in demand as long as the changes are not
hugely different from the forecast. This strategy is used when the cost (or consequences)
FACILITIES STRATEGY.
This Bacardi Rum
factory supplies the
entire North American
market from a single
modern automated
distillery in Puerto
Rico.
Irina Moskalev/123RF
Chapter 11.
Capacity Planning
225 ||]
of running out is approximately in balance with the cost of excess capacity. An appliance manufacturer, with fairly predictable demand, might use a moderate cushion.
3. Small cushion. In this strategy, a small or nearly zero capacity cushion is planned to
maximize utilization. This strategy is appropriate when capacity is very expensive relative to stockouts, as in the case of oil refineries, paper mills, and other capital intensive
industries. These facilities operate profitably only at very high utilization rates between
90 and 100 percent.
When planning the capacity cushion, firms assess the probability of various levels
of demand based on their forecast and then use those estimates to make decisions about
planned increases or decreases in capacity. For example, suppose a firm has capacity to
produce 1200 units. It also has a 50 percent probability of 1000 units of demand, and
50 percent probability of 800 units of demand. Then average demand is estimated to be (.5 x
1000) + (.5 x 800) = 900 units. Producing 900 units results in a (900/1200) x 100% = 75%
utilization rate. Based on existing capacity, the cushion is (100% — 75%) = 25%.
The solved problems section at the end of the chapter provides an example of how to
compute the capacity cushion by using probabilities of demand, existing levels of capacity,
and costs of building capacity. This method provides an analytic basis for estimating the
amount of capacity cushion that may be required.
Size of
Facilities
After deciding on the amount of capacity, a facilities strategy must address how large each
unit of capacity should be. This is a question involving economies of scale, based on the
notion that large facilities are generally more economical because fixed costs can be spread
over more units of production. Economies of scale occur for two reasons. First, the cost of
building and operating large production equipment does not increase linearly with volume.
A machine with twice the output rate generally costs less than twice as much to buy and
operate. Also, in larger facilities the overhead related to managers and staff can be spread
over more units of production. As a result, the unit cost of production falls as facility size
increases when scale economies are present, as shown in Figure 11.2.
This is a good news-bad news story, for along with economies of scale come diseconomies of scale. As a facility gets larger, diseconomies can occur for several reasons. First,
logistics diseconomies are present. For example, in a manufacturing firm, one large facility
incurs more transportation costs to deliver goods to markets than do two smaller facilities
that are closer to their markets. In a service firm, a larger facility may require more movement of customers or materials, for example, moving patients around a large hospital or
moving mail through a regional sorting center. Diseconomies of scale also occur because
coordination costs increase in large facilities. As more layers of staff and management are
added to manage large facilities, costs can increase faster than output. Furthermore, costs
FIGURE 11.2
Optimal facility size.
Economies of scale
Unit
Cost
Diseconomies of scale
Facility Size (units produced per year)
} | 226)
Part Four
Capacity and Scheduling
related to complexity and confusion rise
as more products or services are added
to a single facility. For these reasons, the
curve in Figure 11.2 rises on the righthand side due to diseconomies of scale.
As Figure 11.2 indicates, there is a
minimum unit cost for a certain facility size. This optimal facility size will
depend on how high the fixed costs are
and how rapidly diseconomies of scale
occur. For example, in the wood furni-
ture industry there are plants with over
a thousand employees and other plants
with only a few employees. Each firm
seems to have an optimal facility size,
depending on its demand volume, cost
structure, product/service mix, and par-
ticular operations strategy, which may
emphasize cost, delivery flexibility, or
service. Cost is, after all, not the only
factor that affects facility size.
Timing of
Facility
Decisions
Another element of facilities strategy is
the timing of capacity additions. There
are basically two opposite strategies here.
1. Preemptive
strategy,
Strategy.
In _ this
the firm leads by build-
Starbuck
a ALE apaly TUES
ti
trat
e cusiorier
t
i
Gemane
John Flournoy/McGraw-Hill Education
ing capacity in advance of the
needs of the market. This strategy provides a positive capacity cushion and may actually stimulate the market
while at the same time preventing competition from coming in for a while. Apple
Inc. used this strategy in the early days of the personal computer market. Apple
built capacity in advance of demand and had a large share of the market before competitors moved in. Apple still uses this strategy today in building massive capacity and inventories in advance of new product launches for the iPad and iPhone.
2. Wait-and-see Strategy. In this strategy, the firm waits to add capacity until demand
develops and the need for more capacity is clear. As a result, the company lags market demand, using a lower-risk strategy. A small or negative capacity cushion can
develop, and a loss of potential market share may result. However, this strategy can be
effective because superior marketing channels or technology can allow the follower to
capture market share. For example, in the smartphone market Android phones (e.g.,
Samsung and others) have been able to take away market share from the leader, Apple,
through pricing, advertising, and new products. In contrast, U.S. automobile companies
followed the wait-and-see strategy, to their chagrin, for small autos. While U.S. automakers waited to see how demand for small cars would develop, Japanese manufacturers grabbed a dominant position in the U.S. small-car market.
Facility
Location
Facility location decisions have become more complex as globalization has expanded the
options for locating capacity and developing new markets. For example, Starbucks may
choose to build facilities in regions with heavy coffee drinkers, competing for customers
there. Alternatively, it may choose to locate facilities in regions where people typically do not
Chapter 11
Capacity Planning
227] ||
consume coffee and attempt to create demand for their product and service. Starbucks’s U.S.
competitor Caribou Coffee tries to find facilities locations on the right-hand side of the road
of morning traffic, because customers are more likely to stop frequently when they can pull
over on the right but are less willing to make cumbersome left-hand turns in heavy traffic!
Location decisions are made by considering both quantitative and qualitative factors.
Quantitative factors that affect the location decision may include return on investment, net
present value, transportation costs, taxes, and lead times for delivering goods and services.
Qualitative factors can include language and norms, attitudes among workers and customers, and proximity to customers, suppliers, and competitors. Front office services in particular must often locate near customers for the customers’ convenience, and so this factor
may outweigh most others in determining where to locate new facilities. Examples include
clinics, grocery stores, and restaurants.
Firms often compare potential locations by weighting the importance of each factor that
is relevant to the decision and then scoring each potential location on those factors. Then,
multiplying the factor weight by the location score allows calculation of a weighted-average
score for each site. This score provides insight on how well each potential site meets the
needs of the firm and may be used to make the final facility location decision.
Types of
Facilities
The final element in facility strategy considers the question of what the firm plans to
accomplish in each facility. There are four different types of facilities:
1. Product-focused (55 percent)
2. Market-focused (30 percent)
3. Process-focused (10 percent)
4. General-purpose (5 percent)
The figures in parentheses indicate the approximate percentages of companies in the Fortune 500 using each type of facility.
Product-focused facilities produce one family or type of product or service, usually
for a large market. An example is the Andersen Corporation main facility, which produces
various types of windows for the entire U.S. Product-focused plants often are used when
transportation costs are low or economies of scale are high. This tends to centralize facilities into one location or a few locations. Other examples of product-focused facilities are
large credit card processing operations and auto leasing companies that process leases for
cars throughout the U.S. from a single site.
Market-focused facilities are located in the markets they serve. Many service facilities
fall into this category since services generally cannot be transported. Plants that require
quick customer response or customized products or that have high transportation costs tend
to be market focused. For example, due to the bulky nature and high shipping costs of mattresses, most production plants are located in regional markets. Some international facilities
are market focused because of tariffs, trade barriers, and potential currency fluctuations.
Process-focused facilities have one technology or at most two. These facilities frequently produce components or subassemblies that are supplied to other facilities for
further processing. This is common in the auto industry, in which engine plants and transmission plants feed the final assembly plants. Process-focused plants such as oil refineries
can make a limited variety of products within the given process technology.
General-purpose facilities may produce several types of products and services, using
several different processes. For example, general-purpose facilities are used to manufacture furniture and to provide consumer banking and investment services. Most hospitals
are general-purpose facilities. General-purpose facilities usually offer a great deal of flexibility in terms of the mix of products or services that are produced there. They sometimes
are used by firms that do not have sufficient volume to justify more than one facility.
| | 228| Part Four
Capacity and Scheduling
s
Hess
ae
|vn ha
mo ala
a ae
ol
i ge
a
salen a cl ala
5
Wd
on
Leal Falla
a
sp i
Strategic Capacity Planning at BMW
Analytics using mathematical
optimization models (e.g.,
mixed linear programming) can be helpful in evaluating
various capacity strategies. Given a forecast of demand
over the next several years, the models will determine
the minimum cost plan for meeting the demand.
An example from BMW helps explain this process.
BMW uses a 12-year future planning horizon to represent
the typical development
new BMW
and
production
designs. The mathematical
life cycle for
model calculates
the supply of finished products that should be produced
by each plant to meet the forecasts in all global markets.
As a result, the model determines how much of each
product should be produced in each plant over the next
12 years. All possible
BMWs
are
considered
locations, amounts,
by the
and types of
mathematical
model
in
arriving at the minimum cost plan.
This type of analysis is very helpful in setting overall capacity strategies for how much, how large, where,
when, and what type. For example, the resulting strategy might be to use distributed production to produce
in each national market the amount sold there or a more
production and export strategy. Due to the
centralized
many assumptions made, the models do not determine
the final capacity strategy; however, they are very helpful
in evaluating many different possibilities.
Source: B. Fleischmann, S. Ferber, and P. Henrich, “Strategic
Planning of BMW’s Global Production Network,” /nterfaces
36(3), 2006; and www.bmwgroup-plants.com/en.html, 2020.
Goran JakuS/123RF
\
We have shown how a facilities strategy can be developed by considering questions
of capacity, size of facilities, timing, location, and types of facilities. Analytics using
mathematical optimization models can often be helpful in answering these five strategic
questions as discussed in the BMW Operations Leader box. We now shift from long-term
facilities decisions to medium-term decisions regarding how capacity, once built, is used.
11.3.
LO11.3
S&OP
SALES AND OPERATIONS PLANNING
Explain how
is done.
Sales and operations planning (S&OP) is a term used by many firms to describe the
aggregate planning process. Aggregate planning is the activity of matching supply of
output (production) with demand (market) over the medium time range. The time frame is
between six and eighteen months into the future, or an average of about one year.
The term aggregate implies that the planning is done for a single overall measure of output or at most a few aggregated product or service categories. Products can be grouped, for
planning purposes, into a product family of items with similar production needs and serving
similar markets. The product family might include all refrigerators for an appliance producer
or all surgical procedures for planning at a hospital. The aim of S&OP is to set overall output
levels in the medium-term future in the face of fluctuating or uncertain demand.
S&OP, or aggregage planning, includes the following characteristics and assumptions:
I. A time horizon of about 12 months, with updating of the plan on a periodic basis, perhaps monthly.
Chapter11. Capacity Planning EREYR
2. An aggregate level of demand for one or a few categories of product or service. The
demand is assumed to be fluctuating, uncertain, or seasonal.
3. The possibility of changing both supply and demand variables.
4. A variety of management objectives, which might include low inventories, good labor
relations, low costs, flexibility to increase future output levels, and good customer
service.
5. Facilities that are considered fixed and cannot'be expanded or reduced.
As a result of S&OP, decisions related to the workforce are made concerning hiring,
laying off, overtime, and subcontracting. Decisions regarding production output and inventory levels are also made. S&OP is used not only to plan production output levels but also
to determine the appropriate resource input mix to use. Since facilities are assumed to be
fixed and cannot be expanded or contracted, management must consider how to use existing facilities and resources to best match market demand.
S&OP can involve plans to influence demand as well as supply. Factors such as pricing,
advertising, and product mix may be considered in planning for the medium term. We will
discuss these tactics later in this chapter.
S&OP generally is done by product family, including the equipment and workforce
needed to produce output. Usually no more than a few product families are used for S&OP
to limit the complexity of the planning process. Inconsistencies between supply and
demand are resolved by revising the plan as conditions change.
S&OP matches supply and demand by using a cross-functional team approach. The crossfunctional team consisting of marketing, sales, engineering, human resources, operations,
CALAN
OPERATIONS
can
LEADER
S&OP at Hostess Brands
Hostess knew it needed to manage costs closely.
such as gas stations and grocers, now play
a bigger role in both production planning and delivery.
Retailers,
The
new
model
required
retailers to forecast demand,
order goods, receive goods at their own distribution centers, deliver to their stores, and stock their own shelves.
Retailers signed on to this new approach because their
customers
wanted Hostess products. Hostess became
the first in its segment to use this forecast, planning, and
delivery model.
Under their faster S&OP
their attention toward
new
processes,
Hostess turned
product offerings, which
are
up 80 percent. They tapped customer interest in holidaybhofack2/Getty Images
themed Twinkies, DingDongs, and HoHos. Even pumpkin
spice Twinkies!
Hearts were broken in 2012 when Hostess announced
they were stopping production of such favorites as
Twinkies and HoHos. But under new ownership, the
company relaunched in 2013 as Hostess Brands LLC
and moved quickly to rebuild the brand, with new S&OP
processes and a new supply chain design.
With
limited shelf life, product turnover
must be quick, and Hostess Brands’ new business model
is making that happen.
Source: Abe Eshkenazi, “Sweetening the Deal: How
Hostess Improved Its Supply Chain,” January 15, 2016,
www.apics.org; and www.hostessbrands.com,
2020.
|
230| Part Four
Capacity and Scheduling
and finance meets with the general manager to agree on the sales forecast, the supply plan,
and any steps needed to modify supply or demand. During the S&OP process, demand
is decoupled from supply. For each product family, the cross-functional team must decide
whether to produce inventory, manage customer demand, or provide additional capacity
(internal or external). Once plans to manage demand and supply are in balance, however,
the current plan may not agree with previous financial plans or human resources plans or
budgets, which also may need to be modified.
The resulting sales and operations plan is updated approximately monthly, using a
12-month or longer rolling planning horizon. At its best, S&OP reduces misalignment
among functions by requiring a common plan to be implemented by all parties. Strong
general manager leadership may be required to resolve any conflicts that arise. Read the
Operations Leader box to learn how Hostess Brands is using S&OP to revive their business.
Syngenta is a world leader in agribusiness products such as seeds and fertilizer, with
28,000 employees in 90 countries. The highly seasonal agricultural market is difficult to
forecast and experiences large demand shifts. Syngenta uses S&OP to create collaboration
across functions and with its supply chain partners. Managers across several countries use
the S&OP process and the supporting software to gain better agreement on forecasts, sales
promotions, inventory levels, sales plans, and aggregate production plans.
Since S&OP is a form of aggregate planning, it precedes detailed scheduling, which is
covered in the next chapter. Scheduling serves to allocate the capacity made available by
aggregate planning to specific jobs, activities, or orders.
11.4
CROSS-FUNCTIONAL
NATURE OF S&OP
Aggregate planning for how capacity will be used in the medium-term future is the responsibility of the operations function. However, the planning process requires cross-functional
coordination and cooperation with all functions in the firm, including accounting, finance,
human resources, and marketing.
S&OP or aggregate planning is closely related to other business decisions involving,
for example, budgeting, personnel, and marketing. The relationship to budgeting is particularly strong. Most budgets are based on assumptions about aggregate output, personnel
levels, inventory levels, purchasing levels, and so forth. An aggregate plan thus should be
the basis for initial budget development and for budget revisions as conditions warrant.
Personnel, or human resource planning, is also greatly affected by S&OP because such
planning for future production can result in hiring, layoffs, and overtime decisions. In service industries, which cannot use inventory as a buffer against changing demand, aggregate
planning is sometimes synonymous with budgeting and personnel planning, particularly in
labor-intensive services that rely heavily on the workforce to deliver services.
Marketing must always be closely involved in S&OP because the future supply of output, and thus customer service, is being determined. Marketing, through pricing, promotion, and other activities, will impact demand.
Therefore, it is essential that such sales
plans and their timing be accounted for in the planning process.
S&OP is not a stand-alone system. It is a key input into the enterprise resource planning (ERP) system, which is discussed later. ERP tracks all detailed transactions from
orders to shipments to payments, but requires a high-level aggregate plan for future sales
and operations as an input. The ERP system accepts as input the S&OP plan and projects
the detailed transactions (shop orders, purchase orders, inventories, and payments) that are
required to support the agreed plan. Figure 11.3 captures these inputs and outputs from an
S&OP process.
Chapter 11.
FIGURE
Capacity Planning
/231| |] S
11.3
Relationship of S&OP
with other functions.
S&OP
Process
ERP System
- Production
plan
In some firms, the S&OP process is broken or missing. Top management does not
actively support or participate in the process. Accountability for S&OP is inadequate or in
conflict across functions. The firm’s information system may not support S&OP, and so
the firm is not able to conduct important “what if” analysis. Also, S&OP plans may not be
executed by all organizational functions as has been agreed. Therefore, to be successful,
the S&OP system may require changes in the organization, reporting and accountability,
and the information systems.
11.5
PLANNING OPTIONS
LO11.4 Identify the.
demand and supply
options that are
available for S&OP.
| Operations and marketing must work closely to plan for matching supply and demand over
the medium-term time frame. They do this during aggregate planning when they coordinate their decisions to develop enough demand for products and services without overshooting available facility capacity.
The S&OP process includes discussion of the various decision options available. These
include two categories of decisions: (1) those modifying demand and (2) those modifying
supply.
Demand management entails modifying or influencing demand in several ways:
1. Pricing. Differential pricing often is used to reduce peak demand or to build up demand
in off-peak periods. Some examples are matinee movie prices, off-season hotel rates,
factory discounts for early- or Jate-season purchases, and off-peak specials at restaurants. The purpose of these pricing schemes is to level demand through the day, week,
month, or year.
} | 232| Part Four
Capacity and Scheduling
2. Advertising and promotion. These methods are used to stimulate or in some cases
smooth out demand. Advertising can be timed to promote demand during slow periods
and shift demand from peak periods to slack times. For example, golf courses advertise
to lengthen their season and turkey growers advertise to stimulate demand outside the
peak holiday seasons.
3. Backlogs or reservations. In some cases, demand is influenced by asking customers to
wait for their orders (backlog) or by reserving capacity in advance (reservations). Generally, this has the effect of shifting demand from peak periods to periods with slack capacity. However, backlogs may have the undesirable effect of losing customers or profit.
4. Development of complementary offerings. Firms with highly seasonal demands may try
to develop products that have countercyclic seasonal trends. For example, Toro produces
both lawn mowers and snowblowers, seasonally complementary products. In fast-food
restaurants, breakfast has been added in many cases to utilize previously idle capacity.
The service industries, using all the mechanisms described above, have gone much
further than most of their manufacturing counterparts in influencing demand. Because
they are unable to inventory their output—their services—they use these mechanisms to
improve their utilization of fixed facility capacity.
Supply management includes a variety of ways to increase or decrease supply through
aggregate planning. These include the following:
|. Hiring and laying off employees. Some firms will do almost anything before reducing the size of the workforce through layoffs. Other companies routinely increase and
decrease the workforce as demand changes. These practices, which vary widely by firm
and industry, affect not only costs but also labor relations, productivity, and worker
morale. As a result, company hiring and layoff practices may be restricted by union
contracts or company policies. These effects must be factored into decisions about
whether to change the size of the workforce to match demand more closely.
2. Using overtime and undertime. Overtime sometimes is used for short- or mediumrange labor adjustments in lieu of hiring and layoffs, especially if the change in demand
is considered temporary. Overtime labor usually costs 150 percent of regular time, with
double time on weekends and holidays. Because of its high cost, managers are sometimes reluctant to use overtime. Furthermore, workers may be reluctant to work more
than 20 percent weekly overtime for an extended period. Undertime refers to planned
underutilization of the workforce rather than layoffs, perhaps by using a shortened workweek. Undertime, in the form of furloughs, is common in many industries, including
manufacturing, education, and government during recessions or periods of low demand.
3. Using part-time or temporary labor. In some cases, it is possible to hire part-time
or temporary employees to meet peak or seasonal demand. This option is particularly
attractive when part-time employees are paid significantly less in wages and benefits.
Unions frequently frown on the use of part-time employees because they often do not
pay union dues and may weaken union influence. Part-time and temporary labor is
used extensively in many service operations, such as restaurants, hospitals, supermarkets, and department stores. Temporary labor is also common in agriculture-related
industries.
4. Carrying inventory. In manufacturing firms, inventory can be used as a buffer between
supply and demand. Inventories for later use can be built up during periods of low
demand. Inventory thus decouples supply from demand in manufacturing operations,
allowing for smoother operations with more even use of capacity throughout a time
period. Inventory is a way to store expended capacity and labor for future consumption.
Chapter 11
Capacity Planning
|233 ||]
This option is generally not available for service operations; this results in greater challenges for service industries in matching supply and demand.
5. Subcontracting. Subcontracting is the outsourcing of work (either manufacturing or
service activities) to other firms. This option can be very effective for increasing or
decreasing supply. The subcontractor may supply the entire product or service or only
some of the components. For example, a manufacturer of toys may utilize subcontractors to make plastic parts during periods of high demand. Service operations may subcontract call center operations or catering services during peak periods.
6. Cooperative arrangements. These arrangements are similar to subcontracting in that
other sources of supply are used, but cooperative arrangements often involve partner
firms that are typically competitors. The firms choose to share their capacity, thus preventing either firm from building capacity that would be used only during brief periods.
Examples include electric utilities that link their capacity through power-sharing networks, hospitals that send patients to other hospitals during demand peaks, and hotels
or airlines that shift customers among one another when they are fully booked.
In considering all these options, it is clear that S&OP and the aggregate planning
decisions are extremely broad and affect most parts of the firm. The decisions that are
made are therefore strategic and cross-functional, and should reflect all of the firm’s
objectives. If aggregate planning is considered narrowly, it may not bring to light the
trade-offs that can occur. Some of the multiple trade-offs that should be considered are
customer service level (through back orders or lost demand), inventory levels, stability
of the labor force, and costs. The conflicting objectives and trade-offs among these elements sometimes are combined into a single cost function. A method for doing this is
described next.
11.6
BASIC AGGREGATE
PLANNING STRATEGIES
LO11.5 Contrast
and compare the
For most firms, demand for their product or service changes over time. For example,
demand for lawn fertilizer is very seasonal in nature, but the firm may prefer to keep
chase and level
full-time workers all year. Demand for services can also fluctuate, such as resorts, which
strategies.
depend on good weather and consumer interest. Resort
owners may prefer to frequently hire and lay off workers to closely match demand. In this section, we discuss
Using a level strategy often results in holding inventory
during low demand periods.
Andersen Ross/Getty Images
the aggregate planning strategies used in each of these
instances and explain how firms decide which strategy
is best for them.
Two basic planning strategies can be used, as well as
combinations of them, to meet medium-term aggregate
demand. One strategy is to maintain a level workforce;
the other is to chase demand with the workforce.
With a level strategy, the size of the workforce and
the rate of regular-time output are constant. Any variations in demand must be absorbed by using inventories,
overtime, temporary workers, subcontracting, cooperative arrangements, or any of the demand-influencing
options discussed above. The level strategy essentially
holds the regular: workforce oe
at a fixed number, and so
the rate of workforce output is fixed over the aggregate
Hy| 2341 Part Four
LSA
Capacity and Scheduling
ML LNA
ee
LA
AER
RES
ANE
OPERATIONS LEADER
|#5)
In September 2017, Hurricane Irma became the most
powerful storm to hit the continental U.S. in more than a
decade. Before Irma slammed the Florida coast, Travelers
claims employees from around the country were moving
toward the affected areas so that they would be immediately available to serve affected customers. Even before
the storm made landfall, teams of trained and equipped
claims professionals were ready.
Travelers uses a level strategy, with workers from
many regions pitching in during large-scale disasters and
workers being diverted to training and catch-up activities
during slow times when it might appear to be overstaffed.
The strategy seems to pay off in terms of maintaining
Pororsnarebutiersiocs
high levels of customer service during a peak demand
Travelers offers a wide variety of insurance products and
period. Travelers was able to contact most of its Hurri-
services, covering customers in more than 90 countries. It
sells insurance for auto owners, renters, home owners, and
businesses. Travelers is an example of a service firm that
uses a level strategy to manage its workforce and capacity.
cane
Irma-affected
customers
within 48 hours of their
reporting a claim and to inspect, provide payment, and
close more than 90 percent of claims within 30 days.
Source: www.travelers.com, 2020.
planning period. However, a firm using a level strategy can respond to fluctuations in
demand by using the demand and supply planning options discussed in the previous
section.
With a chase strategy, the size of the workforce is changed to meet, or chase, demand.
With this strategy, it is not necessary to carry inventory or use the demand and supply
planning options available for aggregate planning; the workforce absorbs all changes in
demand. The chase strategy generally results in a fair amount of hiring and laying off of
workers as demand is chased.
These two strategies are extremes; one strategy makes no change in the workforce, and
the other varies the workforce directly with demand changes. In practice, many firms use a
combination of these two strategies. Read the Operations Leader box to see how Travelers
Insurance successfully uses a level strategy even though demand for its service can vary
significantly.
11.7
LO11.6
AGGREGATE
Define
the various costs
associated with
aggregate planning.
PLANNING COSTS
Most aggregate planning methods include a plan that minimizes costs. These methods
assume that demand is given (based on forecasts) but varies over time; therefore, strategies
for modifying demand are not considered. If both demand and supply are modified simultaneously, it may be more appropriate to develop a model to maximize profit rather than
minimize costs, since demand changes affect revenues along with costs.
Chapter11. Capacity Planning
EEEAG)
When demand is given, the following costs should be included:
1. Hiring and layoff costs. Hiring costs consist of the recruiting, screening, and training
costs required to bring a new employee up to full productive skill. For some jobs, this
cost may be only a few hundred dollars; for more highly skilled jobs, it may be thousands of dollars. Layoff costs include employee benefits, severance pay, and other associated costs. This cost may also range from a few hundred to several thousand dollars
per worker.
2. Overtime and undertime costs. Overtime costs consist of regular wages plus an overtime premium, typically an additional 50 to 100 percent. Undertime costs reflect the use
of employees at less than full productivity.
3. Inventory carrying costs. Inventory carrying costs are associated with maintaining
goods in inventory, including the cost of capital, the variable cost of storage, obsolescence, and deterioration. These costs often are expressed as a percentage of the dollar
value of inventory, ranging from 15 to 35 percent per year. This cost can be thought of
as an interest charge assessed against the dollar value of inventory held in stock. Thus, if
the carrying cost is 20 percent and each unit costs $10 to produce, it will cost $2 to carry
one unit in inventory for a year.
4. Subcontracting costs. The cost of subcontracting is the price paid to another firm to
produce the units. Subcontracting costs can be either greater or less than the cost of
producing units in-house but typically would be greater than in-house costs.
5. Part-time labor costs. Due to differences in benefits and hourly rates, the cost of parttime or temporary labor is often less than that of regular labor. Although part-time
workers often receive no benefits, a maximum percentage of part-time labor may be
specified by operational considerations or by union contract. Otherwise, there might be
a tendency to use all part-time or temporary labor. However, the regular labor force is
generally essential for operations continuity as well as to utilize and train part-time and
temporary workers effectively.
6. Cost of stockout or back order. The cost of taking a back order or the cost of a stockout should reflect the effect of reduced customer service. This cost is extremely difficult
to estimate, but it should capture the loss of customer goodwill, the loss of profit from
the order, and the possible loss of future sales.
Some or all of these costs may be relevant in any aggregate planning problem. The
applicable costs can be used to “price out” alternative decisions and strategies. In the
example below, the total costs related to three alternative strategies are estimated. When
this type of analysis is conducted on a spreadsheet, a very large number of alternative strategies can be considered.
11.8
LO11.7
AGGREGATE
Create an
alternative strategy
for the Hefty Beer
Company example.
PLANNING EXAMPLE
The Hefty Beer Company is constructing an aggregate plan for the next 12 months.
Although several types of beers are brewed at the Hefty plant and several container sizes
are bottled, management has decided to use gallons of beer as the aggregate measure of
capacity.
The demand for beer over the next 12 months is forecast to follow the pattern shown in
Figure 11.4. Notice that demand peaks during the summer months and is decidedly lower
in the winter.
)] | 236) Part Four
Capacity and Scheduling
FIGURE 11.4
Hefty Beer
Company—demand
forecast.
600 -—4—
(1000)
Gallons
The management of the Hefty brewery would like to consider three aggregate plans:
1. Level workforce. Use inventory to meet peak demands.
2. Level workforce plus overtime. Use 20 percent overtime along with inventory in June,
July, and August to meet peak demands.
3. Chase strategy. Hire and lay off workers each month as necessary to meet demand.
To evaluate
resource data:
these strategies, management
has collected
the following
cost and
e
Assume the starting workforce is 40 workers. Each worker can produce 10,000 gallons
of beer per month on regular time. On overtime, the same production rate is assumed,
but overtime can be used for only three months during the year.
¢
Each worker is paid $4,000 per month on regular time. Overtime is paid at 150 percent
of regular time. A maximum of 20 percent overtime can be used.
¢
Hiring a worker costs $5,000, including screening, paperwork, and training costs. Laying off a worker costs $4,000, including all severance and benefit costs.
¢
For inventory valuation purposes, beer costs $4 per gallon to produce. The cost of carrying inventory is estimated to be 3 percent per month (or 12 cents per gallon of beer
per month).
*
Assume the starting inventory is 50,000 gallons. The desired ending inventory a year
from now is also 50,000 gallons. All forecast demand must be met; no stockouts are
allowed.
The next task is to evaluate each of the three strategies in terms of the costs given.
The first step in this process is to construct spreadsheets like those shown in Tables 11.1
through 11.3, which show all relevant costs: regular workforce, overtime, hiring/layoff,
and inventory carrying. Notice that subcontracting, part-time labor, and back orders/stockouts have not been used as variables in this example.
To evaluate the level strategy, we calculate the size of the workforce required to meet
the demand and inventory goals. Since ending and beginning inventories are assumed to be
equal, the workforce must be just large enough to meet total demand during the year. When
the monthly demands from Figure 11.4 are summed, the annual demand is 5,400,000 gallons. Since each worker can produce 10,000(12) = 120,000 gallons per year, a level workforce of 5,400,000 + 120,000 = 45 workers is needed to meet the total demand. This means
that five new workers must be hired. The inventories for each month and the resulting costs
have been calculated in Table 11.1.
s}soDaie
|&10L ySOD
pessaidxe
Ul
HOAG7/S4IH
AiojuSsAu| HuiAued
Jejnbay
Sul}
SUIUSAO
AiojUaAu|
pus)Jo
(yjuOW
)
O971$
00
09
O'O7ZL$
(oye)
OS
SpUeSNOU}
Jo
‘Sue||OP
I\\7
O'991$
OO
09
O'O07L$
(one)
0S
OOE
OOE
)
PSINpOld
}SEI9104
SwWIEA©O
(%)
SUA
SOILS
OO€
OO€
O€
qed
O€
uer
SIOHIOM
sedinosay
0
0°09
OOvz$
(oye)
O08lL$
00
00S
OOvV
OSY
oxe)
89V
8L7F
8
OCLL$
(oe)
O6E
OED
OSE
O>
ev
JeW
00
vVOS
vCCC$
OCLI$
00
O¢v
OEV
OOvV
()
ev
idy
A99}V.13S—s}s
‘7
[AAI] oy
OVS
OvET$
00
OO8L$
exe)
OSt
OSE
idy
Sv
UONONpOJ
Pue
OLLL$
OSC
09
OOvL$
00
OS
OSE
OSE
0
SE
4eW
Aew
4
Aew
00
OSL
0¢7$
0
OCLLS
00
OOvV
Of€r
OSV
(0)
Ev
OLLZ$
OSz
09
OO8l$
(exe)
0S
OSV
OSY
0
eaunr
OVS
Ovec$
00
OO08lL$
00
OSt
OOS
OSv
0
SV
Aine
aunr
Aine
OLvEe$
OSL
09
0'097$
0'0
0S
0S9
0S9
0
$9
Aine
00
SEE
LST$
DV
OCLI$
OLS
C8C
9LS
0S9
(06
Ev
Bny
0'997$
00c
09
OO0rc$
00
0S
009
009
0
09
6ny
00
SEC
LYTS$
DV
OCLI$
QLS
86l
QOLS
009
(Ya
Ev
Bny
ydas
99v7$
O'S
99
OC6L$
00
GS
O8v
SLYV
0
8vV
ydas
00
vst
vO6L$
OCLL$
00
Esl
OE
SLU
0
Ev
jdas
06
O'681$
00
O'O08L$
00
GL
SLY
OSD
0
SV
po
‘suo||26
Ov
O09
O86L$
O'ssl¢
(ox)
0S
OLD
SLY
)
LV
po
0'0
OEL
OS8L$
OCLI$
00
80L
OEY
SLY
0)
Ev
po
00
09
O98L$
OO8L$
00
OS
SLV
OSv
0
Sv
AlojUSAU!
S!
OO0'0G
Org
O'C6L$
0'0
O'O8lL$
00
OOL
009
OS’
@)
Sv
spuesnou)
JO
‘suo||e6 Bulieys
OLET$
O'Sc
09
0'007$
00
0S
00S
00S
0
OS
dunce
00
66
GELTS$
OCLI$
91S
OL
OLS
00S
(YA
Ev
OO€E
OOLe$
00
O08l$
00
0Sz
0S9
OS’
)
Sv
IIOJYAOAA
YIM IIAQ PWT
009
OOrc$
(exe)
OO8lL$
00
00S
OSY
OSv
0
Sv
Aew
ITOJYIOA
xO
AiojusAu!
SainBy
aleul
OL6L$
OSC
09
O'O09lL$
00
(os)
OOvV
OOv
0
OV
idy
aseYy)—sjso~xPUuBlUa(]
OCLL$
(oe)
00
Hee
60C$
7
aesaissy suluuelg
OSL
SIE,
9°80¢$
OCLL$
(exe)
OLE
O8l
)
JejnBay
FISGVLSLL
Aiojusnu| BuyAued
|E}O] JSOD
T/S4IH SOAP
Jejnboy
auwi}
SUWISAC
PSINpOld
}SEISIOY
AioyUsAu|
Pua)Jo
(yyUOW
OE
OOE
@)
SwISAO
(%)
ev
qed
OCY
OC?l7$
00
OO8L$
00
OEv
OO€
ev
SHU
SS|CS
uer
sadinosay
1ejnBoy
JOM SIS
aese1ssy Suruuelg
OVC
0'677$
A1ojuenu| BuiAues
JEVOL SOD
AISWL
LL Z
OS?
OOsL$
00
OO€E
OSE
OSt
0
Sv
JEW
Ad9}8.13G—s}
“T
[OAV]soD
AON
08
09
OV6L$
O08L$
00
oS
OSY
OSY
0
Sv
AON
OCLI$
00
00
9OL
9C8L$
88
OEV
OS’
0)
Ev
AON
00
09
O98L$
O08l$
00
0S
OSV
OSL
@)
Sv
eq
(one)
09
O981$
OO081$
00
OS
OSY
OSV
0
SV
2eaq
OCLI$
(exe)
(exe)
C8
CTO8lL$
OE
OSV
89
re)
ev
.eq
00
09
O98L$
O'O8L$
00
0S
OSv
OSV
0
Sv
O'S6Z
9CL
9LES9'T$
O'091't$
00
OOS
OOvS
[EOL
O'v907$
SSL
OSL
OL9E
vS6S'7$
8LVS
OOvS
JeVOL
OSTe
O'EIE
Os8rS'c$
O'091'7$
(exe)
OOvS
OOvVS
JEIOL
Capacity Planning
hv
sjsoz
s}so5
OO€
00z
7/341 $$0A0
sejnBay
Sw!)
SUWILSAC
sjso5
JSEISIOY
A1oyUaAu|
pus)Jo
SSIES
(yyUOW
0
OSYv
0
PSoNpOld
SWIUSAQ
(%)
SHU
OSD
Sv
qed
Gv
uer
SIBIOM
sadinosay
aesaissy Suruuelg
1ejnBay
J1ISVLLLL
Chapter 11
237 ||) MA
)} | 238| Part Four
Capacity and Scheduling
Consider the calculations for January. With 45 regular workers entered at the top of the
table, 450,000 gallons of beer can be produced, since each worker produces 10,000 gallons
in a month. The production exceeds the sales forecast by 150,000 (450,000 — 300,000)
gallons, which is added to the beginning inventory of 50,000 gallons to yield an inventory
level of 200,000 gallons at the end of January.
Next the costs in Table 11.1 are calculated as follows: Regular time costs are $180,000
in January (45 workers x $4,000 each). There is no overtime cost, but 5 workers have been
hired, since the starting workforce level was 40 workers. The cost of hiring these five workers is $25,000. Finally, it cost 12 cents to carry a gallon of beer in inventory for a month, and
Hefty is carrying 200,000 gallons at the end of the month, which costs $24,000.' These costs
total $229,000 for January. The calculations are continued for each month and then summed
for the year to yield the total cost of $2,548,000 for the level strategy.
The second strategy, level workforce plus overtime, is a bit more complicated. If X is the
workforce size for option 2, we calculate based on nine months without overtime and three
months with 20 percent overtime, as allowed in the example assumptions.
9(10,000X) + 3[(1.2)(10,000X)] = 5,400,000 gallons
For nine months Hefty will produce 10,000X gallons per month, and for three months it
will produce 120 percent of 10,000X, including overtime. When the above equation is
solved for X, we round off to X = 43 workers. The 43 workers each produce 10,000 gallons per month during nine months. Working 20 percent overtime during the summer,
each produces 12,000 gallons per month during those three months. In Table 11.2 we have
calculated the inventories and resulting costs for this option.
The third strategy, the chase strategy, varies the workforce each month by hiring and
laying off workers to meet monthly demand exactly. When this strategy is used, a constant level of 50,000 gallons in inventory is maintained as the minimum inventory level, as
shown in Table 11.3. The number of workers in January starts at only 30 that are needed
to meet the demand of 300 units. In April, the number of workers is increased to 40 to pro-
duce the demand of 400 units. A peak of 65 workers is reached in August, and so on, until
the number of workers is finally reduced to 45 in December.
The annual costs of the three strategies are summarized in Table 11.4. On the basis of
the given costs and assumptions, the chase strategy is the lowest-cost strategy. However,
TABLE 11.4
Cost Summary
Strategy 1—Level
Regular-time payroll
$2,160,000
Hire/layoff
Inventory carrying
25,000
363,000
Total
$2,548,000
Strategy 2—Level with overtime
Regular-time payroll
Hire/layoff
Overtime
Inventory carrying
Total
'
$2,064,000
15,000
154,800
361,600
$2,595,400
Strategy 3—Chase
Regular-time payroll
Hire/layoff
Inventory carrying
Total
$2,160,000
295,000
72,600
$2,527,600
|
?
For convenience,
we use the end-of-month inventory to calculate inventory carrying costs rather than
the average monthly
1
Chapter 11.
Capacity Planning
/239 ||)
cost is not the only consideration. For example, the chase strategy requires building from
a minimum workforce of 30 to a peak of 65 workers, and then layoffs are used to reduce
the number back to 45 workers. Will the labor climate and availability permit this amount
of hiring and laying off each year, or will this lead to unionization and potentially higher
labor costs, or even shortages in production? Maybe a two-shift policy or the use of parttime or temporary workers should be considered for peak demand portions of the year.
These alternatives, along with others, should be considered in attempting to evaluate and
possibly improve on the chase strategy. Alternatively, the firm may decide to use the level
strategy in view of the small additional cost with no labor disruption.
We have shown how to compare costs in a very simple case of aggregate planning by using
three potential strategies. Advanced analytic methods have been developed to consider many
more strategies and more complex aggregate planning problems. These methods, which can
be quite complex, include linear programming, simulation, and various decision rules.
11.9
KEY POINTS AND TERMS
The following key points are discussed in the chapter:
*
Facilities decisions are long-range in the hierarchy of capacity decisions. A facility
strategy should be implemented rather than a series of incremental facility decisions.
A facilities strategy answers the questions of how much, how large, where, when, and
what type of capacity is needed.
¢
The amount of capacity planned should be based on the desired risk of meeting forecast demand. The capacity cushion is a decision that firms must consider, determining
whether they will use a large cushion and not run out of capacity or a small cushion
with the possibility of a capacity shortage.
¢
When timing capacity expansion, a firm can choose to be preemptive by building capacity sooner or wait and see how much capacity is needed.
*
Both economies and diseconomies of scale should be considered in setting an optimal
facility size. The type of facility selected is focused on product, market, process, or
general-purpose needs.
¢
Sales and operations planning (S&OP), or aggregate planning, serves as the link
between facilities decisions and scheduling. S&OP decisions set output levels for the
medium time range. As a result, decisions regarding workforce size, subcontracting,
hiring, and inventory levels are also made. These decisions must fit within the facilities
capacity available and are constrained by the resources available.
¢
Aggregate planning is concerned with matching supply and demand over the medium
time range. There are many options or factors for managing or changing supply and
demand.
¢
Supply factors that can be changed by aggregate planning are hiring, layoffs, overtime,
undertime, inventory, subcontracting, part-time labor, and cooperative arrangements.
Factors that influence demand are pricing, promotion, backlog or reservations, and
complementary products.
¢
There are two basic strategies for adjusting supply: the chase strategy and the level strategy. Firms can also use a combination of these two main strategies. A choice of strategy
can be analyzed by estimating the total cost of each of the available strategies. Factors,
other than costs, to consider are customer service levels, possible demand changes, the
labor force, and forecasting accuracy.
| | 240) Part Four
Capacity and Scheduling
Terms
Kev
y
LEARNING
ENRICHMENT
(for Sein StuOVias
LOCATE ROI GST Ah)
Economies of sca le 225
Diseconomies of scale 225
Preemptive strategy 226
Wait-and-see strategy 226
Product-focused facilities 227
Market-focused facilities 227
Process-focused facilities 227
General-purpose facilities 227
of capacity
i
Hierarchy
ier
decisions 220
Capacity 221
Utilization 221
Effective capacity 222
Facilities decisions 223
Facilities strategy 223
Capacity cushion 224
Sales and operations
planning 228°
Aggregate planning 228
Product family 228
| Demand management 231
Supply management 232
Level strategy 233
Chase strategy 234
Video
Overview of Capacity Planning
4:47
https://youtu.be/i-pzxlICRJRO
How to Measure Capacity
https://www.allaboutlean.com/production-capacity/
Website
Capacity Planning for Surgical Suites
https://youtu.be/CBDf{METpaH4
Video
9:48
Choosing Global Locations for Capacity
https://www.strategy-business.com/article/10403?gko=e029a
Website
Steps for S&OP Process
Website
http://www.apics.org/apics-for-individuals/
apics-magazine-home/magazine-detail-page/20 1 3/09/09/s-op-step-by-step
Aggregate Planning Supply and Demand Options
Video
https://youtu.be/HOtai_3EdN Y
eo,
SOLVED PROBLEMS
Problem
1. Probabilistic Capacity Planning
The XYZ Chemical Company estimates the annual
demand for a certain product as follows:
Thousands of Gallons
Probability
100|
cies
110
Pee
120
130
140
430)
£30)
©)
. If capacity is 130,000 gallons, how much of a capacity cushion is there?
. What is the probability of idle capacity?
. What is the average utilization of the plant?
2
FT
ao
. If lost
business (stockout) costs $100,000 per thousand gallons and it costs $5,000 to
build 1000 gallons of capacity, how much capacity should be built to minimize total
costs?
Solution
. Capacity cushion = 100% — utilization
= 130 —[(.1 x 100) + (.2 x 110) + (.3 x 120)
+ (.3 x 130) +-¢.1 x 140)]
= 9 thousand gallons, or 9/130 = 6.9% of capacity
Chapter 11.
Capacity Planning
241 | fe
b. Probability of idle capacity = probability of demand < 130
eS een ee)
.6 (or 60%)
c. Average utilization = (.1 X 100/130) + (.2 x 110/130) + (.3 x 120/130)
+ (.3 x 130/130) + (.1 x 140/130)
=193,1%
:
d. To determine the amount of capacity that minimizes total costs, we must determine
the cost of capacity and then add on the penalty cost for not supplying the quantity
demanded. This is done for each amount of possible capacity.
To build 100,000 gallons of capacity:
Total cost = Capacity cost + Penalty cost
= (100 x $5,000) + {$100,000 x [0 x .1) + (10 x .2)
+ (20 x .3) + 30 x .3) + (40 x .1)]}
= $2,600,000
To build 110,000 gallons of capacity:
Total cost = Capacity cost + Penalty cost
= (110 x $5,000) + {$100,000 x [(O0 x .1) + (O x .2)
+(10'x .3) + 20.3) +G0x.1)T}
= $1,750,000
To build 120,000 gallons of capacity:
Total cost = Capacity cost + Penalty cost
= (120 x $5,000) + {$100,000 x [(0 x .1) + (0 x .2) + (0 x .3)
(10 X23)
20x sh)}}
= $1,100,000
To build 130,000 gallons of capacity:
Total cost = Capacity cost + Penalty cost
= (130 x $5,000) + {$100,000 x [(0 x .1) + (0 x .2) + (0 x .3)
+ (0 Xi.3) +.CLOX<1)]4
= $750,000
To build 140,000 gallons of capacity:
Total cost = Capacity cost + Penalty cost
= (140 x $5,000) + {$100,000 x [(O0 x .1) + (O x .2) + (O x .3)
+-(0'%43) + (OX. D]}
= $700,000
The choice that minimizes total costs is to build 140,000 gallons of capacity, which has
an estimated total cost of $700,000.
Problem
2. Services Aggregate Planning
Ace Accounting Associates (AAA) provides income
tax filing services for individuals. Customers pay for the service according to the type
of form they file. Customers with complex forms (i.e., 1040) are charged $200. Customers with simpler forms are charged $50. Five permanent accountants work for AAA at
H | 242| Part Four
Capacity and Scheduling
a rate of $600 per week. During the busy season—the five weeks before the due date
for tax filing—temporary accountants can be hired for $600 per week. A one-time fee
of $200 is paid to an employment service each time the company hires a temporary
accountant. All accountants (permanent and temporary) access a computerized expert
system, which costs the company $175 per accountant per week. On the average, any
accountant (permanent or temporary) can process 4 complex forms per week or 20
simple forms per week. Demand for simple and complex forms for the upcoming tax
season is given in the table below. All demand must be met (serviced) by the end of the
fifth week.
Week
Demand (simple)
Demand (complex)
eae
hes
40
10
60
17
a
aes
4
5
80
21
100
30
100
20
a. Determine the total profit that would result from having enough accountants to meet
demand each week.
b. Find a more profitable arrangement in which all forms will be completed by the end
of the fifth week (demand does not have to be met during the week it occurs).
c. If one or more permanent accountants will work up to an additional 40 hours overtime per week at 1.5 times the regular pay rate for the overtime hours worked, how
will that change your solution to part b?
d. Are there any limitations or assumptions in your answers to parts a, b, and c that
might affect the decisions regarding whom and when to hire or lay off?
Solution
The revenue and permanent employee costs are the same for all the solutions, and so they
are shown here once: .
Week
1
2
3
4
5
2000
2000
4000
3000
3400
6400
4000
4200
8200
5000
6000
11,000
5000
4000
9000
3000
875
3875
3000
So
3875
3000
875
3875
3000
875
3000
875
3875
Ss75
Revenue
Simple forms
Complex forms
Total revenues
Costs of 5 permanent employees:
Accountants
Computer system
Permanent costs
a. First, the number of total accountants needed is computed as follows:
Number of accountants:
Simple forms
Complex forms
Round-up total
2.00
2 |S\0)
3.00
4.25
4.00
325)
5.00
7250
5.00
5.00
4.50
5
7.25
8
O25
10
12250
13
10.00
10
To determine the number of temporary accountants, subtract five permanent accountants from the round-up total.
Chapter 11.
Capacity Planning
243 ||
Costs of Temporary Accountants
Number of Temps
O
3
5
8
5
Hiring fees
Accountants (pay)
Computer system
Temporary costs
Permanent costs
Total costs
Total revenue
O
0)
O
O
SSyp
3875
4000
600
1800
525
2925
3875
6800
6400
400
3000
875
4275
3875
8150
8200
600
4800
1400
6800
38/5
10,675
11,000
O
3000
875
3875
3875
7750
9000
125
(400)
50
325
1250
Profit (loss)
Five-week profit (loss) = $1350
b. In this case we do not need to round up the number of accountants to meet demand
in each period.
Costs of Temporary Accountants
Number of Temps
O
2
4
8
5
Hiring fees
O
400
400
800
O
Accountants
O
1200
2400
4800
3000
Computer system
Temporary costs
O
O
350
1950
700
3500
1400
7000
875
3875
3875
Permanent costs
3875
3875
3875
3875
Total costs
3875
5825
S75
10,875
7750
Total revenue
4000
6400
8200
11,000
9000
Profit (loss)
125
575
825
125
1250
Five-week profit (loss) = $2900
c. Suppose we use only the five temporary accountants
demand in that week by using overtime.
in week 4 and meet total
Costs of Temporary Accountants
Number of Temps
Hiring fees
Temp. accountants
Overtime (perm. accts.)
O
2
4
5
5
O
O
0
400
1200
O
400
2400
O
200
3000
2700
O
3000
0)
Computer system
PatO!
mS5©
700
875
BOv5)
Temporary costs
O
1950
3500
6775
So75
Permanent costs
3875
3875
3875
3875
S875
Total costs
3875
5825
7375
10,650
7750
Total revenue
4000
6400
8200
11,000
9000
Profit (loss)
25
SIS
825
350
1250
Five-week profit (loss) = $3125
P|) | 244) Part Four
Capacity and Scheduling
d. Aggregate planning is rarely as simple as this sample problem. For example, in this
problem, the costs of termination are not considered. It is doubtful that the customers
of this imaginary firm would allow their filings to be put off until it was convenient
for the firm to file them—especially if they were expecting a refund!
Problem
Manufacturers Inc. (MI) currently has a labor
3. Manufacturing Aggregate Planning
force of 10, which can produce 500 units per period (50 units per worker per period).
The cost of labor is now $2400 per period per employee. The company has a longstanding rule that does not allow overtime. In addition, the product cannot be subcontracted due to the specialized machinery that MI uses to produce it. As a result, MI
can increase or decrease production only by hiring or laying off employees. The cost
is $5000 to hire an employee and $5000 to lay off an employee. Inventory carrying
costs are $100 per unit remaining at the end of each period. The inventory level at the
beginning of period 1 is 300 units and ending inventory should be similar. The forecast
demand in each of six periods is given in the table below.
Period
1
2
¢}
4
5
6
Aggregate demand
630
520
410
270
410
520
a. Compute the costs of a chase strategy using only full-time workers.
b. Compute the costs of a level strategy, again with full-time workers.
c. Compare the two strategies.
Solution
a. First, decide on the workforce level to be used for the chase strategy so that production meets demand in each period. For example, in period 1, demand is 630 units,
which requires 12.6 workers (each worker can make 50 units). The answer is rounded
up to 13 workers. The tables for number of units produced and for cost calculations
are as follows:
Period
1
2
3
4
5
6
630
is
650
320
520
10
500
300
410
8
400
290
270
5
250
270
410
8
400
260
520
10
500
240
31,200
°24,00025:19, 2001) 112,000
15,000°'15,000)4-..10,000.4
15:000:
32,000
30,000
29,000
27,000
78,200
69,000
58,200
54,000
249,200
51 50007a1
26,000
60,200
524 000
110.000.
24,000
58,000
Units:
Aggregate demand
Number of workers
Units produced
Ending inventory
Costs:
Cost of labor
Hiring/layoff cost
Inventory carrying cost
Cost per period
Total costs
$377,600
b. For the level strategy, decide first on the workforce level. To make a fair comparison
between the costs of the chase and level strategies and because we round off to whole
numbers of workers, we must produce the same number of total units over six periods. The total units produced in part a are
650 + 500 + 400 + 250 + 400 + 500 = 2700 units
Chapter
11
Capacity Planning
245 ||)
A level strategy produces an equal number of units each period, or 2700/6 = 450 units per
period. This requires exactly nine workers (450/50) each period. The resulting number of
units and cost calculations are
Period
Units:
Aggregate demand
Number of workers
Units produced
Ending inventory
Costs:
Cost of labor
Hiring/layoff cost
Inventory carrying cost
Cost per period
Total costs
2
3
4
5
6
630
3)
520
9
410
9
270
9
410
9
520
9
450
120
450
50
450
90
450
270
450
310
450
240
21,6005
5,000
12,000
38,600
21-600
O
5,000
26,600
2-1. 60035
O
31,000
52,600
221-600
O
24,000
45,600
=2 1:600—-2.1,600"
O
O
9,000
27,000
30,600
48,600
$242,600
c. The level strategy is much less costly. This is the case because it is fairly expensive
to hire and lay off workers.
Discussion Questions
1. Approximately how far ahead would one need to plan
for the following types of facilities?
Restaurant
. Hospital
Oil refinery
Toy factory
Electric power plant
0)
Ph:
ps
Kot
E.G!
Public school
g. Private school
2. What problems are created by simultaneously considering the capacity questions of how much, how large,
where, when, and what type?
3. A school district has forecast student enrollment for
several years into the future and predicts excess capacity for 2000 students. The school board has said that the
only alternative is to close a school. Evaluate.
4. Why are facilities decisions often made by top management? What is the role in these decisions of operations,
marketing, finance, accounting, information systems,
engineering, and human resources?
5. In what ways does corporate strategy affect capacity
decisions?
6. S&OP or aggregate planning sometimes is confused
with scheduling. What is the difference?
7. The XYZ Company manufactures a seasonal product.
At the present time, the company uses a level labor
force as a matter of company policy. The company
is afraid that if it lays off workers, it will not be able
to rehire them or find qualified replacements. Does
this company have an aggregate planning problem?
Discuss.
. It has been said that aggregate planning is related to
personnel planning, budgeting, and market planning.
Describe the nature of the relationship among these
types of planning.
Firms often have multiple objectives such as good labor
relations, low operating costs, high inventory turnover,
and good customer service. What are the pros and cons
of treating these objectives separately in an aggregate
planning problem versus combining them all into a
single measure of cost?
10. What factors are important in choosing the length of the
planning horizon for aggregate planning?
Ie A barbershop has been using a level workforce of barbers five days per week, Tuesday through Saturday. The
barbers have considerable idle time on Tuesday through
Friday, with certain peak periods during the lunch hours
and after 4 p.m. each day. On Friday afternoon and all
day Saturday, all the barbers are very busy, with customers waiting a substantial amount of time and some
customers being turned away. What options should
this barbershop consider for aggregate planning? How
would you analyze these options? What data should be
collected, and how should the options be compared?
EG
Part Four
Capacity and Scheduling
Problems
Three Excel spreadsheets are provided on Connect for
assistance in solving the chapter problems.
Ike Suppose we are considering the question of how much
capacity to build in the face of uncertain demand.
Assume that the cost is $20 per unit of lost sales due to
insufficient capacity. Also assume that there is a cost
of $7 for each unit of capacity built. The probability of
various demand levels is as follows:
Demand—X
Units
Probability of X
WN—O
NOOR
~
05
10
aS
.20
.20
aS
10
05
How many units of capacity should be built to minimize the total cost of providing capacity plus lost sales?
b. State a general rule regarding the amount of capacity
to build.
c. What principle does this problem illustrate?
. The Ace Steel Mill estimates the demand for steel in
millions of tons per year as follows:
Millions of Tons
10
12
Probability
10
25
14
.30
16
18
20
15
a. If capacity is set at 18 million tons, how much of a
capacity cushion is there?
b. What is the probability of idle capacity, and what is
the average utilization of the plant at 18 million tons
of capacity?
c. Ifit costs $8 million per million tons of lost business
and $80 million to build a million tons of capacity, how
much capacity should be built to minimize total costs?
A barbershop has the following demand for haircuts on
Saturday, which is its busiest day of the week.
Number of Haircuts
20
25
30
35
40
Probability
1
cS
4
1
1
a. What is the average demand for haircuts on Saturday?
b. If the capacity is 35 haircuts, what is the average utilization of the shop?
c. If capacity is 35 haircuts, how much of a capacity
cushion does it have?
d. If it costs $50 per lost haircut due to customer dissatisfaction and $100 for each unit of capacity
provided, how much capacity should be built to
minimize costs?
4. Assume a restaurant operates from 11 a.m. to 11 p.m.
seven days per week.
a. How much capacity does the restaurant have on a
weekly basis and an annual basis, in hours?
b. If the restaurant can serve a maximum of 40 customers per hour, how much weekly capacity does the
restaurant have in terms of customers?
c. What implicit assumptions are made in your calculation for part b?
. An urgent care clinic is staffed by two physicians who
can each see four patients per hour. The clinic is open
6 p.m. until midnight, seven days per week. The clinic
tracked the average number of patients arriving by hour
for a month and observed the following:
Time
Demand
6-7
7-8
8-9
9-10
10-11
11-12
8
10
10
4
4
2
a. Sketch a graph with one line showing capacity and
another line showing demand.
b. What observations do you make from the graph in
part a?
c. What suggestions would you make to the clinic for
managing its capacity?
eXce | 6. The Chewy Candy Company would like to
determine an aggregate production plan for the
next six months. The company makes many different
types of candy but feels it can plan its total production
in pounds provided that the mix of candy sold does not
change too drastically. At the present time, the Chewy
Company has 70 workers and 9000 pounds of candy
in inventory. Each worker can produce 100 pounds of
candy per month and is paid $19 per hour (use 160
hours of regular time per month). Overtime, at a pay
rate of 150 percent of regular time, can be used up to a
maximum of 20 percent in addition to regular time in
Chapter 11.
Capacity Planning
247 | fy
a worker. The forecast for the next six months are 8000,
. Analyze these options from the standpoint of an aggregate planning problem. What are the pros and cons of
each option?
10,000, 12,000, 8000, 6000, and 5000 pounds of candy.
. Which option do you recommend? Why?
a. Determine the costs of a level production strategy
for the next six months, with an ending inventory of
8000 pounds.
. How does this problem differ from the other aggregate
planning problems above?
any month. It costs 80 cents to store a pound of candy
for a year, $1,200 to hire a worker, and $1,500 to lay off
b. Determine the costs of a chase strategy for the next
six months.
. The Restwell Motel in Orlando, Florida, is preparing
an aggregate plan for the upcoming 12 days. The motel
has a maximum of 200 rooms, but demand varies dur-
c. Calculate the costs of using the maximum overtime
for the two months of highest demand.
ing the week. Demand is listed in terms of rooms rented
each day. The motel requires one employee, paid $105
per day for each 12 rooms rented. It can utilize up to
20 percent overtime at 150 percent pay and also can
hire part-time workers at $120 per day. Each part-time
worker can also clean 12 rooms per day. There is no
hiring and layoff cost for the part-time workers. A frac-
. A company has seasonal demand, with the forecast for
the next 12 months as given below. The current labor
force can produce 500 units per month. Each employee
can produce 20 units per month and is paid $2,000 per
month. The inventory carrying cost is $50 per unit per
year. Changes in the production quantity cost $100 per
unit (in months when production volumes change) due
tional number of part-time workers (e.g., 3.4 workers)
can be employed since less than a full day of employment is possible for each worker.
to hiring or layoffs, line changeover costs, and so forth.
Assume 200 units of initial inventory.
a. What is the cost of carrying inventory for the month
of January for the level strategy?
b. What is the total cost of the level strategy including
regular time, inventory carrying cost, and changes in
production level?
c. What is the total cost of the chase strategy?
Day
Meese
Demand
185
W Selinger
190 170
Seat Sute Mie ele
160 110 100
100
160 180
Veco nee
170 150
100
a. Assume a steady regular workforce of 10 employees,
20 percent overtime when needed, and the balance
of demand met by part-time workers. How much
does this strategy cost over the 12-day planning
horizon? When needed to meet demand, assume
Month
J
Demand
651
F
Mi
A
Mi
ed
rs
S 20 —N5)'D
700 850 702 650 500 600 850 803 900 703 600
8. Approximately 40 percent of a medical clinic’s weekly
incoming calls for appointments occur on Monday. Due
to this large workload, 20 percent of the callers receive
a busy signal and have to call back later. The clinic has
one clerk for each two departments to handle incoming
calls. Each clerk handles calls for the same departments
all week and thus is familiar with the doctors’ hours,
scheduling practices, and idiosyncrasies. Consider the
following alternatives to solve this problem:
Continue the present system, which results in some
customer inconvenience, loss of business, and perceived
poor service. About 1000 patients attempt to call the
clinic on Mondays. The clinic has 50,000 patients in all.
Expand the phone lines and add more people to handle
the peak load. The estimated cost for adding two more
lines and two clerks is $60,000 per year.
Install new software to speed up making appointments.
In this case, the peak load could be handled with the
current personnel. The estimated cost to lease and
maintain the software is $50,000 per year.
Expand the phone lines and ask people to call back later
in the week for an appointment. Add two lines and two
phone-answering clerks part time at $30,000 per year.
maximum overtime is used before part-time workers
are hired.
b. What is the total cost over the 12-day planning
horizon if the regular workforce of 10 employees
and only part-time workers are used? No overtime is
used?
eXce / 10. The Bango Toy Company produces several types of toys. The forecast for the next six
months in thousands of dollars is given below:
July
Aug.
Sept.
Oct.
Nov.
Dec.
Forecast $1000 $1500 $2000 $1800 $1500 $1000
A regular employee can produce $10,000 worth of toys
per month, and the company has 80 regular employees
at the end of June. Regular-time employees are paid
$3800 per month, including benefits. An employee on
overtime produces at the same rate as on regular time
but is paid at 150 percent of the regular pay. Up to 20
percent overtime can be used in any one month. A
worker can be hired for $1000, and it costs $2000 to lay
off an employee. Inventory carrying costs are 30 percent per year. The company wishes to end the year with
80 employees. Beginning inventory of toys is $900,000.
a. Calculate the cost of a chase strategy.
b. Calculate the cost of a level strategy.
) | 248 | Part Four
Capacity and Scheduling
c. Using the Excel template, simulate several other
strategies.
d. Determine the effect on the chase strategy, in part
a, of changing the hiring cost to $1500, $2000, and
$2500. What do these changes suggest the relationship is between hiring cost and total cost?
e. Use the Excel template to study the effect of demand
changes on the total cost of the chase strategy.
Assume various percentage increases and decreases
in demand (110 percent, 120 percent, 210 percent,
220 percent, etc.).
11. A small textile company makes several types of sweaters. Demand is very seasonal, as shown by the following quarterly demand estimates. Demand is estimated in
terms of standard hours of production required.
Forecast
Fall
Winter
Spring
Summer
OOOO
=1:5:000
8000
5000
An hour of regular time costs the company $12.
Employees are paid $18 per hour on overtime, and labor
can be subcontracted at $14 per hour. A maximum
of 1000 overtime hours is available in any month. A
change in the regular level of production (increase or
decrease) incurs a one-time cost of $5 per hour for adding or subtracting an hour of labor. It costs 2 percent per
month to carry an hour of finished work in inventory.
Materials and overhead costs in inventory are equal to
the direct labor costs. At the beginning of the fall quarter, there are 5000 standard hours in inventory and the
workforce level is equivalent to 10,000 standard hours.
a. Suppose management sets the level of regular workers for the year equal to the average demand and
subcontracts out the rest. What is the cost of this
strategy?
b. What is the cost of a chase strategy?
eXce/
12. Beth’s Broasted Chicken shop offers a
variety of fast-food items. Beth uses regular
and part-time workers to meet demand. The demand for
the next 12 months has been forecast in thousands of
dollars, as follows:
Joon
Aca Min Jind ~Ate Sh OrNanD
Demand
25 33 40 57 50 58 50 48 37 33 28 32
Assume that each employee can produce $5000 worth
of demand in a month. The company pays regular
workers $10 per hour, including benefits, and part-time
workers $7 per hour. There are 167 working hours in
a month. Management would like to use as many parttime workers as possible but must limit the ratio to one
regular worker to no more than one part-time worker
to provide adequate supervision and continuity of the
workforce. A fractional number of part-time workers
(e.g., 2.6 workers) can be employed since less than a
full day of employment is possible for each worker. For
example, if 10.6 total workers are needed, 6 regular
workers and 4.6 part-time workers is a feasible combination, but 5 regular workers and 5.6 part-time workers
is not, because the part-time workers exceed the fulltime workers. Demand must be met on a month-bymonth basis. It costs $500 to hire and $200 to lay off a
regular worker. No costs are associated with hiring and
laying off part-time workers.
a. Develop a strategy for this problem by using the
minimum integer number of regular workers along
with the maximum amount of part-time workers in
each month. What is the total cost of this strategy
including regular-time labor, part-time labor and hiring/firing cost?
b. What is the total cost of a strategy with a constant
level of 6 full-time workers used in each month?
Excess labor is allowed when demand is less than
six full-time workers.
Valley View Hospital faces somewhat seasonal demand.
As a result, the forecast of patient days of demand is as
follows (a patient day is one patient staying for one day
in the hospital):
Forecast
Fall
Winter
Spring
Summer
90,000
70,000
85,000
65,000
The hospital uses regular nurses, part-time nurses
(when the hospital can get them), and contract nurses
(who are not employees). Contract nurses work a number of hours, which varies depending on their contract
established with the hospital. Regular nurses are paid a
sum of $15,500 per quarter for 60 days of work; parttime nurses are paid $6500 per quarter for 30 days of
work. Contract nurses get an average of $17,200 per
quarter for 60 days of work. It costs $1000 to hire or lay
off any of these three types of nurses.
Suppose that regular nurses are set at a level of
800 nurses for the year. Each regular nurse works the
equivalent of 60 days per quarter. The remainder of
the demand is made up by 50 percent part-time and
50 percent contract nurses on a quarter-by-quarter
basis. What is the cost of this plan starting at the
beginning of fall with 800 regular nurses, 200 parttime nurses, and 200 contract nurses? Assume it takes
0.8 nurse day to provide around-the-clock care for
each patient day.
Scheduling Operations
Sey
s
a
&
LO12.2
Construct a Gantt chart.
LO12.4
Explain the implications of the theory of constraints for scheduling.
LO12.5
Describe the important factors to consider when designing a scheduling system.
A teaching hospital must create a schedule for physicians in resident training that assigns
them to overnight call shifts across three hospitals over a 365-day planning horizon. The work
shifts must meet constraints such as assigning residents to six consecutive weeks at a particular hospital and not assigning them to be on call more than once every four days. A manufacturing manager in a bottling plant must schedule a variety of beverages to allow intensive
cleaning setups between batches while also considering how much inventory to hold.
Organizations in every industry must schedule work to meet demand for products and services while supporting longer-term strategies. Scheduling decisions allocate available capacity or resources (equipment, labor, and space) to jobs, activities, tasks, or customers over
time. Since scheduling involves allocation decisions, it uses the resources made available by
facilities decisions and aggregate planning. Therefore, scheduling consists of short-term decisions that are constrained by previous decisions regarding facilities and aggregate planning.
In practice, scheduling results in a time-phased plan of activities, indicating what is to be
done, when, by whom, and with what equipment. Scheduling should be clearly differentiated
°
ey
ad
9
9
° °
o
o
ants
Pe
9
°
249||
=
. | | 250| Part Four
Capacity and Scheduling
from aggregate planning. Aggregate planning seeks to determine the resources needed,
while scheduling allocates the resources made available through aggregate planning in the
best manner to meet operations objectives. Aggregate planning is done on a time frame of
about one year; scheduling is done on a time frame of a few months, weeks, or hours.
Scheduling seeks to achieve potentially conflicting objectives: high efficiency, low
inventories, and good customer service. Efficiency is achieved by a schedule that maintains high utilization of labor, equipment, and space. Of course, the schedule should also
seek to maintain low inventories; unfortunately, this may lead to low efficiency due to
lack of available material or high setup times. Thus, a trade-off decision in scheduling
between efficiency and inventory levels is required in the short run. In the long run, however, efficiency can be increased, customer service improved, and inventory simultaneously reduced by changing the production process itself through cycle time reduction and
quality improvement efforts. Scheduling, then, is primarily an activity that involves potential trade-offs between conflicting objectives in the short run.
Because of the conflicting objectives, all functional areas are interested in scheduling.
Marketing wants to make sure the most important customers are scheduled first. Finance
and accounting want to be sure that the schedule is cost-efficient and makes the best use of
available resources. Operations is often at the intersection of a cross-functional scheduling
challenge that requires coordination across all the business functions.
12.1
LO12.1
BATCH SCHEDULING
Describe
the concept of batch
scheduling.
For scheduling batch operations, much of the terminology (“shop,” “job,” and “work center’) comes from manufacturing job shops. The concepts, however, apply to batch operations of all types, including factories, hospitals, offices, and schools. For service operations,
99 66
99 66
“job” can be replaced by “customer,” “patient,”
“client,” “paperwork,” or whatever type of
work flows through the process. Furthermore, “work center” can be replaced by “room,”
“office,” “facility,” “skill specialty,” or whatever the processing centers are. In this way, the
concepts can be generalized to all types of operations.
A few examples of batch scheduling might be helpful. In a university students are processed
in batches, with the batch size equal to the number of students in a class. Batch scheduling
consists of assigning classes to classrooms and instructors. For manufacturing Louisville Slugger aluminum bats, batch sizes of 100 are used to make a variety of models. Different models
require processing at different work centers. For example, some bats have a metal end produced
by heating and rotating the bat while others receive a plastic end-cap that is snapped into place.
Batch scheduling, the plan for assigning work to the necessary resources, is a very complex
management problem. First, each batch flowing through a batch process typically moves along
with many starts and stops, not smoothly. This irregular flow is due to the layout of the batch process by machine group or skills into work centers. As a result, jobs or customers wait in line as
each batch is transferred from one work center to the next, and work-in-process (WIP) builds up.
The batch scheduling problem can be thought of as a network of queues. A queue of
WIP inventory is formed at each work center as jobs wait for resources to become available. These queues are interconnected through a network of material or customer flows.
The problem in scheduling batch processes is how to manage these queues.
One of the characteristics of a batch operation is that jobs or customers spend most of
their time waiting in line. The amount of time spent waiting will, of course, vary with the
load on the process. If the process is highly loaded (high utilization rate), a job may spend
as much as 95 percent of its total production time waiting in queues. Under these circumstances, if it takes one week to actually process an order, it will take 20 weeks on average to
move completely through production. However, if the process is lightly loaded, the waiting
Chapter12
Scheduling Operations
251] |)
Soccer Scheduling Goal!
Soccer is far and away the most popular sport in Brazil.
Each year, the Brazilian Football Confederation faces the
challenge of scheduling its 20 teams in an eight-month
tournament to determine a champion. A fair and balanced
schedule
is important, both to maximize
revenue
and to instill confidence in the tournament outcome.
The goal is to schedule all teams in the tournament
while meeting more than 30 different constraints. Exam-
ples of constraints are:
*
Each team faces each other team twice, playing once
at home and once away.
*
Every pair of teams plays once in the first half of the
tournament and once in the second half.
*
Elite teams must play as often as possible on weekends rather than weekdays.
*
Teams
home
city must have comple-
mentary patterns (i.e., when
with the same
one plays at home, the
other plays away).
*
No
team
plays
more
than
two
consecutive
home
games or two consecutive away games.
Numerous
Koji Aoki/Aflo/Getty Images
other constraints include the rights of certain
players, league officials, media sponsors, and city administrators to approve or reject a proposed schedule.
Recently, the league began using software to assist
with scheduling. Formulating the scheduling problem
while including all constraints was computationally
:
;
:
Hn eesI ie
5
SOUL SONING
TOF Sole a
OneioMmts
iter oe,
then using feasible solutions to resolve other constraints
proved fruitful. The league is able to identify
a schedule
that met all constraints while allowing nonconflicting TV
broadcasts of the most attractive games, particularly
important for revenue purposes. However, since league
rules prohibit repeating a previously used schedule,
every year’s tournament requires a new schedule!
cis
:
Source: Adapted from Celso C. Ribeiro and Sebastian
Urrutia, “Soccer Scheduling Goaaaaal,’ OM/MS Today, April
— 20910, pp. 52-57; and https://www.capterra.com/sports—league-software/, 2019.
time will be reduced since all the jobs will flow through the process more rapidly. Regardless of process load, the challenge is to develop scheduling procedures that will effectively
manage the flow of jobs, customers, and work.
We continue this chapter with a discussion of Gantt charts, a simple form of scheduling. This
is extended to more complex and realistic settings with discussions of finite capacity scheduling and the theory of constraints. The chapter is completed with dispatching rules and examples
of scheduling systems used in actual practice. The Operations Leader box on the scheduling of
games in a soccer league provides an interesting example of scheduling in practice.
12.2
GANTT CHARTS
LO12.2 Constructa
Gantt chart.
One of the oldest scheduling methods is the Gantt chart. Although there are many variations of the Gantt chart, we restrict its use in this chapter to the batch scheduling problem.
The Gantt chart is a table with time across the horizontal dimension and a limited
resource, such as machines, people, or machine hours, along the vertical dimension. In the
example below, we assume that machines are the limited resource to be scheduled.
| | 252| Part Four
Example
Capacity and Scheduling
Suppose
we have three work centers (A, B, C) consisting of one
machine
each
and five
jobs (1, 2, 3, 4, 5) to be scheduled. The processing time of each job in each work center
is shown in Figure 12.1 along with sequencing of the jobs through the work centers. For
example,
job 2 is processed in work center C for six hours followed by four hours in work
center A. This is denoted by C/6, A/4 in Figure 12.1.
The jobs are scheduled forward in time within the finite capacity of one machine of
each type (A, B, and C). We assume, arbitrarily, that the jobs should be scheduled in the
sequence
1, 4, 5, 2, 3.
The Gantt chart resulting from these assumptions is shown in Figure 12.2. This chart is
constructed by first scheduling job 1 on all three machines. Job 1 starts on machine A for
two hours; then it is placed on machine B for three hours (time 2 to 5); finally, it is processed
on machine C for four hours (time 5 to 9). There is no waiting time for the first job scheduled, since there can be no job interference or waiting. Next, according to the assumed
sequence,
job 4 is scheduled on the Gantt chart for machines C, B, and then A. Job 4 can
begin immediately on machine C, since the machine is open until time 5 and only four
hours is needed. After a waiting time of one hour, job 4 can start on machine B. Job 4 can
then be scheduled on machine A from time 8 to 11. Next, job 5 is scheduled on the Gantt
chart. Job 5 is processed on machine A first, but machine A is already scheduled until time
2. So job 5 starts at time 2 and is completed by time 7. After completion on machine A,job
5 is moved to machine B, which is busy until time 8. Job 5 is then scheduled from time 8 to
11 on machine B. Jobs 2 and 3 are then scheduled.
FIGURE 12.1
Job data for scheduling.
Job
DAS SIONS
Machine hours
Due
date
A/2, B/3, C/4
C/6, A/4
B/3, C/2, A/1
C/4, B/3, A/3
A/S, B/3
FIGURE
12.2
Gantt chart. Jobs are
scheduled in sequence
ey
2
id
©
ROMER
78 @u
Rae Roe araie
15
ipa
19@20
ary: -
Cele
FIGURE 12.3
Machine idle and job
waiting times.
Maching
idle (hr)
Job waiting
Delivery
time (hr)
time (hr)
Chapter 12
VISIBLE SCHEDULES.
,
A good schedule must be available
for all employees to see.
Ingram Publishing
Scheduling Operations
253 ||
After the Gantt chart has been constructed, it should
be evaluated with respect to both job and machine performance. One way to evaluate machine performance is
on the basis of the time it takes to complete all work, the
makespan. In Figure 12.2, the makespan is 20 hours,
since it takes 20 hours to complete all five jobs.
Another measure of Gantt chart performance is machine
utilization. In the five jobs, a total of 43 hours of processing time is required (simply add machine times for all jobs
from Figure 12.1). Utilization is 43/60 = 71.7 percent,
because a total of 60 hours of machine time is available
(makespan 20 x 3 machines). The 43 hours of processing
time is a constant regardless of the schedule used. Notice
:
;
Lee
also that idle time = 3 x (makespan) — 43. Therefore, mini-
mizing makespan will also minimize machine idle time.
A measure of job performance is the sum of the waiting times for all jobs. In Figure 12.3, the delivery times and job-waiting times are listed
for each job. These quantities are obtained directly from the Gantt chart. The delivery time
and waiting times will, of course, depend greatly on the job sequence used. Since job | was
scheduled first, it has no waiting time and is completed ahead of the due date. Jobs 2 and 3,
which were scheduled last, have considerable waiting time.
In general, the waiting times of jobs and machine utilization are highly dependent on
the sequence of the jobs scheduled. In this case, we have five jobs to schedule and 5! (five
factorial) = 120 possible sequences of jobs. If we construct 120 Gantt charts, one for each
possible sequence, we could determine the job sequence with the minimum makespan,
or the one with the minimum total job waiting time. In general, for n jobs there will be n/
possible sequences to evaluate to find the optimal sequence by complete enumeration of
all the possibilities. Complete enumeration is often cumbersome or impossible in practical
situations in which there may be several hundred or more jobs.
Optimal job-sequencing algorithms attempt to find an optimal solution without enumeration of all possible job sequences. One particular problem 1s called the m X n machinescheduling problem, where m is the number of machines and n is the number of jobs.
Optimal solutions can be found for only relatively small values of m and n. Fairly good
heuristics are available to develop good solutions for any values of m and n.
In summary, the following conclusions about batch scheduling can be drawn from Gantt
chart scheduling:
1. Schedule performance (makespan, job waiting times, job delivery times, machine utilization, and inventory level) is highly sequence dependent (which job is scheduled first,
second, third, etc.).
2. The waiting time of a job depends on the job interference encountered in the schedule
and the capacity available on machines.
3. Finding the optimal schedule is computationally intensive and cannot be done for
most practical size applications. However, good heuristic procedures are available that
closely approximate the optimal schedule.
Batch scheduling methods have many applications in both manufacturing and service
industries. A service example is the scheduling of patients in a hospital. Patients flow
through the hospital and require the services of multiple resources (people and equipment).
At one cancer treatment unit, for example, waiting time for appointments for radiology
|
254| Part Four
Capacity and Scheduling
treatments was reduced from 40 days to 16 days despite a 14 percent increase in workload.
The new scheduling system allowed the radiology staff to schedule all sequential treatments and other linked appointments. This also meant that patients could see the full extent
of their treatment plan at the outset.!
12.3.
LO12.3
FINITE CAPACITY SCHEDULING
Create
work schedules
using forward
and backward
scheduling.
Finite capacity scheduling (FCS) is an extension of Gantt charts. It establishes a work
schedule to be produced in a certain time period, considering relevant limitations of
resources. Software is often used to determine a good schedule with efficient flow, while
accounting for trade-offs of inventory levels and customer service.
FCS schedules jobs through a number of work centers, each with one or more machines
that perform the same function. Jobs can pass each other, or change their sequence, as they
are processed, depending on their priority. A job can be split if this will facilitate scheduling. For example, if the job consists of making 100 parts, the job could be split into two
lots of 50 parts each to help the schedule. Alternate routings of a job through the work
centers may also be allowed. In FCS, attention is paid to scarce resources to facilitate job
flow and improve the performance of the shop. Let’s take a look at a simple example which
is an extension to FCS of Figure 12.1. See the highlighted example box below.
This addition of capacity in the form of additional machines substantially improves the
completion dates of jobs and reduces their waiting times. The makespan is now 11 hours
instead of 20 as before. This type of forward scheduling provides a feasible estimate of
a completion date for current orders. Alternatively, backward scheduling can be used
to assign orders to machines or other resources, working backward from a due date to
determine when work must be started in order to be completed by the due date. Backward
scheduling is used when meeting due dates is more important than machine efficiency.
Aside from forward or backward scheduling, it is important to understand the effect of
bottlenecks on scheduling. The formal definition of a bottleneck is a work center whose
capacity is less than the demand placed on it and less than the capacities of all other resources.
A bottleneck resource will constrain the capacity of the entire shop, and an hour added to a
bottleneck will add an hour of capacity to the entire plant. An hour added to a non-bottleneck
work center will not help the schedule at all, since excess capacity already exists there.
Example
Assume the same situation as in Figure 12.1 that was used for Gantt charting, except now
there are two machines of each type in each work center. In other words, there are two
machines of type A, two of type B, and two of type C.
We can now construct a Gantt chart with this new information, as shown in Figure 12.4.
In this figure, the two machines of type A are labeled A, and A,. We use the same sequence
of scheduling as before, with jobs scheduled in the sequence 1, 4, 5, 2, 3. The Gantt chart
will not change for job 1, since this job was scheduled without job interference and no
job waiting time. Next, job 4 can be completed one hour earlier by the additional machine
added to each work center and is scheduled on machine C,, followed by B, and A,. Previously job 5 could not start immediately because machine A was already scheduled. Now
that we have two machines, job 5 is scheduled on machine A, followed by B,. Next we
schedule
job 2, which is benefited substantially by the addition of a second machine of
type C; it can be scheduled to start immediately. Also job 3 can start earlier than before.
'N. Huber, “Scheduling System Slashes Radiotherapy Waiting Times by 60% at Wirral Cancer
Unit,”
Computer Weekly, May 4, 2004, pp. 39-40.
Chapter12
4ee
o
E
es
sp =
:
Scheduling Operations
255] ||
‘) |ene
Te ga
=
S)
S
2
FIGURE
12.4
Work center Gantt chart (jobs are sequenced in order
lis Gey Ashe
Machine center
idle (hr)
Job waiting
Delivery
time (hr)
time (hr)
9
10
11
10
8
i)
i
In the example in Figure 12.4, work center C is a bottleneck from time 3 to time 6 since
job 3 is waiting to be processed.2 Demand exceeds capacity, and no other work center
provides a constraint at that time. Work center A is a bottleneck from time 8 to time 10
for similar reasons, and job 3 is again waiting to be processed. Note, the bottleneck is not
constant; it shifts from one work center to another.
Scheduling can be improved by adding capacity at the bottleneck work center. If this
were done in the example in Figure 12.4, the makespan could be further reduced from
11 hours to 10 hours, the minimum makespan because job 2 has 10 hours of processing
time. Capacity can be added in many ways: by adding machine time, reducing setup time,
adding overtime, and subcontracting, to name a few. An important principle in scheduling
is to find the bottlenecks and work to remove them by improving flow through the bottleneck resources. FCS can be used to identify the bottlenecks at any point in time.
One way that managers can improve throughput at the bottleneck is to increase the flexibility of capacity resources; for example, using equipment that can complete a variety of
tasks or cross-training workers so they can work where they are needed most at any given
time. The burden on the bottleneck can be eased by shifting its workload to other machines
or workers. The resources targeted for increased flexibility must be closely coordinated
with the job schedule.
? We assume for this example that job 3 cannot be split into two or more lots for processing.
| | 256| Part Four
12.4
LO12.4
Capacity and Scheduling
THEORY OF CONSTRAINTS
Explain the
implications of the
theory of constraints
for scheduling.
In his popular book The Goal,* Eliyahu Goldratt argues that making money from operations
can be broken down into three measurable quantities: throughput, inventory, and operating
expenses. He defines these terms in rather nontraditional ways. Throughput is defined as the
sales of the plant minus the cost of raw materials used to produce those sales. It is not enough
just to make a product; it must be sold to the customer in order to make money. So, if operations has excess capacity, the task of operations is to help the sales department increase sales
(and thus throughput). In contrast, if the plant is operating at capacity, it must push orders
through the plant faster to increase throughput. This is done by identifying the bottleneck
operations in the plant and increasing the capacity of bottlenecks, often without buying more
equipment but through more creative scheduling, overtime, better workforce policies, and so
forth. Goldratt calls this the theory of constraints (TOC) since the most important constraint,
either sales or the production bottleneck, is being relieved in order to increase throughput.
A plant must also reduce inventory to make money. Goldratt defines inventory as only
the raw-material value of any goods being held in inventory, a rather unconventional definition. He puts all labor and overhead costs into the operating expense category not into
inventory. Operating expense is the cost of turning raw materials into throughput. His logic
is that true costs are distorted by putting labor and overhead into inventory on the assumption that the inventory will be sold, when in fact no money is made for the company until
the inventory actually is sold.
TOC is based on several principles observed in organizations:
¢
Companies, departments, and teams have unbalanced capacities. Annual
attempt to provide resources to balance capacities, but rarely succeed.
budgets
¢
There is always a constraint in the system somewhere in the plant, human resources,
purchasing, or sales that prevents the company from making more money.
¢
One hour of capacity lost at the constraint (bottleneck) is an hour lost to the whole organization and can never be recovered.
¢
An hour gained at a nonconstraint does not add to the output or profit of the organization.
¢
Constraints must be managed differently than nonconstraints.
TOC has many implications for scheduling. First, the bottleneck is the critical resource
and constraint that should be scheduled to achieve maximum throughput. As was noted
above, each hour of capacity gained at the bottleneck is an hour gained for the entire plant.
All non-bottleneck resources should be scheduled so that the bottleneck is not starved
(waiting) for materials and is kept busy processing orders needed for sale. Also, a queue
of jobs should be formed in front of the bottleneck resource to ensure that it stays busy.
Non-bottleneck resources do not need to operate at full capacity provided that they process
enough to keep the bottleneck busy. Thus, some of the non-bottleneck work centers may
have idle time in their schedule. Non-bottleneck resources should not produce inventory
just to increase resource utilization. They should be idle when their capacity is not needed
to supply the bottleneck. See Figure 12.5 for an illustration of the bottleneck constraint and
non-bottleneck resources.
Many steps can be taken at the bottleneck resource to increase capacity. These steps include
reduction of setup time so that the work center can be quickly changed from one job to the
next. Another step is to be sure that the bottleneck resource is used as much as possible and is
not shut down for breaks, lunch, or even maintenance that can be deferred. Resources should
* Eliyahu M. Goldratt and Jeff Cox, The Goal: A Process of Ongoing Improvement, 3rd ed. (Great
Barrington, MA: North River Press, 2014).
Chapter12
Scheduling Operations
257] |
FIGURE 12.5
The bottleneck
constrains the flow of
the entire system.
be added to the bottleneck work center via additional labor or machines on a temporary basis
if possible. When this is done, the operation will progress toward its goal of making money.
In summary, there are four steps to eliminate constraints:
1. Identify the system constraint that prevents the company from making more money.
2. Decide how to reduce the system’s constraint, so it is no longer a constraint.
3. Subordinate everything else (other tasks, work centers,
departments) to reducing the constraint.
sales, purchasing or other
4. Once the constraint is eliminated, find the next constraint and start over.
An example might be helpful here. Delta Airlines’ maintenance, repair, and overhaul
(MRO) operations used TOC to improve the scheduling of jobs based on their capacity
constraint—the repair and support shops. Based on the rate of work done in the repair and
support shops, both upstream and downstream work processes were paced to match the
bottleneck rate. This helped to reduce inventory in all related work centers. Using TOC,
Delta MRO increased throughput by 18 percent and decreased inventory by 50 percent.
Traditional cost accounting attempts to maximize the utilization of all resources and work
centers even if they build inventory that is not needed and even if the work centers are not
bottlenecks. TOC argues that non-bottleneck resources can remain idle some of the time, provided that they do not constrain the bottleneck. Maximizing the efficiency of each resource,
or reducing their standard cost variances to zero, does not make more money for the company.
Firms should consider the following ways to reduce bottlenecks:
*
Can more capacity be added at the bottleneck, even if it is less efficient, or uses anti-
quated equipment or expensive overtime, or is outsourced to a vendor?
*
Can the work that does not absolutely need to go through the bottleneck be diverted to
another non-bottleneck resource?
¢
Can work that has poor quality and will be scrapped later be prevented from reaching
the bottleneck?
*
Can the output of the bottleneck be increased by running larger batches or reducing the
setup time?
/]| 258. Part Four
Capacity and Scheduling
The theory of constraints has had a major impact on
scheduling software design and practice. Software using
finite capacity scheduling can identify bottlenecks and
make it possible to configure schedules that achieve the
goal of making more money.
TOC can also be applied to service operations. The
Odessa, Texas, police department used TOC to improve
its process for hiring new police officers.* The old process consisted of eight steps: application, written exam,
background
exam,
The police department in Odessa, Texas, used TOC to
MPL CNS tne emg esCccss.
Blend Images/Alamy Stock Photo
investigation,
medical
exam,
oral
interview,
psychological
exam,
polygraph
and drug
screening. The entire hiring process took 117 days from
initial application to successful hiring of a candidate.
While most of the steps were completed in a few days,
the background investigation took 104 days and became
the constraint or the bottleneck in the process. Of the
60 applicants each year, only 10 were able to complete the entire process and were hired.
Because of the long waiting time, many applicants became discouraged and found jobs
elsewhere. As a result, the Odessa Police Department had a shortage of qualified police
officers.
To relieve the bottleneck, it was decided that the background check should be divided
into two parts, a cursory initial background check followed by a complete background
check. The cursory initial check could be done in one day and would eliminate some applicants immediately. The complete background check would be done only after an applicant
passed most of the other steps. As it turned out, it took only 3 days of processing to complete the entire background check and most of the 104 days was waiting time. As a result
of reducing the workload on the bottleneck, the total throughput time was reduced from
117 days to 16 days, and the yield of the process improved to provide the badly needed 20
officers per year.
12.5
LO12.5
PRIORITY DISPATCHING RULES
Compare
various dispatching
rules.
Dispatching is a technique used during scheduling to determine the job priority at any
particular work center. The priority of a job may change from one work center to another,
depending on the particular dispatch rules chosen. In finite capacity scheduling, the dispatch rule is used to select the particular job to schedule next in a work center when more
than one job is waiting in line to be processed.
In practice, schedules are difficult, if not impossible, to maintain because conditions
often change: A machine breaks down, a qualified operator is ill, materials do not arrive on
time, and so on. Real-time adjustments to the schedule are made by use of dispatch rules to
determine which job to process next.
A dispatch rule specifies which job should be selected for processing next from among
a queue of jobs. When a machine or worker becomes available, the dispatch rule is applied
and the next job is selected. A dispatch rule is thus dynamic in nature and continually sets
the priority on the basis of changing conditions.
“L. J. Taylor, Ill, B. J. Moersch, and G. M. Franklin, “Applying the Theory of Constraints to a Public Safety
Hiring Process,” Public Personnel Management 32, no. 3 (2003), pp. 367-382.
Chapter12
Scheduling Operations
259 ||
In practice, several types of dispatch rules are used. For services three types of rules are
common:
1. First come, first served (FCFS) is very common and gives customers a feeling of fairness. Using this rule, customers are served in the exact sequence in which they enter the
waiting line. FCFS is rarely used in manufacturing since it performs poorly in meeting
due dates and minimizing makespan.
2. Priority rule assigns some customers to be served before others. The priority can be
based on the price the customer paid (for example, flying using a first-class ticket), the
expected time to service their needs (for example, the “10 items or less” lane at the grocery store), or other factors.
3. Preemptive rule is used to interrupt service for one customer and shift capacity to a
customer with more urgent needs. Hospitals, police, and fire services often use a preemptive rule to shift resources to a customer in a life-or-death situation.
For manufacturing, two types of dispatch rules are commonly used:
1. Critical ratio (CR) is computed as
CR= remaining time until due date
remaining processing time
The job with the minimum value of CR is scheduled first, the job with the next smallest
value of CR is scheduled next, and so on. When the ratio exceeds a value of 1, there is
sufficient time available to complete the job if the queue times are managed properly. If
the ratio is less than 1, the job will be late unless processing times can be compressed.
The CR rule has a precise meaning; for example, a ratio of 2 means there is twice as
much time remaining until the due date as the processing time.
i)
. Shortest processing time (SPT) means the job with the least time needed on the
machine (or resource) is selected. This rule is based on the idea that when a job is finished quickly due to its short processing time, other machines downstream will receive
work, resulting in a high flow rate and high utilization.
While the SPT rule is good at achieving efficiency and high throughput, it is poor at
meeting due dates, primarily because due dates are not considered in its calculation. However, due dates are very critical in practice. The CR does a much better job of meeting due
dates since it explicitly considers them in its calculation.
Lead times can vary depending on how managers
set them and how work is scheduled. If a job is given
a very tight due date, perhaps just a little greater than
total processing time, a rule such as CR will speed
the job through the shop because of its high priority.
Depending on the dispatching rules used, lead time can
be expanded or contracted to a great extent.
It comes as a surprise to some people that lead times
can be managed. The common view is that lead time is
relatively fixed or a statistical phenomenon. There is little realization that lead time is a function of both priority
and capacity. If an operation is producing at near capacity, average lead times will be extended. Even though
the average lead time is long, an individual job can be
quickly delivered if its priority
is high.
peice Thus,ot lead time is
j
This dispatcher determines the priorities of work being done.
Rick Brady/McGraw-Hill Education
a function of both capacity and priority decisions.
1] | 260, Part Four
Capacity and Scheduling
iE
re ail
the first choice
The Villages is a retirement community in Florida with 666 holes of
golf consisting of 12 championship
courses with 27 holes each and 38
nine-hole executive courses. Scheduling over 7000 golfers per day on |
these courses presents a tremendous challenge. The Villages accomplishes this by utilizing an online
reservation system that uses a prior-
and
totals are given the lowest priority for
their requested times when scheduling. The assignment process is
continued
linda nolan/Alamy Stock Photo
ity dispatching rule.
Each time a resident schedules a round of golf, he/
she is assigned one point. If the reservation is canceled
prior to play, the player is assigned another point, and if
of golf course
requested tee times, since they have
recently played the least. Golfers that
play the most and have high point
until all 7000 golfers are
scheduled by the reservation system
or the courses run out of capacity.
This point system helps avoid
wasting golf course capacity by encouraging players
to use their scheduled tee times and also provides fair
a third point is assigned. A rolling total count of points is
and equal access to the courses for all residents. Since
executive golf courses can be played free of charge, the
point system prevents individual golfers from overusing
kept for the most recent seven days. When requesting
tee times, players with the /owest point totals are given
the courses. This golf community is successful
priority rules to allocate tee times to residents.
the player does not show up for the scheduled tee time,
in using
In summary, dispatching rules are used to determine the priority of a job during scheduling and in real time during processing. The priority of a job can be changed dynamically
as it is processed through the shop. See the Operations Leader box for an example of using
a priority rule for scheduling golf tee times.
12.6
LO12.6
PLANNING AND CONTROL SYSTEMS
Describe
the important
factors to consider
when designing a
scheduling system.
When scheduling operations, a planning and control system is needed. This system should
facilitate the development of good schedules (planning) and ensure that schedules are
implemented and corrected as needed (control). These scheduling systems incorporate the
methods described above. But without managers’ attention to the overall system, the methods are useless.
Every scheduling system should answer several questions:
1. What delivery date do we promise? The promised delivery date should be based
on both marketing and operations considerations, including available capacity and customer requirements. We have seen how the promised date can be derived through Gantt
chart scheduling or FCS. Marketing needs to input appropriate customer priorities into the
scheduling process. Finance needs to make the capital available in time. All functions need
to work together for the best interests of customers and the bottom line of the business.
2. Where is the bottleneck? Capacity of the entire facility will be limited by the bottleneck work center. The scheduling method must therefore find the bottleneck and work
to remove it by adding more capacity with approaches such as reducing the setup time,
adding equipment, overtime, subcontracting, and reassigning work at the bottleneck.
Chapter12
Scheduling Operations
261] |)
3. When should we start each job, activity or task? This question is answered by
using dispatching rules or a Gantt chart/FCS schedule.
4. How do we ensure that the job is completed on time? Dispatching helps answer
this question partially, but the answer also requires constant activity monitoring. Corrective action can be taken as needed to make sure that jobs are finished on time.
Advanced planning and scheduling (APS) is a process and software that includes
scheduling methods such as finite capacity scheduling, constraint-based bottleneck scheduling, and dispatching at the plant-floor level. APS systems consider materials requirements and plant capacity when generating a daily production schedule, helping to achieve
goals such as maximum throughput with minimum inventory while meeting customer due
dates.
To illustrate the principles of a scheduling planning and control system, an example is
discussed.
Courtroom Scheduling
Courtroom
scheduling
requires scheduling planning and control systems.
Sometimes
court calendars are overloaded, with the result that police officers, witnesses, lawyers, and
defendants spend long times waiting. After a day of waiting without being heard, some
witnesses will not return.
At the same time, the court may suffer from case underload, with judges kept waiting
and inefficient use of courtrooms. This condition occurs when cases finish earlier than
expected or cases scheduled to appear are delayed.
In an attempt to correct these problems, a court-scheduling system was developed for
the New York City Criminal Court to achieve the following objectives: (1) high probability
that judges will be kept busy, (2) high probability that cases will be heard when scheduled,
(3) several cases for a police officer should be batched on the same day, (4) high-priority
cases should be scheduled as soon as possible, and (5) a maximum
waiting-time limit
should be set for all cases.
The heart of the system was a priority dispatching rule. Cases are not scheduled until
they meet the priority rule based on factors such as seriousness of the charge, whether
the defendant was in jail, and elapsed time since arraignment. When the priority of a case reaches a certain threshold, it is inserted into the court calendar; otherwise it remains
unscheduled.
When a case was scheduled, the capacity reserved for it was
predicted by a multiple regression equation that utilized causal
variables such as the plea of the defendant, seriousness of the
offense, number of witnesses, and presiding judge. The predicted time was then scheduled for the first available spot on
the calendar. Some slots, however, were kept open for emergencies and future rescheduling.
Each day the planning and control system allows priorities
to be revised on the basis of the current conditions. Any new
cases with a high enough priority are added to the schedule.
This example illustrates how a service operation can use a
dispatching rule as part of a planning and control system, and
how the system answers the crucial questions above. Scheduling embedded
in an information system helps answer the crucial
c
Courtrooms can be scheduled with better systems.
questions required to develop and implement the schedule.
Guy Cali/Corbis/Getty Images
|} | 262|
12.7
Part Four
Capacity
and Scheduling
KEY POINTS AND TERMS
In this chapter, we have treated scheduling decisions for batch operations. The chapter’s
theme is that all scheduling decisions deal with the allocation of limited resources to jobs,
activities, tasks, or customers. We assume, for scheduling purposes, that resources are
fixed as a result of aggregate planning and facilities decisions.
The following are among this chapter’s key points:
*
Within the available resources, scheduling seeks to satisfy the conflicting objectives of
low inventories, high efficiency, and good customer service. Thus, trade-offs may be
*
¢
implicitly or explicitly made whenever a schedule is developed. Because of the potential for conflicting objectives, cross-functional coordination is required for effective
scheduling.
Gantt charting is the simplest form of scheduling. It is used to schedule jobs one at a
time according to the resources available. Gantt charting will determine the waiting
time of each job, job completion dates, resource (machine) utilization, and the makespan of all jobs.
Finite capacity scheduling is used to schedule multiple jobs through a number of different work centers. The jobs are scheduled in a manner similar to Gantt charting except
that each work center may have multiple machines (or resources). In finite scheduling,
efforts are made to improve job flow through bottlenecks by splitting jobs, alternative
routings, overtime, and other methods.
¢
The theory of constraints (TOC) is aimed at making money by managing throughput,
inventory, and operating expenses. The bottleneck resource is scheduled to increase
throughput, while non-bottleneck resources are scheduled to keep the bottleneck
busy.
¢
Dispatching is used to decide on the priority of jobs as they pass through the factory or
service process. Various dispatching rules can be used to decide which job or activity to
process next at each work center.
*
Scheduling systems should answer the following questions: (1) What delivery date do
we promise? (2) Where is the bottleneck? (3) When should we start each job, activity,
or task? (4) How do we ensure that the job is completed on time? To handle constantly
changing situations, these systems are usually part of an information system and they
operate as a daily planning and control system.
*
Key Terms
Lead time for completion of a job is not a statistical phenomenon. Lead time is a function of both capacity and priority decisions.
Time-phased plan 249
Trade-off decision 250
Batch scheduling 250
Network of queues 250
Gantt chart 251
Makespan 253
Machine utilization 253
Machine-scheduling
problem 253
Finite capacity scheduling 254
Forward scheduling 254
Backward scheduling 254
Bottleneck 254
Throughput 256
Inventory 256
Operating expenses 256
Theory of constraints 256
Dispatching 258
Dispatch rule 258
First come, first served 259
Priority rule 259
Preemptive rule 259
Critical ratio 259
Shortest processing
time 259
Advanced planning and
scheduling 261
Chapter 12
LEARNING
Theory of Constraints—Introduction
Video
https://youtu.be/1UqWejurWwU
536
ENRICHMENT
(for self-studyor
DIOS GIES)
Scheduling Operations
Round Robin CPU Scheduling Algorithm
https://youtu.be/aWIQYIIBZDs
Video
4:23
Linear Programming for Staff Scheduling
https://youtu.be/3I8B2 1M-ob8
Video
3:58
nMetric—Smart Job Scheduling Priority
https://youtu.be/y9DqLj9MGzU
Video
1:59
Job Shop Scheduling
Scheduling: Washburn Guitar
Video
10:40
263 ||)
SOLVED PROBLEMS
Problem
1. Gantt Charting The table below includes information about four jobs and three work
centers. Assume that six hours is required to transfer a job from one work center to
another (move time not including waiting time for busy machines). Sequence the jobs
with a Gantt chart. Use the following priority order: 1, 2, 3, 4. Can you tell if any jobs will
be late? Which ones? Can a simple reordering of job priorities alleviate the problem?
Solution
Job
Work Center/Machine Hours
Due Date (days)
1
2
3
4
A/2, B/1, C/4
C/4, A/2
B/4, A/2
B/4, A/2, C/3
3
2
2
3
jabiat!
center
Time (in hours)
OZ
OS
2
sea
See
nO
ara
Se
Oo LO PITS
OF
sa
Ss
9
ISI 4S aI6 7s 1S 195202212223
245
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Job 4 will be one day late, since it is due at the end of day 3, and it will still need one hour
of processing on day 4 to be completed. But moving job 4 in front of job 3 in the schedule
priority would allow job 4 to be completed one hour before it is due. Job 3 would still be
done on time.
Problem
2. Priority Dispatching Rules The five jobs listed below are waiting for their final operation at a work center. Determine the order of processing for the jobs by using the following rules: SPT, FCFS, and CR.
Job
Processing Time
Due Date
Order of Arrival
A
B
G
D
E
5
7
6
2
10
10
18
29
12
19
3
1
2
5
4
:||264| Part Four
Solution
Capacity and Scheduling
The order of processing for each rule is given below. The SPT rule assigns jobs with the
smallest amount of processing time first. The job with the next smallest amount of processing time is assigned to be produced next, and so on. According to this rule, the following
sequence is used:
D—A-—C-B-E
The CR rule assigns the jobs on the basis of the minimum ratio of total time remaining
to total processing time remaining. In this case we take the ratio of due date to processing
times in the table and arrange the jobs in sequence from the minimum to the maximum CR
ratio, resulting in the following sequence:
E-—A-—B-—C-—D
The FCFS rule assigns the job that arrived first to be processed first, and so on. According to this rule, the following sequence is used:
B—C-—A-E-—D
Problem
3. Finite Capacity Scheduling
A medical clinic has three departments (A, B, and C)
that can process patients. Each department has one clinician, except Department C,
which has two. Currently there are four patients who must be scheduled through the
three departments on a first come, first served basis in the sequence patient 1, 2, 3,
and 4. The times required to process each patient in each department are shown below.
Patient
1
z
3
4
Department/Time
B/2, C/2, A/3
Ai 2-C/3B 7A
Bis} C/T, A/2
C/2, B/3
Using finite capacity scheduling, draw a Gantt chart for the schedule.
a. What is the makespan?
b. Which department is the bottleneck?
c. If capacity is added to relieve the bottleneck, what is the new makespan?
Solution
The Gantt chart is shown below.
a. The makespan is 10.
b. Department A is the bottleneck. Patients 2 and 3 could be completed earlier if capacity were added to the bottleneck. Department A constrains the output (makespan) of
the entire facility.
c. The new makespan is 8 if one more clinician is added to Department A.
Chapter12
Scheduling Operations
|265 ||)
Discussion Questions
it What types of scheduling decisions are management likely to encounter in the following operations?
Describe the differences between Gantt charting, FCS,
and the theory of constraints.
Describe the scheduling decisions in terms of the types
of resources to be scheduled and the associated customers or jobs scheduled.
Why are m X n machine-scheduling algorithms not
widely used in practice?
a. Hospital
What is the purpose of a planning and control system
related to scheduling?
b. University
What is the goal as stated by the theory of constraints
c. Moviemaking
(TOC), and how is that goal achieved?
d. Make-to-order factory
What is the definition of a bottleneck according to
TOC?
. Specify the kinds of objectives that might be appropriate for each of the situations listed in question 1.
What scheduling rule should be applied to bottleneck
work centers, and what scheduling rule should be
applied to non-bottleneck work centers?
. Why is it important to view a batch-process operation
as a network of interconnected queues?
. How is the scheduling of patients in a doctor’s clinic similar to and different from the scheduling of jobs in a factory?
. What measures can be taken to provide more capacity
at a bottleneck work center?
Problems
ite Students must complete two activities to register for class:
registration and payment of fees. Because of individual dif-
a. Develop a Gantt chart to determine the total time
required to process all six jobs. Use the following
ferences, the processing time (in minutes) for each of these
sequence ofjobs: 1, 2, 3, 4, 5, 6.
two activities for five students varies, as shown below:
b. Can you develop a better sequence to reduce the
total time required for processing?
Minutes
Student
Registration
Pay Fees
12
7
5
3
5
ing sequence of students: D, E, B, C, A.
b. Can you construct a better sequence to reduce the
total time required?
c. What problems might be encountered in using this
approach to registration in colleges?
. Six jobs must be processed through machine A and then
machine B as shown below. The processing time for
each job is also shown.
=
BWN
Due Date (days)
A/3, B/2, C/2
C/2, A/4
B/6, A/1, C/3
C/4, A/1, B/2
S)
2
4
3
a. What is the makespan?
b. How much machine idle time is there?
c. When is each job delivered compared with its due
date?
d. How much idle time (waiting time) is there for each
job?
e. Devise a better job sequence for processing.
. In problem 3, assume there are two machines of type A,
two of type B, and two of type C.
Machine (minutes)
A
Work Center/Machine Hours
NOON
a. Construct a Gantt chart to determine the total time
required to process all five students. Use the follow-
Job
hour. Sequence the jobs in priority order 1, 2, 3, 4.
Job
4
MmMOQOWD
Sequence the jobs shown below by using a Gantt chart.
Assume that the move time between machines is one
B
1
2)
10
6
6
12
3
7
Vf
a. Prepare a finite capacity schedule.
b. Compare the FCS to the Gantt chart in problem 3.
. In problem 2, assume there are two machines of type A
and two of type B.
4
8
4
a. Prepare a finite capacity schedule.
5
3
9
6
6
8
b. How does your FCS compare to the performance of
the Gantt chart in problem 2?
|| 266| Part Four
Capacity and Scheduling
6. The Security Life Insurance Company processes all new
life insurance policies through three departments: incoming mail (I), underwriting (U), and policy control (P).
The incoming mail department receives applications and
customer payments and then routes the files to the underwriting department. After checking on the applicants’ qual-
8. A secretary is considering three dispatching rules for
typing term papers. The following information is given
on jobs that are waiting to be typed:
ifications for life insurance, the underwriting department
forwards the file to policy control for the issue of the policy.
At the present time, the company has five new policy
applications waiting to be processed. The time required for
processing in each department is shown below.
Policy
=
OBWN
Department/Hours
1/3, U/6, P/8
1/2, P/10
1/1, U/3, P/4
1/2, U/8, P/6
1/1, P/6
Paper
Hours
until Due
Date
Total
Remaining
Processing
Time* (hours)
Processing
Time
(typing
hours)
Order
of
Arrival
20
19
16
10
18
72
15
11
5
11
10
12
6
5
7
4th
3d
2d
1st
5th
A
B
ce
D
=
*Includes typing, corrections, and copying.
Use the following dispatching rules to determine
sequence for term paper processing.
ae
Sample
Test Station/Hours
Due Date (hours)
1
2
3
4
5
A/1, B/2, C/3, D/1
B/2, C/3, A/1, D/4
C2zANSeDiarsG/z2
A/2, D/2, C/3, B/1
D/2, C/1, A/2, B/4
6
10
8
14
12
(OR
9) Suppose you are in charge of dispatching for the
University Hospital laboratory described in problem 7.
Use the following dispatching rules for the first
station (A) to determine which job should be processed
first through station A. Hint: Decide between jobs
1 and 4.
Ay
and D) is doubled. Prepare an FCS for this situation.
c. What are the bottleneck work centers in part a of the
problem? Suggest capacity additions that might be
needed.
Ser
bs CR
10. Using the Gantt charts developed in the following problems, which is the bottleneck work center?
a.
Problem 2
b. Problem 3
a. Using a Gantt chart, schedule these five samples in
priority order of earliest due date first.
b. Assume that the capacity of each test station (A, B, C,
SUPE
b. FCEFS
a. Prepare a Gantt chart schedule for these policies.
. At the University Hospital, five blood samples must
be scheduled through a blood-testing laboratory. Each
sample goes through up to four different testing stations. The times for each test and the due dates for each
sample are as follows:
the
c.
I:
Problem 6
How would the theory of constraints be used to improve
output for the operations in the following problems?
a. Problem 2
b. Problem 3
c. Problem 6
CAs
Peeler
Project Planning
and Scheduling
a
LEARNING
OBJECTIVES
heaek
hi ow eB
After reading this chapter, you should be able to:
Se
LO13.2
Describe the four activities included in project management.
LO13.4
Calculate the ES, EF, LS, LF for an example network.
A significant amount of the work that is carried out by firms, nonprofit organizations, and
governments is managed as projects. Projects undertaken by business firms include setting up new production facilities, putting on concerts at large stadiums, as well as nearly
all consulting projects. Nonprofit organizations use projects to tackle problems, such as
the Gates Foundation goal to eradicate malaria, which is approached through multiple and
often overlapping projects. Government projects are many, including managing the implementation of a new electronic health record for the Veterans Administration.
While projects have varying degrees of complexity, most will benefit from specific tools
and techniques designed to help manage projects. Large projects, in particular, can be very
unwieldy without such tools for scheduling and managing the many different elements of
the project. Here, we introduce helpful ways to manage the project type of operation, which
produces a unique product or service.
°
=
o
o
°
°
°o
°
°
°
°
°o
°
°
°
*
a
°
(Oo
o
ee
°
6
-
°o
°
°
2
°
°
°
°o
9
9
°
\
fel
o
is]
| | 268| Part Four
TABLE 13.1
Capacity and Scheduling
Examples of Projects
a
Building construction
New product introduction
Research and development
Computer system design
Installation of equipment
NASA mission
Fund-raising campaign
te
Movie making
Teaching a course
<
Designing an advertising campaign
Start-up or shutdown of a plant
Manufacture of aircraft, ships, and large machines
Auditing accounts
Planning a military invasion
In this chapter, you will learn how to manage both
small and large projects. A project is an operation or
process undertaken to create a unique product, service,
or outcome. It should be noted that projects have a specified beginning and end; they are not ongoing activities.
Because of this, the management of a project differs
considerably from that of an ongoing operation.
Although many decisions in projects differ from those
in ongoing operations, our main concern in this chapter
is with project planning and scheduling decisions. In the
first part of the chapter, a broad framework for project
planning is established; this includes the project objectives and the planning and control activities they require.
Planning fora large fund-raising gala is managed asa project.
[In the second part of the chapter, specific scheduling
a
ee
methods are described in detail.
Projects include a wide range of manufacturing and
service activities. Large objects such as ships, passenger airplanes, and satellite launchers
are manufactured on a project basis. Each unit is made as a unique item, and the manufacturing process is often stationary, so that materials and labor must be brought to the project site. The construction of buildings typically is organized on a project basis. Services
such as making movies and fund-raising campaigns are also produced on a project basis.
Table 13.1 lists a wide range of manufacturing and service activities that are managed as
projects.
13.1
OBJECTIVES AND TRADE-OFFS
LO13.1 Explain the
In projects, there are usually three distinct project objectives: cost, schedule, and perfornature of trade-offs
mance. The project cost is the sum of direct and allocated costs assigned to the project. The
among the three
project manager and project team’s job is to control those costs that are directly controlobjectives of project —_ lable by the project organization. These costs typically cover labor, materials, and some
management.
support services. Ordinarily, the project will have its own budget, which includes the costs
assigned to the project.
The second objective in managing projects is schedule. A project completion date
and intermediate milestones frequently are established at the outset. Just as the project
team and manager must control the project costs within budget, they must also control the
schedule to meet established dates. Frequently, the budget and schedule conflict. For example, if the project is behind schedule, overtime may be needed to bring it back on schedule.
But there may be insufficient funds in the budget to support the overtime costs. Therefore,
a trade-off decision between time and cost must be made. Management must determine
whether the schedule objective is of sufficient importance to justify an increased cost.
Chapter 13
Project Planning and Scheduling
/269| |J
The third objective in project management is performance, that is, the performance
characteristics of the product or service being produced by the project. For constructing a
building, performance can be meeting the specifications and blueprints for the building.
If the project is a movie, performance refers to the quality of the movie produced. In this
case, performance may be specified by a variety of movie standards regarding casting,
sound, filming, and editing. Generally speaking, performance is meeting some measures of
outcome success for the project.
Performance may also require trade-offs with Sh schedule and cost. In a movie, for
example, if the picture is not meeting performance expectations, additional shots or script
revisions may be required. These performance requirements may, in turn, cause cost and
schedule changes. Since it is rarely possible to predict performance, schedule, and cost
requirements accurately before a project begins, numerous trade-offs may be required
while the project is under way.
13.2
PLANNING AND CONTROL IN PROJECTS
LO13.2 Describe
the four activities
included in project
management.
TABLE
Project
13.2
A general sequence of management decisions required in all projects is planning, scheduling, control, and closing. Details of some of the activities involved in each of these project
stages are shown in Table 13.2.
Project planning refers to those decisions required in the beginning of a project that
establish its general character and direction. Project planning establishes the major project objectives, the resources required, the type of organization used, and the key people
who will manage and implement the project. Project planning is usually a function of
top and middle managers, with a cross-functional team often making all major decisions.
When completed, project planning should be documented by a project authorization form
or letter, which is used to initiate further project activities. The project authorization form
should specify all the planning decisions listed in part A of Table 13.2.
In the project scheduling phase of project management, the cross-functional team
specifies the project plan in more detail. This phase begins with the construction of a
detailed list of project activities, called a work breakdown structure. A detailed time
‘
A. Planning
Identify the project customer
C. Control
Monitor actual time, cost, and performance
Management
Activities and
Establish the end product or service
Set project objectives
Compare planned to actual figures
Determine whether corrective action is needed
Decisions
Estimate total resources and time required
Evaluate alternative corrective actions
Decide on the form of project organization
Make key personnel appointments (project
manager, etc.)
Define major tasks required
Establish a budget
Take appropriate corrective action
B. Scheduling
Develop a detailed work breakdown structure
Estimate time required for each task
Sequence the tasks in the proper order
Develop a start/stop time for each task
Develop a detailed budget for each task
Assign people to tasks
D. Closing
Finish all work
Close contracts
Pay all accounts payable
Turn over project to owners
Reassign personnel and equipment
H] |270) Part Four
Capacity and Scheduling
FIGURE 13.1
Work breakdown
structure
example—Banquet.
Source: Reprinted from
ies Dae a
“Level 1? =>
“Level 2”
1.1
1.2
1.0)
1]
1.3
1.4
1.5
1.6
Room &
Guests
Staff
Speakers
;
www.hyperthot.com/pm_
Planning &
wbs.htm
Supervision
Dinner
Equipment
Planning
Menu
Site/Room
Guest List
Shoppers
Invite
Budget
Shopping
List
Tables/Chairs
RSVPs
Cooks
Transport
Setting/
Utensils
Name Tags
Servers
Coordinate
Topics
Special Needs
Hosts
Backup for
No-Shows
Cleanup
Thank-yous
Disbursements/
Reconciliation
[ Shopping
Coordination
Cooking
Decorations
SAA se Ss
Serving
Equipment,
Pots, etc.
schedule for each activity in the work breakdown structure is then established using the
methods described later in this chapter. When the time schedule is completed, a timephased budget, which is keyed to the start and completion times of each of the project
activities, can be developed. Finally, the project personnel can be assigned to individual
project activities.
A work breakdown structure (WBS) is a hierarchical listing of all the tasks needed to
complete a project. [t is constructed by organizing the project into activities and subactivities, as shown in Figure 13.1 for a banquet. Note how Level 2 consists of all the activities
needed to complete the banquet, including planning and supervision, dinner, the room and
equipment, guests, staff, and speakers. Each of these activities, in turn, is broken down
into Level 3 subactivities that have to be completed. From the WBS, a schedule can be
prepared and a budget and personnel can be assigned to
each activity. This makes it possible to assign responsibilities for each part of the project and even subcontract
portions of the project if desired. The WBS becomes
the basis for planning, scheduling, budgeting, and controlling the project.
Project control is maintained by the cross-functional
team, which monitors each activity as the work is performed on the project. Activities should be monitored
for time, cost, and performance in accordance with the
project plan. When a significant discrepancy exists
between actual results and the plan, corrective action
should be taken. These corrective actions might include
revision of the plan, reallocation of funds, personnel
changes,
and other changes in resources. Such correcBOEING PRODUCTION. Large aircraft are produced by using
:
;
aks
pe
arolect nisnngenientinernod
r,
tive actions should make the plan feasible
and realistic
© Larry W. Smith/Getty Images
once again.
Chapter 13
Project Planning and Scheduling
271) [oe
Project closing is concerned with formally ending a project. It includes finishing all
work, closing all subcontracts, paying all bills, the turnover of the project to its “owners,” and reassignment of the personnel and equipment used on the project. It is important
to close a project so that a specific ending is defined. In road construction the closing phase
is often ended with the “ribbon cutting.” All the phases of a typical project at Churchill
Downs are shown in the Operations Leader box.
Project management is a profession. As such, it has a body of knowledge, a professional organization—the Project Management Institute (PMI), and a variety of recognized
certifications. The body of knowledge is extensive and forms the basis for certification
as a project manager. See Table 13.3 for the topics in the body of knowledge defined by
the PMI.
We have already discussed some of the subjects in the body of knowledge. Owing
to space limitations, the remainder of this chapter will be restricted primarily to project
scheduling methods and related concepts such as the critical path and slack.
OPERATIONS LEADER
Churchill Downs Embraces Project Management
Churchill Downs is famous for the annual Kentucky Derby
horse race—“the most exciting two minutes in sports.”
Prior to establishing a project management office (PMO),
they managed projects in an informal way. Churchill
Downs then hired a project director and established a
PMO to formalize the process of project management. To
ease the organization into the new process, a racetrack
analogy was developed as shown below. The starting
gate consists of initial approval and prioritization along
5 Risk & Issues Logs
6 Scope Change Control (Requests/Logs)
7 Testing/Defect Tracking
with a budget and a designated project manager. Then
at the next gatepost, the charter and work breakdown
structure are developed. This is followed by monitoring
progress and risk assessment as the project progresses
around the racetrack until it reaches the finish line. Since
this approach was so successful for an initial IT project, it
was expanded to other projects at Churchill Downs.
Source: PM Network, July 2009, pp. 40-45 and
Churchilldowns.com, 2020.
3 Charter
4 Work Breakdown Structure
Starting Gate
2 Approval & Prioritization
1 Investment Request Worksheet
2
ro
Paddock
8 IT Approval (as applicable)
9 Production Turn/Implementation
10 Sponsor Approval
i | 272| Part Four
Capacity and Scheduling
TABLE 13.3
Project Management
Institute Body of
Knowledge
1. Integration Management
Develop project charter
Project planning
Project execution
Monitor and control
2. Scope Management
Scope planning
Work breakdown structure
Scope verification
Scope control
4. Cost Management
Cost estimating
Cost budgeting
Cost control
5. Quality Management
Quality planning
7. Communications
8. Risk Management
Risk management planning
Risk resource planning
Risk monitoring and control
Management
Communications planning
Information distribution
Performance reporting
Quality assurance
Quality control
3. Schedule Management
Sequencing of activities
Resource estimating
Activity duration estimating
Schedule control
6. Resource Management
Organize project team
Lead project team
Manage project team
9. Procurement
Management
Contracting
Select sellers
Contract administration
Contract closure
10. Stakeholder
Management
Identify stakeholders
Analyze expectations
Analyze impact
Engage stakeholders
13.3
LO13.3
SCHEDULING
Distinguish
the advantages and
disadvantages of a
network over a Gantt
chart for project
scheduling.
Example
METHODS
Several types of scheduling methods are in use. These may be generally classified as Gantt
chart or network methods. Gantt chart methods utilize a bar chart, as shown in Figure 13.2.
The network methods‘use a graph or network to show precedence relationships.
The Gantt chart method of scheduling has a great deal in common with Gantt chart
scheduling for batch processes as described in the previous chapter. In each case, the activity durations are shown on the chart by a bar or line. These charts also show when each
activity is scheduled to begin and when it will be completed.
Figure 13.2 is a simplified Gantt chart for establishing a new office location for a business.
Time is shown across the top, and activities are shown down the side. Each activity in the
project is depicted as a bar on the chart over the period of time for which the particular
activity is scheduled.
The first activity in the chart is to decide on the location and lease the office space,
which is started and completed in Week 1. After this is decided, the company can begin
to hire workers and arrange for furnishings and phones. Thus, activities 2, 3, and 5 occur
after activity 1. The chart also shows that arranging for furnishings, hiring, and arranging for
phones occur in parallel during the same time frame. The Gantt chart therefore shows not
only how much time is required for each activity but also when each activity takes place.
Gantt charts are commonly used in project scheduling because they are easy to use and
quite widely understood. In complex projects, however, a Gantt chart becomes inadequate
because it does not show the interdependencies and relationships between activities. For
complex projects, it is difficult to schedule the project initially and even more difficult to
reschedule it when changes occur. The network method of project scheduling overcomes
these difficulties.
Chapter 13
FIGURE 13.2
Gantt chart project
example.
Project Planning and Scheduling
|273 |||
Week
No.
Activity Description
1
2
=
ire the workers
area
Arrange for the furnishings
|
ce)lees)
ied
Arrange for the phones
;
[6 |testatephones
Move into the office
The advantage of the network method over the Gantt chart is that the precedence relationships in network scheduling are shown explicitly on the network. This permits the
development of scheduling algorithms that account for all precedence relationships when
the schedule is developed. With Gantt charts, the precedence relationships must be kept
in the scheduler’s head. On complex projects, this cannot be done easily, and Gantt charts
become unwieldy. Furthermore, when a single activity time changes on the Gantt chart,
the entire chart must be rescheduled by hand. Rescheduling can be done automatically
by a network algorithm. In contrast, networks are more complex, more difficult to understand, and more costly to use than Gantt charts. Thus, networks should be used in complex
projects that have many different interrelated activities. However, when using scheduling
software even very simple projects can take advantage of network methods.
Network scheduling methods involve the use of some important scheduling concepts,
such as critical path and slack. These network scheduling concepts will be described next
by means of the constant-time network. More complicated networks using time-cost tradeoffs will be discussed later in the chapter.
13.4
CONSTANT-TIME
LO13.4 Calculate
the ES, EF,LS,LF for
an example network.
NETWORKS
In constant-time networks, the time for each activity is assumed to be a constant. This is
— the simplest case from the standpoint of scheduling. Other, more complicated methods are
— then derived from these constant-turme network methods.
First, we illustrate the construction of a simple network. Table 13.4 shows the activities
required to write a typical business report. The first activity, A, is to decide on the topic
and scope of the report. Then two activities, B (collect data) and C (search the Internet),
proceed in parallel or at the same time. Once B and C are completed, the report can be
written (activity D). Table 13.4 indicates the immediate predecessors of the activities we
TABLE
13.4
Write a Business
eas
Immediate
Duration in
Activity
Description
Predecessors
Days
A
B
Gj
D
Decide topic and scope
Collect data
Search the Internet
Write the report
None
A
A
BandC
1
2
3
5)
P| 274
Part Four
Capacity and Scheduling
have just described. Activity A has no immediate predecessor, since it is the first thing
that must be done. Activities B and C each have activity A as the immediate predecessor.
Activity D has B and C as immediate predecessors, because the final report cannot be writ-
ten until the data are collected and the Internet is searched. The duration times for each
activity are also shown in Table 13.4.
The above information can be put into a network representation. The activity-on-node
(AON) network representation is shown in Figure 13.3.'! Here, each of the four activities
is shown as a node (circle) in the diagram and the arrows indicate the precedence relationships between the activities. In each circle the duration of the activity is shown below its
label. Activities
B and C cannot start until activity A is completed, and activity D cannot
start until both activities B and C are completed. The convention is that all preceding
activities must be completed before a successor activity can start.
We use the business report example from Table 13.4 and Figure 13.3 to illustrate
constant-time network scheduling. Once the network (precedence of relationships) has
been defined, the scheduling calculations can be made. To calculate activity start and finish times the following notation and definitions are needed:
ES(a) = early start time of activity A based on the early finish times of all immediate
predecessors.
EF(a) = early finish time of activity A, which is constrained by its early start time.
LS(a) = late start time of activity A, which is constrained by its late finish time.
LF(a) = late finish time of activity A without delaying the late start time of all
immediate successors.
Each activity has four scheduled times, as defined above. For convenience we abbreviate
them as ES, EF, LS, and LF. These times may be calculated by a forward pass and a backward pass through the network.
First, a forward pass is made to calculate the early start (ES) and early finish (EF) times
directly from the network diagram. This is done by starting at the beginning of the network
and proceeding through to the end of the network in precedence order. We illustrate this calculation by using the example from Figure 13.3. A convention is that we place the ES and EF
times on top of the nodes, as shown in Figure 13.4. Starting at node A, we assign an ES time
of zero since it is the first activity (ES = 0 on node A). The EF of activity A is the ES plus
the activity duration (from Table 13.4), which is 0 + 1 = 1. The earliest any activity can possibly finish is just the time it takes to conduct the activity (its duration) added to its ES time.
FIGURE 13.3
Network for writing a
business report.
' There is also an activity-on-arrow convention for drawing networks. We use the AON convention
here
because it is easier to understand and is the basis for most scheduling software (e.g., Microsoft
Project).
Chapter 13
FIGURE 13.4
Forward pass for
writing a business
Project Planning and Scheduling
275 ||
13
report.
01
49
ES EF
The ES of activities B and C, the successors
of activity A, is the EF of activity A
because B and C cannot start until A is finished. It follows that the EF of activity B is its
ES time (1) plus the activity duration of B, which is 2 days, or EF(B) = 3. Likewise, the
EF(C) is its ES time (1) plus the activity duration of C, which is 3 days, EF(C) = 4.
Now we can schedule the ES of activity D, which cannot begin until both activities B
and C are completed. Therefore, the earliest start time of activity D is determined by the
maximum of the EF of both B and C. In this case, ES(D) = Max (3, 4) = 4. The meaning
of precedence is that both the preceding activities have to be finished before activity D can
start, therefore taking the maximum of EF times of these predecessor will yield the earliest
time that D can start. The earliest finish of activity D is its ES plus the activity D duration
time [EF(D) = 4 + 5 = 9]. We have now completed the forward pass for this network. The
project completion time is 9 days, which is the EF of the last activity.
The logic we have been using in these calculations can be expressed by the following
formulas:
ES(a) = 0 for starting activities
EF(a) = ES(a) + t(a), where ¢(a) denotes the duration of activity a
ES(a) = Max [EF (all predecessors of a)]
Project completion time = Max [EF (all ending activities)]
To summarize, the reason we take the maximum EF of all predecessors to each activity
for its early start time is that an activity cannot start until all its predecessors are finished.
Even one predecessor that is not finished can hold up the early start of the next activity,
which leads to the maximum EF of all predecessors. Once the ES of an activity has been
determined the early finish (EF) of that activity is just its ES + the activity duration time,
t(a). So the activity cannot finish until its duration has passed from the early start time.
A backward pass is needed to calculate the late start (LS) and late finish (LF) times.
This is done to determine the latest that activities can be started and completed, without
affecting other activities. The backward pass is based on the following calculations:
LF(a) = Min [LS (all successors of a)]
LS(a) = LF(a) — t(a)
These late times are computed starting with the last activity in the network and proceeding backward through the entire network. As shown, LS and LF are positioned below the
node. The LF time for activity D in Figure 13.5 is the same as the EF since this is the last
i|| 276| Part Four
Capacity and Scheduling
And
FIGURE 13.5
Forward and backward
pass for writing a
business report.
01
49
01
49
ES EF
14
LS LF
activity, which is 9. Also, the LS for activity D is the same as the ES, which is 4. The LF
for activities B and C is equal to the LS of activity D since it is the only successor. The LS
for activities B and C is obtained by subtracting their activity duration times from their LF.
For activity A there are two successors, B and C, so the LF for activity A should be the
minimum of the LS for the successors, Min (1, 2) = 1. The reason we use the minimum is
that activity A must finish in time for both B and C to achieve their late start times or the
project completion will be delayed. Since the late start of either activity B or C could delay
the project completion, we take the minimum of their LS times to ensure the late finish
of activity a is finished on time. The LS for activity A is its LF minus the activity time,
1-—1=0.
On the backward pass, the same logic is applied as the forward pass except we have to
consider now the minimum LS time of all successors. As we work backwards through the
network, the LF of each preceding activity must be completed in time to not delay any of
the LS of the successors. That leads to the formula LF(a) = Min [LS (all successors of a)].
Once the LF(a) has been determined, the LS(a) is just the LF(a) minus the duration time
of activity (a).
LO13.5
Explain
the significance of
the critical path and
Slack.
As a check on your computations, determine whether the ES = LS and EF = LF for
activity A, the first one in the network. After the backward pass, you should always end up
with the same ES and EF times that you started with when going forward.”
We can now identify the critical path as the longest path in the network from start to
finish. It consists of all activities where ES = LS and EF = LF. In this case the critical path
of activities is A-C—D. For these activities the earliest they can start is also the latest they
can start. There is no slack in the activities on the critical path. The critical path constrains
the completion time of the project since it is the longest path of activity times from the start
to the end of the project. In the example in Figure 13.5 there are only two paths through
the network: A-B—D and A-C_D. The length of these paths is 8 and 9, respectively. These
lengths are obtained by adding the times of activities along each path. It is apparent from
these calculations that path A-C—D is the longer of the two and is therefore the critical
path. Notice that the length of the critical path is the project completion time we have just
computed.
In large examples it is not possible to enumerate all paths, as we have just done to find
the longest path, because there are simply too many. Therefore, the forward and backward
“This assumes that the project starts with a single activity.
Chapter 13
Project Planning and Scheduling
2771)
passes are used to find the critical path by noting those activities where ES = LS or equivalently EF = LF. In other words, the critical activities comprise the longest chain because
the earliest they can start is also the latest they can start and the earliest they can finish is
also the latest they can finish.
Slack is defined as the mathematical difference between LS and ES, or equivalently
LF-EF. In Figure 13.5 there is only one activity with slack—activity B with one unit of
slack. This means the duration of activity B can slip by one day and still not affect the project completion date. In this case it is easy to see that once activity B has slipped by one day,
it also is on a critical path and the project has two critical paths at that point.
In managing a project, all activities on the critical path must be carefully monitored. If
any of the critical activities slips (takes more time than planned), the completion date of the
project will slip by a like amount. In a typical project, with a few hundred activities, only
5 to 10 percent of all activities are on the critical path. Therefore, the focus of monitoring
on critical activities provides a significant reduction in managerial effort. This is basically
management by exception, paying more attention to those activities that are going to affect
the completion date of the project. For activities not on the critical path, they can slip by
the amount of slack in the activity before affecting the final project completion date.
It is now apparent that network calculations have several advantages over the Gantt
chart. Networks allow precise determination of the critical path and slack, and they allow
the rapid evaluation of proposed schedule changes.
Example
We take the example of opening a new office used in the Gantt chart (Figure 13.2) and
schedule it using a network representation. First, we need to specify the precedence relationships and the activity times. These are shown in Table 13.5. It’s logical that we can’t
install the furnishings before we arrange for them, and we can’t install the phones before
we make the phone arrangements. Assuming we are using hard-wired phones, we have
also specified that the furnishings be installed before the phones are installed in order to
know where to place the phone jacks, and of course, everything must be completed before
we can move into the office. These precedence relationships are shown in the network in
Figure 13.6.
Activity
Description
Immediate
Predecessors
Activity
Time
Computed
Slack
1
2
3
4
5
6
7
Lease the site
Hire the workers
Arrange for the furnishings
Install the furnishings
Arrange for the phones
Install the phones
Move into the office
None
1
1
3
1
45
2,6,4
1
5
1
2
1
1
2
O
O
1
1
3
1
0
A forward and backward pass was
and LF for each activity. When making
ties 5 and 4 both are finished. Activity
therefore, ES(7) = Max{EF(2), EF(6) and
made on Figure 13.6 to determine the ES, EF, LS,
the forward pass, activity 6 cannot start until activi7 cannot start until activities 2, 6, and 4 are finished;
EF(4)] = Max[6,5,4] = 6. When making the backward
pass, we start with the LF of activity 7 and work backward. Take a close look at the LF for
activity 4. It is the Min[LS(6), LS(7)] = Min (5,6) = 5. The remaining backward calculations are
straightforward. As a check on our math, we note for activity 1 that ES = LS and EF = LF.
P] |278
Part Four
Capacity
and Scheduling
1 6
FIGURE 13.6
Network for opening
a new Office.
LS
LF
The critical path consists of those activities where ES = LS and EF = LF, that Is, path
1-2-7. To check our logic, there are four paths in this network with the following lengths
obtained by adding the times along each path.
Activities
Total time
o1
0
N
©
As we can see, path 1-2-7 is the longest path through the network and thus the critical
path. Those activities not on the critical path have some slack, as shown in Table 13.5.
These figures are obtained by simply subtracting LS — ES or LF — EF for each activity. Slack
is the amount of time an activity can slip before affecting the project completion date.
It is possible to put the results from the critical path calculations into a Gantt chart format.
In this case we start each activity at its early start time and follow it with a dashed line to
show the slack in each activity. This provides a useful display of the critical path along with
the slack; see Figure 13.7. Next, we expand on the constant-time network to consider CPM.
FIGURE
13.7
Week
Gantt chart project
No.
Activity Description
example.
1 | Lease the site
1
Ea
Hire the workers
bed
Arrange for the furnishings
be
Install the furnishings
peal
Arrange for the phones
6 | Install the phones
Rie
Ee
Rin
7 |Move into the office
Critical path =>
Slack «===
Chapter 13
13.5
Project Planning and Scheduling
279 ||)
CPM METHOD
LO13.6 Calculate
the cost of crashing
a network by one or
two days.
The critical path method (CPM) was developed by E.I. du Pont de Nemours & Co. as
a way to schedule the start-up and shutdown of major plants. Since these plant activities
— were repeated frequently, the times were fairly well known. However, the time of any activity could be compressed by expending more money. Thus, CPM uses a time-—cost trade-off
rather than a constant time.
The CPM method of project scheduling uses a time—cost function of the type shown in
Figure 13.8 for each activity. The activity can be completed in proportionally less time if
more money is spent. To express this assumed linear time-—cost relationship, four figures
are given for each activity: normal time, normal cost, crash time, and crash cost.
The following definitions are provided.
Normal time: the planned activity duration.
Normal cost: the budgeted cost for the normal time.
Crash time: the minimum activity duration for additional cost.
Crash cost: the cost needed to achieve the crash time.
The project network is solved initially by using normal times and normal costs for all
activities. If the resulting project completion time and cost are satisfactory, all activities will
be scheduled at their normal times. If the project completion time is too long, the project can
be completed in less time at greater cost by crashing (using less time) for certain activities.
For any given project completion time that is less than the normal time, a great number
of network possibilities exist, each at a different total cost. This occurs because a variety of
different activity times can be decreased to meet any specified project completion time. All
these possibilities can be evaluated by means of a linear programming (LP) problem. The
LP problem is used to find the solution representing the minimum total project cost for any
given project completion time.
To illustrate the principles involved, an example is given below. The example shows
how to calculate normal times and normal costs and how to determine the best way to
reduce project completion by one day. Although this simple example is easily evaluated,
FIGURE 13.8
Time-cost relationship
in CPM.
s.
& g
Cost
cost
Normal
Crash
time
Normal
time
Time
1} | 280)
Part Four
Capacity
and Scheduling
it will, as the network becomes more complex, be necessary to use linear programming to
evaluate all combinations.
Note from this example that the project completion time can be decreased one unit at a
time by incurring more cost. This can be continued until all activities on the critical path
are “crashed” to their minimum times or until other paths become critical. Remaining
activities can have some slack in them. Management can determine how much it will cost
to obtain any given project completion time between the normal time and the minimum
full-crash time by simply decreasing project completion time one unit at a time until full
crash times are reached.
Example
A project network—along with activity times and costs—is given below. Calculate the normal
project time and normal cost. Also, calculate the least-cost way to reduce the normal project completion time by one day.
Note that when there is more than one beginning node (in this case A, B, and C), a start
node is added to the network. Also, when there is more than one ending node (in this case
D, C, and E), an end node is added to the network.
Activity
A
B
ce
D
E
Solution
Normal
Time
Normal
Cost
Crash
Time
Crash
Cost
S
2
6
4
3
40
50
100
80
60
1
1
4
2
1
80
120
140
130
140
The normal project completion time is computed by setting all activities at their normal
times and making a forward pass. The resulting normal project completion time = 7. See
the ES, EF time calculations below.
Chapter 13
03
Project Planning and Scheduling
281] |]
B07
The normal project cost is the sum of normal costs for all activities, which equals $330.
Crashing one or more activities on the critical path can reduce the project completion
time by one day. Therefore, we calculate the activity cost per day for each activity on the
critical path that can be crashed by at least one day and then choose the least cost per day
of those activities. The following formula is used:
Activity cost/day =
Crash $ —- Normal $
Normal days — Crash days
The project completion time can be reduced from seven to six days by crashing either activity
A or activity D, the only two on the critical path, by one day. It costs $20 per day = (80 — 40)/
(3 — 1) to crash activity A and $25 per day = (130 — 80)/(4 — 2) to crash activity D by one
day. Therefore, it is less costly to crash activity A by one day in order to achieve an overall
project completion time of six days at a cost of $350. This process can be continued, crashing the least-cost activity on the critical path (which may itself change) one day at a time until
the minimum project completion time is reached.
13.6
USE OF PROJECT MANAGEMENT
LO13.7 Contrast
and compare the use
of constant-time and
CPM networks.
CONCEPTS
Project management requires a great deal more than scheduling. Planning for the project
is required before the scheduling begins, and control is required after the schedule is
developed. Project management requires a blend of behavioral and analytic skills, often
involving the use of cross-functional teams. Thus, scheduling methods should be seen as
only one part of a complete approach to project management. See an example of project
complexity in the Operations Leader box for a major infrastructure project in California.
In selecting project scheduling methods, we should make a conscious trade-off between
sophisticated methods and cost. Gantt chart methods should not be seen as outdated or naive.
Rather, Gantt charts are justified for projects in which the activities are not highly interconnected or for small projects. In these cases when the Gantt chart is warranted, a network
method may not provide enough additional benefits in relation to its costs. When using scheduling software, a Gantt chart will normally be provided in the output similar to Figure 13.7.
If a network method is justified, a choice must be made between constant-time, CPM,
or more advanced methods. The constant-time method is adequate for cases when activity
)} | 282| Part Four
,
Capacity and Scheduling
!
OPERATIONS LEADER
|
{|}
SOA
Carlsbad Desalination Plant
The
Claude
“Bud”
opened ahead
Lewis
Carlsbad
Desalination
Plant
of schedule in 2015 to provide a fresh
water supply to the San Diego County Water Authority.
Referred
to as the nation’s
largest, most advanced
and
energy-efficient seawater desalination plant, the project
cost about $1 billion for the plant, pipelines, and upgrades
to existing facilities. Ongoing California droughts created
urgency for completing the project ahead of its original
2016 planned completion.
Named
for the former mayor of Carlsbad,
California,
the total project time line was 17 years, including planning, permitting, construction, and six months of final
testing. The project involved construction and expertise from numerous
global engineering
firms. A variety
of project management tools were necessary to manage the long and detailed schedule. In total, 1.5 million
hours of work were involved, and the project employed
needs for the San Diego region. With a paltry 3.3 inches
of rain at the city’s official weather station in 2018, this
2500 workers.
This complex
Reed Kaestner/Getty Images
project resulted
in a plant that uses
reverse osmosis to produce 50 million gallons of fresh
output per day, about 7 percent of the water
water
investment in water supply helps to sustain the region’s
3.3 million people.
Source: https://www.carlsbaddesal.com, 2020.
times are constant or nearly so. CPM methods, by contrast, should be used when activity
times are fairly constant but can be reduced by spending more money. CPM might apply
in cases such as construction projects, installation of equipment, and plant start-up and
shutdown. More advanced network methods include PERT (probabilist times), generalized
networks, resource-constrained networks, and project management based on the theory of
constraints. These methods are beyond the scope of this textbook.
Computerized network scheduling methods are used in practice. A large number of
different standard software packages are available to cover the entire range of scheduling
methods. These packages also assist in project accounting and in controlling progress.
Chevron Corporation, a multinational energy firm, manages much of its work as projects. For its major construction projects, from Kazakhstan to Australia, the use of sophisticated project management scheduling systems is crucial. In contrast, personal wedding
planners also use some of the tools discussed in this chapter, so that everything is ready
for the big day! Keeping activities on time and on budget is important in all industries, and
there is tremendous demand for professional project management skills.
13.7
KEY POINTS AND TERMS
This chapter is concerned with the planning and scheduling of projects. The major points
include the following:
*
A project is a production activity geared toward the creation of a unique product, service, or result.
Chapter 13
Project Planning and Scheduling
|283 |||
¢
The three objectives in projects are time, cost, and performance. Because these objectives
are conflicting, trade-offs among them must be made in the course of managing projects.
¢
All projects go through four phases: planning, scheduling, control, and closing. The
planning phase establishes the objectives, organization, and resources for the project.
The scheduling phase establishes the time schedule, cost, and personnel assignments.
The control phase monitors the progress of the project in cost, time, and performance; it
also corrects the plan and schedule as necessary to achieve project objectives. The closing phase shuts down the project and turns it over to the owners.
¢
Project management is a profession that includes a body of knowledge and certification
as a project manager.
¢
The Gantt chart is a scheduling method for displaying project activities in a bar-chart
form. The Gantt chart is useful for small projects or projects in which activities are not
highly interrelated.
*
Two network scheduling methods are covered in this book: constant-time and CPM.
Both methods rely on a network or graph to represent the precedence relationship
between activities.
¢
A network allows us to identify the critical path, slack, and activities that need to be
rescheduled. The critical path is the longest time path of activities from the beginning to
the end of the network. Activities on the critical path have zero slack—they must be completed on time to prevent slippage of the project completion date. Slack is the amount of
time an activity can be extended while still allowing the project to be completed on time.
¢
The early start, late start, early finish, and late finish times for each activity can be com-
puted by means of a forward pass and a backward pass through the network.
¢
CPM is a network-based method that uses a linear time—cost trade-off. Each activity
can be completed in less than its normal time by crashing the activity for a given cost.
Thus, if the normal project completion time is not satisfactory, certain activities can be
crashed to complete the project in less time at greater cost.
Key Terms
Project 268
Project objectives 268
Project planning 269
Project scheduling 269
Work breakdown
structure
Project control 270
Project closing 271
Constant-time networks
Activity-on-node 274
Early start 274
269
273
Early finish 274
Late start 275
Late finish 275
Critical path 276
Slack 277
Critical path method
(CPM)
279
LEARNING
ENRICHMENT
Gantt Chart Excel Tutorial
https://youtu.be/xsxi4qaEnOg
Video
2.35
(for self-study or
instructor assignments)
30 Global Megaprojects
https://www.popularmechanics.com/technology/g2 12 1/the-worlds-
Website
25-most-impressive-megaprojects/
CE
SP
SS
ETE
Review of Forward and Backward Pass
https://youtu.be/40DLMs 1 1Exs
Video
Tze
Project Crashing in Microsoft Project
https://youtu.be/TCNS Yur2000
Video
6:03
TSE
TE
aE CPPPT
TN
EL
TE
LC
TE
A
IEET
TETT TSE IIT SD EE TE EPODOT EIT EEE ESTE LCE SIE,
hs|| 284| Part Four
Capacity and Scheduling
SOLVED PROBLEMS
Problem
1. Constant-Time Network A list of activities, precedence relations, and activity times
for a project is given below:
Activity Name
Predecessor
Activities
A
_
5
B
A
4
C
B
Z
D
A, ©
6
8
E
D
ie
E
5
G
iS
4
H
|
J
K
D, E, G, |
e
G,H
Ranke
13
2
1
6
a. Draw an activity-on-node network diagram.
Sy.
C
Compute the early start (ES), late start (LS), early finish (EF), and late finish (LF)
times for each activity in the network.
. What is the critical path? What is the expected completion time for the project?
Solution
b. Activity
Name
ES
LS
EE
LF
Slack
A
B
©
D
E
=
G
H
O
5
9
HM
Wy,
25)
11
25
11
38
39
5
9
11
7
25
30
1S
38
18
39
45
5
9
11
ily;
25
39
25
38
25
39
45
()
O
O
O
J
O
5
9
11
WW
34
21
25
23
38
39
K
O
9
10
O
2
O
O
Chapter 13
Project Planning and Scheduling
|285 ||)
c. The critical path is the activities that have no slack:
A-B-C-D-E-H-J-K
Problem
2. Constant-Time Network
The times are shown below for each activity.
a. What is the critical path for the project?
b. What is the expected completion time of the project?
c. What is the slack in each activity?
Solution
The forward and backward passes are shown on the figure below with early start, early
finish, late start, and late finish times.
ES EF
i |] 286. Part Four
Capacity and Scheduling
a. The critical path is A-C-D-F. This is the longest path in the network
ES = LS, EF = LF for all activities on the critical path.
and
b. Expected completion time is 24, which is the LF of the last activity.
c. Activity B has a slack of 11 and E has a slack of 2. For all other activities slack is
Zero.
Problem
3. CPM Network
Activity
A
B
Cc
D
The following information is given for a CPM network.
Predecessor
Successor
Normal
Time
Normal
Cost
Crash
Time
Crash
Cost
—
A
A,B
BES
BAG
DIG
D
—
4
3
5
2
$50
60
70
30
2
2
3
1
$100
80
140
60
Draw the network showing the normal times.
. Make a forward and backward pass to calculate ES, LS, EF, LF for each activity.
. What are the normal project completion time and the normal cost?
stp
Ge
. What should be done to crash the network by one day? By two days?
Solution
a.
b. Activity
ES
EF
LS
LF
A
B
G
D
0
4
7
12
4
7
di2
14
0
4
7
12
4
u
12
14
c. Normal project completion time = 14, normal cost = $210.
d. To crash by one day, any of the activities on the critical path (A, B, C, or D) can be
crashed. The cost of crashing per day is (A = $25, B = $20, C = $35, D = $30). The
lowest cost of crashing one day is therefore activity B. In this case project cost is
$210 + $20 + $25 = $230 for the 13-day project completion.
To crash two days, also consider the same critical path, since no other path is longer than 13 days, Activity A can be crashed for one day with the next lowest cost of
$25. The total cost for the 12 day project completion is $210 + $25 + 20 = $255.
Chapter 13
Project Planning and Scheduling
|287 ||
Discussion Questions
le How does project scheduling differ from the scheduling
of ongoing operations?
How would you, after the fact, audit a project to determine whether it was successful?
Give three examples of projects not given in the text. Are
these projects unique, or are they repeated in some way?
. Contrast and compare CPM and constant-time network
as project-scheduling techniques.
. Define the term critical path.
. What is the management significance of the critical
path through a network?
How is the Gantt chart used as a scheduling tool? When
should the Gantt chart be used in preference to networkbased methods?
What is meant by the need to make trade-offs between
cost, performance, and schedule? Give examples.
Why are a forward pass and a backward pass needed to
produce a project schedule?
. What is the definition of the early and late start times
of an activity and the early and late finish times of an
activity?
Problems
iN A public accounting firm requires the following activities for an audit:
Immediate
Activity
Predecessor
Activity
Time
8
a. Draw a network for this project.
b. Make a forward and backward pass to determine
ES, LS, EF, and LF.
c. Calculate slack.
d. Prepare a Gantt chart for this project.
. A construction project has the following network and
activity times:
a
ee
a. Make a forward pass and a backward pass on the
activity times.
@)
b. Find the slack for each activity.
DY
QO7m7mmMooOW
90
wo
mop
AONBRNUN
d. Activity D will be delayed by one day. What effect
will this have on the project?
a. Draw a network for this project.
b. Make a forward pass and a backward pass to determine ES, LS, EF, and LF.
c. What are the critical path and the project completion
time?
d. If the project completion must be reduced by two
days, which activities might be affected?
. The following activities are required in starting up a
new plant:
Activity
A
B
c
D
E
fF
G
c. Prepare a Gantt chart for this project.
Immediate
Predecessor
Activity
Time
=
=—
A
B
B
© |B
ECD
3
2
2
5
4
wv
3
e|| 288| Part Four
Capacity and Scheduling
a. Find the completion time of the project.
4. A clinic start-up is based on the following network:
b. What is the critical path?
Immediate
Predecessor
Activity
Time
A
os
4
B
€
D
E
5
G
—
_
A
A
Se
B,D
8
3
3
6
5
6
Activity
7. Construction of a building is based on the following
CPM network:
Activity
Normal
Normal
Predecessor
Successor
Time
Cost
es
B,C, D
CID
G
$100
80
40
80
80
60
ERD
_
6
8
2
3
5
5
6
a. Draw a network for this project.
b. What is the project completion time?
c. Identify the critical path.
ees
desia
ye
Ye
doL
(le
ah
(Aye
. An entrepreneur is starting a new business. The activities and times required are given below:
120
Crash
Time
Crash
Cost
|=
NON
|]?
WN
NN
$150
140
60
120
140
100
160
Draw the network for this project and label the
activities.
What are the normal project completion time and
normal cost?
Identify the critical path.
d. How much will it cost to crash the project completion by one day? by two days?
e.
What is the minimum time for project completion?
The professor from problem 6 is considering reducing
the time for completion of the project. He can hire a
teaching assistant (TA) to help with the new course,
but this will cost additional money. The TA can help
reduce the times only for activities A, D, and C. In each
case the TA can reduce the time required by one day for
$150 and two days for $300.
a.
Which activity should the TA perform to reduce the
project completion time by one day?
Can the network be crashed by two days in total?
Explain why or why not.
a. Find the critical path and the project completion time.
b. What are the ES, LS, EF, and LF for each activity?
A software development project is based on the
following:
c. How much slack is there in activity D?
6. In preparing to teach a new course, the professor has
estimated the following activity times.
Activity
Normal
Time
Normal
Cost
€
8
D
E
10
2
E
—
4
15
Predecessor
Successor
A
—
B
ic
—
A
D
E
B
c,D
Crash
Time
Crash
Cost
$200
7
$300
300
800
6
1
400
1000
200
500
2
Wi
300
800
Draw the network for this software development
project.
What are the normal project completion time and
normal cost?
What is the critical path?
d. Teams of software engineers working on activities
A and B agree to combine their efforts and to label
their new activity as “F.” They believe that, by working together, they can complete their combined work
in nine periods. Draw the new network.
Chapter 13
e. What is the new project completion time and the
new critical path?
a. If she decides to crash the project by one day, what
activity should be crashed?
b. What is the total project cost after crashing by one
day?
d. What is the total project cost after crashing by two
days?
11. The project manager for making an action movie
involving a very highly paid star is very concerned
about managing the project well. The following information is provided.
A
Successor
_
Normal
Normal
Crash
Crash
Time
Cost
Time
Cost
4
$2000
Not
_
available
B
A
2
3000
1
$7000
Cc
D
E
F
B
B
c,D
E
3
4
5
3
1000
2000
2000
1000
1
2
4
= Not
2000
3000
6000
_
G
H
|
J
E
F,G
H
|
2
3
4
3
4000
2000
1000
1000
available
=
1
22
2
2
b. What are the normal project completion time and
normal cost?
Cc.
12
What is the critical path?
. Using the project information provided in problem 11,
answer the following:
a.
To shorten the project by one day, what activity
should be crashed?
What is the total project cost after crashing by one
day?
c. If she decides to crash by two days, what activity
should be crashed next?
Predecessor
289 ||)
Complete the successor column.
10. Using the original information in problem 9, the project manager determines it is necessary to shorten the
completion time for the project.
Activity
Project Planning and Scheduling
5000
4000
4000
3000
~~
———
\
~
c
on)
wou
P
rT ore
a
Si eed
;
oo
j
ing lacie “won
wat
:
nosworiotn
—_
tearehiorsdi-odtee.
{
Ai
a
:
Gf oe
CYTE
SST
'
TSISty
ry
=
‘
WOU
i
7
;
i
eet
=
+ ia ©
:
a2
;
i
:
fF
=
a?
i?
rs
=i
{
f
—
Sg
i
te
se
Y
=>
DSH
-
=
Povttad
:
TP
ia
:
a
PART
Inventory
Hew
eee
ae
14.
Independent Demand
Inventory
15.
Materials Requirements Planning and ERP
Part Five addresses decisions and tools for managing inventory in organizations.
The discussion is organized so that inventory with independent demand is handled
in Chapter 14. Independent demand refers to the market forces that drive demand
for these items (e.g., finished goods and spare parts). Chapter 15 covers inventory
with dependent demand, which is derived from the demand for another item or
component (e.g., demand for car engines is dependent on demand for complete
cars). MRP and ERP systems are used to manage dependent demand inventories.
CG deb JN TP a le IR
Independent Demand
Inventory
meee
LEARNING
LO14.2
O
ECTIVES
oh
oe
ee
After reading this chapter, you should be able to:
Explain the costs incurred by inventory.
n independent
tee 7 Differentiate betwee
and
LO14.4
Calculate the economic order quantity and identify the underlying assumptions.
LO14.7
Describe
how inventory and service k
LO14.8
Define vendor managed inventory (VMI) and the ABC system.
LO14.9
Solve advanced inventory problems.
Inventory may be the most visible sign of supply chain management for end consumers. A
case in point is that college students expect their favorite foods to be available when they
are picking up groceries. Few things disappoint consumers more than advertised products
that are out of stock when a customer shops in a retail store or online. If the supply chain
works well, goods are available when and where people need them.
Inventory management is among the most important operations management responsibilities because inventory requires a great deal of capital and affects the delivery of goods
Fe
Chapter 14
Independent Demand Inventory
|293 ||)
to customers. Inventory management affects all business functions, including operations;
marketing, which needs inventory for sales; accounting and information systems, which
track inventory; and finance, which provides necessary funds. Decisions related to managing inventory can be improved significantly through the use of the basic tools presented in
this chapter.
Many technologies are used to help manage inventory. Bar coding, for example, is
ubiquitous in most organizations (components are scanned in the production process;
nurses scan drugs before administering them to patients) and has reduced hand counting
and recording of inventory data significantly. Bar coding enables the use of point of sale
data, which are collected as items are scanned and sold, by firms and their supply chains.
Radio-frequency identification (RFID) is a technology for tracking the movement of
goods. The ability to locate inventory is valuable for both accounting and loss prevention.
Technologies such as these help to track and manage inventories in complex global supply
chains. See the Learning Enrichment feature at the end of the chapter for a video on how
Walmart uses robots to monitor store inventory.
14.1.
DEFINITION OF INVENTORY
LO14.1 Define
inventory types
and the purpose of
inventory.
FIGURE
14.1
Inventory is a stock of materials used to facilitate production or satisfy customer demands.
Typical inventories include raw materials, work in process, and finished goods. In Figure 14.1,
an operation is shown as a materials-flow process with raw materials inventories waiting
to enter the production process, work-in-process inventories in an intermediate stage of
transformation, and finished goods inventories that have been transformed completely by
the production process.
Inventory stocks are located at various points in the production process, with flows of
materials connecting one stock point to another. The rate at which stock is replenished is
the supply, and the rate of stock depletion is demand. Inventory acts as a buffer between the
supply rate and the demand rate.
The water tank shown in Figure 14.2 is a good analogy for these concepts of flows and
stocks. In this figure, the level of water in the tank corresponds to inventory. The rate of
flow into the tank is analogous to the supply rate, and the rate of flow out of the tank corresponds to the demand rate. The water level (inventory) is the buffer between supply and
demand. If the demand rate exceeds the supply rate, the water level drops until the supply
A materials-flow process.
Productive Process
Work in
process
Suppliers
Raw
materials
Work in
process
Work in
process
Finished
goods
Customer
P| | 294)
Part Five
FIGURE
Inventory
14.2
A water tank analogy
for inventory.
and demand rates come back into balance or until the water (inventory) is depleted. If the
supply rate exceeds the demand rate, the water (inventory) level rises.
Imagine a series of these tanks connected to one another, all with varying rates of inputs
and outputs. This situation, which is illustrated in Figure 14.3, is analogous to the challenge of inventory management. Here, one tank represents raw materials, there are two
tanks for work in process, and there is one tank for finished goods. The tanks serve as buf-
fers to absorb variations in flow rates within this pseudo-production system.
Inventories in the supply chain serve the same purpose as inventories in the factory—to
buffer the difference in flows between supply and demand. However, in a supply chain,
typically, one firm does not control all inventories; rather, inventory must be coordinated
across the supply chain partners. Many of the concepts covered in this chapter apply to the
broader context of the supply chain, in which coordination between parties with different
interests and objectives must also be considered.
Inventory is often described as a “necessary evil” and a form of waste. This perspective has led many firms to strive relentlessly to reduce their inventory levels of items that
they manage, often without careful consideration of items themselves (e.g., uncertainty,
stockouts, etc.).
Two points deserve emphasis. First, reducing excessive inventory is a laudable goal;
eliminating inventory completely is not wise. Inventory is needed as a buffer to prevent
stockouts whenever the demand rate is not equal to the supply rate. Second, within a supply chain there will be demand and supply variation seen by each member of the supply
chain. One member of the supply chain cannot reduce inventory without effecting the other
members, possibly to carry more inventory. Thus, inventory changes will be felt across the
supply chain.
FIGURE
14.3
Water tanks as pseudo-production system.
Rawmaterial
Supplier
usage rate
‘|
First-stage
|
Secondstage
output rate
|
]} output rate
Customer
supply
demand
rate
rate
Lowquality
materials
Scrap
Scrap
Scrap
Chapter 14
14.2
Independent Demand Inventory
295 |||
PURPOSE OF INVENTORIES
The primary purpose of inventories is to uncouple the various phases of operations and the
supply chain. Raw materials inventory uncouples a manufacturer from its suppliers, meaning that the supplier can produce parts at a convenient time within its own schedule and
the manufacturer can later use those materials at the appropriate time for its transformation
process. Similarly, work-in-process inventory uncouples the various stages of manufacturing, and finished goods inventory uncouples a manufacturer from its customers.
There are four primary reasons for firms to carry inventory:
1. To protect against uncertainties. In inventory systems, there are uncertainties in supply, demand, and lead time. Safety stock is inventory that is maintained to protect against
these uncertainties. If customer demand is known, it is feasible—although not necessarily
economical—to produce at the same rate as consumption. In this case, no finished goods
inventory would be needed; however, every change in demand would cause the production
system also to change, resulting in a very uneven workload. Instead of such tight coupling,
safety stocks of finished goods are maintained to absorb changes in demand so that production
can maintain a separate, more even pace. In a similar way, safety stocks of raw materials are
maintained to absorb uncertainties in delivery by suppliers in terms of both the quantity and
the timing of delivery. Safety stocks of work-in-process inventories are maintained to allow for
unexpected breakdowns, unreliable workers, and schedule changes. Most safety stocks can be
reduced by improving coordination with suppliers and customers in the supply chain.
2. To allow economic production and purchase. It is often economical to produce
inventory in lots (batches) as it allows production at one point in time, and then no further
production of the same item is done until the lot is nearly depleted. This makes it possible to
spread the setup cost of production over a large number of items. Producing or ordering in
lots also permits the use of the same production equipment for different products. A similar
benefit holds for the purchase of raw materials. Owing to ordering costs, quantity discounts,
and transportation costs, it is sometimes economical to purchase in large lots even though
part of a lot is held in inventory for later use. The inventory resulting from the purchase or
production of material in lots is called cycle inventory, since the lots are produced or purchased on a cyclic basis. Most firms are working to reduce setup times and costs by altering
the product or process. This effort can result in smaller lot sizes and much lower inventories.
3. To cover anticipated changes in demand or supply. There are several situations
in which changes in demand or supply are expected, causing firms to hold anticipation
inventory. Expected changes in the price or availability of raw materials may cause stockpiling of raw materials; for example, firms often stockpile steel before an expected strike in
the steel industry. Another source of anticipation is a planned market promotion in which a
large amount of finished goods may be stocked before a sale. Firms in seasonal businesses
often hold anticipation inventory in order to smooth employment. For example, a producer
of air conditioners may use a level production strategy even though most air conditioners
are sold during the summer.
4. To provide for transit. Inventories that are moving from one point to another in
the supply chain are called pipeline inventory or transit inventory. These inventories are
affected by production location decisions and by the choice of carrier. These inventories
can be significant in size when ships rather than planes are used to transport goods across
the world. Obviously, firms compare the cost and time factors in evaluating various transportation options. Time-sensitive goods may be more profitable when transported by using
costly but rapid air freight because they reach the market so quickly.
i] | 296 | Part Five
Inventory
Overall, it is easy to see that there are many good reasons for firms to hold inventories.
Inventory helps firms satisfy customer demand while also allowing a firm and its supply
chain partners to schedule production economically.
14.3
LO14.2
COSTS OF INVENTORY
Explain the
costs incurred by
inventory.
Many inventory decisions can be made by using economic criteria. One of the most important prerequisites, however, is an understanding of the relevant costs. Inventory cost structures incorporate the following four types of costs:
1. Item cost. This is the cost of buying or producing the individual inventory items.
The item cost usually is expressed as a cost per unit multiplied by the quantity procured or
produced. Item cost may be discounted if enough units are purchased at one time.
2. Ordering (or setup) cost. The ordering cost is incurred when ordering a lot (batch)
of items and generally does not depend on the lot size ordered; it is assigned to the entire
batch. This cost includes creating the purchase order, expediting the order, transportation
costs, receiving costs, and so on. When
the item is produced within the firm, there are
also costs associated with placing an order that are independent of the number of items
produced. This is called the setup cost, and it includes the costs to set up the production
equipment for a run as well as record-keeping costs. In some cases, setup costs can be
thousands of dollars, leading to significant economies for large batches. Setup cost often
is considered fixed when, in fact, it can be reduced by changing the way operations are
designed and managed. Reducing setup time is an important way to align the production
rate with the demand rate, and thereby reduce inventory.
3. Carrying (or holding) cost. The carrying or holding cost is associated with keeping
items in inventory for.a period of time. Typically, the carrying cost is charged as a percentage of dollar value per unit time. For example, a 15 percent annual holding cost means that
it costs 15 cents to hold $1 of inventory for a year. In practice, holding costs often range
from 15 to 30 percent per year. The carrying cost usually consists of three components:
*
Cost of capital. When items are carried in inventory, the capital invested in them is
not available for other purposes. This represents a cost of forgone opportunities for
other investments, which is assigned to inventory as an opportunity cost.
*
Cost of storage. This cost includes variable space cost, insurance, and taxes.
¢
Costs of obsolescence, deterioration, and loss. Obsolescence costs are assigned
to items that have a high risk of becoming obsolete, for example, fashion and tech-
nology items that quickly lose their appeal in the market. Perishable products are
charged with deterioration costs when they deteriorate over time, for example, food
and blood. Many products have an expiration date printed on them and become obsolete at that time. The costs of loss include pilferage and breakage costs associated
with holding items in inventory.
4. Stockout cost. Stockout cost reflects the economic consequences of running out of
stock. As a result, the current sale can be lost if the customer will not accept a backorder
and/or a dissatisfied customer may be less likely to buy in the future.
In light of these costs, it should be apparent why firms do not want to hold more inventory than is necessary to serve their customers. Also, it is easy to see why inventory
management is a cross-functional challenge. Marketing may be particularly interested in
minimizing the stockout costs associated with lost sales. Accounting and finance may be
Chapter 14
Independent
Demand Inventory
297 ||)
OPERATIONS LEADER
Target Sets the Standard for Inventory Management
- With inventory valued at more than $5 million in each
Target manages the flow of finished goods invenretail store, Target Corporation stands out as one of the _ tory from the factory to the end consumer in more than
most advanced retailing firms in managing inventory in
1800 stores in all 50 states, and via online sales with
the United States.
ship-to-home or -store options. Its supply chain begins in
thousands of factories around the world.
Target has more than 40 distribution centers. Some
centers while others ship
directly to stores. It is all about getting the right inventory
to the right place at the right time.
In-store inventory management is technology driven.
Point of sale (POS) data are used to derive a list of sold
items every hour so that shelves can be restocked immediately. Orders are triggered by either stock levels or
percentages of stock sold. Workers use hand-held technology to provide customers with immediate information
on inventory availability in other stores or online.
suppliers ship to distribution
artzenter/Shutterstock
Source: www.target.com,
2020.
interested in minimizing the amount of inventory that has to be financed and held. Operations may want a sufficient level of inventory to assure smooth scheduling and production
control. Since these objectives may be at odds, it is important that the total cost minimization approach be taken of the four costs described above.
The minimum cost principle also applies when supply chain partners attempt to minimize the total cost across the entire supply chain. However, it may be difficult to achieve
a minimum when lower costs to one party incur higher costs for another. Nevertheless,
market forces, financial incentives, and negotiations among parties can be used to seek the
minimum cost across the supply chain. The Operations Leader box explains how Target
Corporation manages inventory in its complex supply chain.
14.4
INDEPENDENT VERSUS DEPENDENT DEMAND
LO14.3 Differentiate
between
independent and
dependent demand.
A crucial distinction in inventory management is whether demand is independent or
dependent. Independent demand is influenced by market conditions outside the firm; it is
therefore independent of demand for any other inventory items. Finished goods inventories
and spare parts for replacement usually have independent demand. Dependent demand
items have demand that is related to another item and is not independently determined by
the market. When final products are assembled from components, the demand for those
components is dependent on the demand for the final product.
A toy wagon illustrates the difference between independent demand and dependent
demand. Demand for wagons is independent because it is influenced by the market and
must be forecast. Demand for wagon wheels is dependent because it has a direct mathematical relationship to the demand for wagons; it takes four wheels to complete each wagon.
Likewise, demand for wagon handles is dependent on the demand for finished wagons.
1] | 298) Part Five
Inventory
FIGURE 14.4
Demand patterns.
Independent demand
Finished goods
Dependent demand
Spare parts
Raw materials
Work in process
Usage
Demand
Usage
Demand
Time
Time
Independent and dependent demand items exhibit very different usage or demand patterns. Since independent demand is subject to market forces, it often exhibits both a fixed
pattern and random influences stemming from customer preferences. In contrast, dependent demand exhibits a lumpy on-again, off-again pattern because production is scheduled
in lots. A quantity of parts is required when a lot is made; then no parts are required until
the next lot is produced. These demand patterns are shown in Figure 14.4.
Different demand patterns call for different approaches to inventory management. For
independent demand, a replenishment philosophy is appropriate. As the stock is used, it
is replenished so that products are always on hand for customers. Thus, as inventory begins
to run out, an order is triggered for more material, either by ordering from a supplier or by
producing more, and the inventory is replenished.
For dependent demand, a requirements philosophy is used. The amount of stock ordered
is based on requirements for higher-level items. As dependent demand items or materials are
used, additional inventory is not ordered. More items or materials are ordered only as required
by the scheduled production for the higher-level or end items they are used to produce.
The nature of demand therefore leads to two different philosophies of inventory management, and those philosophies generate different sets of methods and software. In this
chapter, decisions related to independent demand items are covered, including the following types of inventories:
1. Finished goods inventories and spare parts in manufacturing firms.
2. Maintenance, repair, and operating supplies (MRO) inventories.
3. Retail and wholesale finished goods.
4. Service industry (e.g., hospitals and schools) inventory.
14.5
LO14.4
ECONOMIC ORDER QUANTITY
Calculate
the economic order
quantity and identify
the underlying
assumptions.
The economic order quantity (EOQ) and its variations are still widely used in industry
for independent demand inventory management. The EOQ helps to determine a logical
order size or batch size for a single inventory item. It does this by balancing cost trade-offs.
The EOQ model is based on the following assumptions:
1. The demand rate is constant, recurring, and known. For example, demand (or usage) is
100 units per day with no random variation, and demand is assumed to continue into the
indefinite future.
2. The lead time is constant and known. The lead time, from order placement to order
delivery, is therefore always a fixed number of days.
Chapter 14
FIGURE 14.5
EOQ inventory levels.
Independent Demand Inventory
/299 ||]
Order
interval
——
=
e
sf
Average inventory
.
Lot size = Q:
level = Q/2
s)
Time
3. No stockouts are allowed. Since demand and lead time are constant, one can determine
exactly when to order inventory to avoid stockouts.
4. Items or materials are ordered or produced in a lot or batch, and the lot is placed into
inventory all at one time.
5. The unit item cost is constant, and no discounts are given for large purchases. The carrying cost is linearly related to the average inventory level. The ordering or setup cost
for each lot is fixed and is independent of the number of items in that lot.
6. The item is an independent item, without interactions with other inventory items.
Under these assumptions, the inventory level over time is as shown in Figure 14.5,
where the dark line represents the amount of inventory on hand over time. Notice that the
figure shows a perfect sawtooth pattern, because demand is assumed to be constant and
items are ordered in fixed lot sizes.
In choosing the lot size, there is a trade-off between ordering frequency and inventory
level. Small lots will lead to frequent reorders and a low average inventory level. If larger
lots are ordered, the ordering frequency will decrease but more inventory will be carried.
This trade-off between ordering frequency and inventory level can be represented by a
mathematical equation using the following symbols:
D = demand rate, units per year
S = cost per order placed or setup cost, dollars per order
C = unit cost, dollars per unit
i = carrying rate, percentage of dollar value per year
QO = lot size, units
TC = total of ordering cost plus carrying cost, dollars per year
The annual ordering cost is
Ordering cost per year = (cost per order) X (orders per year) = SD/Q
In the above equation, D is the total demand for a year, and the item is ordered Q units at a
time; thus, D/Q orders are placed in a year. This is multiplied by S, the cost per order placed.
The annual carrying cost is:
Carrying cost per year = (annual carrying rate) X (unit cost)
X (average inventory) = iCQ/2
In this equation, the average inventory is Q/2.
A maximum of Q units is carried in inven-
tory (when a new batch arrives); the minimum amount carried is zero units. Since the stock
is depleted at a constant rate, the average inventory is Q/2. The carrying rate per year (/)
1} | 300] Part Five
FIGURE
Inventory
14.6
Total cost curve.*
Total Cost
—
(SD/Q + iCQ/2)
Cost
Minimum
($/year,
TC)
Cost
Annual
7
EOQ
Carrying Cost
(iCQ/2)
Ordering cost
a
(SD/Q)
Q (units)
*Notice that the item cost of procurement is the constant CD, which is independent of Q and therefore can be removed from further
consideration. It will not affect the minimum of TC.
multiplied by the unit cost (C) gives the cost of holding one unit in inventory per year. This
unit charge multiplied by the average inventory level gives the total annual carrying cost.
Given the annual ordering and carrying costs above, the total cost of an inventory item is
TC = SD/Q + iCQ/2
(14.1)
Figure 14.6 is a plot of TC versus Q, showing the carrying and ordering costs along
with the total, which is the sum of the other two lines. As Q increases, the annual ordering
cost decreases because fewer orders are placed per year; at the same time, however, the
annual carrying cost increases because more inventory is held. Ordering costs and carrying
costs, therefore, offset one another; one decreases while the other increases. Because of
this trade-off, the function TC has a minimum.
Finding the value of Q that minimizes TC is a classic problem in calculus. We take the
derivative of TC, set it equal to zero, and then solve for the resulting value of Q:
Gr
tle + eo=0
SD
ie
On
gi2
One ai
ya OSD
7 Ga
25D
Oe pega
(14.2)
Equation (14.2) is the economic order quantity (EOQ), which minimizes the cost of man-
aging an item in inventory. Although we have minimized cost on an annual basis, any
unit of time can be used provided that the demand rate and carrying rate are compatible.
For example, if demand is expressed on a monthly basis, the carrying rate also must be
expressed on a monthly basis.
Although the EOQ formula is based on rather restrictive assumptions, it is a useful
approximation in practice. The formula provides a ballpark figure as long as the assumptions are reasonably accurate. Furthermore, the total cost curve is rather flat in the region
Chapter 14
Example
Independent Demand Inventory
3014] |
To illustrate the use of the EOQ formula, suppose we are managing a carpet store and want
to determine how many yards of a certain type of carpet to buy. The carpet has the following characteristics:
D = 360 yards per year
S = $10 per order
i = 25 percent per year
C = $18 per yard
Thus:
Q=
2(10)(360)
= V 1600
.25(18)
= 40 yards
The manager should order 40 yards of carpet at a time. This will result in D/Q = 360/40 = 9
orders per year, or one order every 1.33 months.
The minimum cost of managing this inventory will be $180 per year, as follows:
TC = 10(360/40) + .25(18)(40/2) = 90 + 90 = 180
Notice that the minimum cost occurs when the annual ordering cost component equals the
annual carrying cost component.
The total cost curve for inventory is very flat in the neighborhood of the minimum. For
example, if 50 units of carpet are ordered instead of the EOQ of 40 units, a 25% increase,
the change in total cost is small, about a 2.5 percent increase. Thus, while calculating the
EOQ provides a good estimate for planning order size, the inventory manager can adjust
the order quantity if necessary, with a limited effect on the total cost of maintaining an
inventory item.
of the minimum; thus, the order quantity can be adjusted somewhat based on circumstances without greatly affecting total costs.
It is very important that inventory decisions be made by considering total cost.
Regardless of the situation, if one can identify the relevant total cost equation, an economic lot size can be found. The idea of minimizing the total cost equation is basic to
all lot-sizing formulas. For example, the supplement to this chapter shows the total cost
equation and associated minimization procedure when price discounts are available for
large orders.
Hewlett-Packard has extended the total cost concept to its entire
supply chain. In producing personal computers (PCs), it identifies
four relevant costs:
* Component devaluation costs.
¢ Price protection costs (lowest price guarantees given to retailers).
¢ Product return costs.
¢ Obsolescence costs (end-of-life write-off).
Poorly managed inventory may result in the
need to discount prices.
TY Lim/Shutterstock
All these costs are associated with the declining value of a PC once it
is placed in inventory due to the risk of not selling the product before a
new model is introduced to the market. We referred to these costs previously as the cost of obsolescence and deterioration that are included as
part of holding cost. These costs often exceed
the profit margin on a prodoe
uct and must be considered in establishing the EOQ amount.
| |302,
Part Five
Inventory
The EOQ is used frequently in manufacturing to calculate appropriate order sizes (from
suppliers) and lot sizes (for production), and it also is used in many service industry supply
chains. Restaurants use EOQ for estimating order sizes for food and other supplies, larger
firms use EOQ to manage their office supply stocks, and pharmaceutical mail order firms
use EOQ to estimate order sizes to replenish their warehouses. Below, we describe two
inventory management systems that are based on the EOQ model.
14.6
LO14.5
CONTINUOUS
Compute
the parameters for
a continuous
review
and periodic review
inventory control
system.
REVIEW SYSTEM
In practice, one of the most serious limitations of the EOQ model is the assumption of constant demand. This assumption can be relaxed to design a practical system for managing
inventory that allows for random demand. We build a system based on the EOQ that is sufficiently flexible to use in practice for independent demand items. All EOQ assumptions
except constant demand and no stockouts remain in effect.
In managing inventory, decisions about when to reorder stock are based on the total of
on-hand inventory plus inventory that is on order. On-order inventory is counted the same
as on-hand inventory for reorder decisions because the on-order inventory is scheduled
and expected to arrive. The total of on-hand inventory and on-order inventory is called the
stock position. Be careful on this point! A common mistake in inventory calculations is
failure to consider amounts already on order.
In a continuous review system (also known as a fixed order quantity system or the Q
system), the stock position is monitored after each transaction, or continuously. When the
stock position drops to a predetermined level, or reorder point, an order is placed for a fixed
quantity. Since the order quantity is fixed, the time between orders varies in accordance
with the random nature of demand. See the Operations Leader box on Cantaloupe Systems, a provider of hardware and software to monitor vending machine inventory levels.
A formal definition of the decision rule embedded in the Q system is as follows:
Continually review the stock position (on hand plus on order). When the stock position drops
to the reorder point R, the fixed quantity Q is ordered.
A graph of this system is shown in Figure 14.7. The stock position drops as inventory is
used to fulfill irregular demand until it reaches the reorder point, R, when an order for Q
units is placed. The order arrives later, after a lead time, L, and the cycle of usage, reorder,
and order arrival is repeated.
The Q system is completely determined by two parameters, Q and R. In practice, these
parameters are set by using certain simplifying assumptions. First, Q is set equal to the
EOQ value from Equation (14.2). In more complex models, Q and R must be determined
FIGURE
14.7
A continuous review
system (or Q system).
Position
Stock
Time
Chapter 14
Independent Demand Inventory
|303 |{7
nua
OPERATIONS LEADER
Cc.
Are Your Vending Machines Full or Empty? Ask Cantaloupe Systems
cus-
Cantaloupe Systems builds and installs a device that
tomers find what they want to buy. Keeping track of items
that are in stock and replenishing items that have sold
out are therefore activities crucial to making money from
owning vending machines.
Owners
of vending
sits inside a vending machine. This device, called a seed
McGraw-Hill Education
machines
make
money
when
device,
monitors
all transactions
on
an
ongoing
and transmits the data wirelessly to servers
basis
hosted
by
Cantaloupe Systems.
The seed device acts essentially like a continuous
review system for monitoring inventory of snacks and
beverages sold through vending machines. When an
item nearly sells out, the device notifies the owner of the
vending machine. Cantaloupe Systems expects owners
of vending machines with its seed devices to save an
average of $35,000 annually per route. Think of a route
as comprising a number of vending machine locations
that a replenishment truck has to traverse. The seed
devices from Cantaloupe Systems can, moreover, help
to reduce fuel consumption by approximately 40 percent
since replenishment trucks can avoid making unnecessary stops and to be able to refill 80 percent more vending machines per week.
Source: www.cantaloupesys.com, 2019.
simultaneously. However, using the EOQ to estimate Q is a reasonable approximation provided that demand 1s not highly uncertain.
The value of R is based on either the stockout cost or the stockout probability. Formulations that utilize the stockout cost become quite difficult to estimate, and so the stockout
probability is commonly used to determine R.
To calculate R, management must determine a desired service level, which is the percentage of customer demand satisfied from inventory. The service level also is called the
fill rate. A 100 percent service level means that all customer demand is satisfied from
inventory, but as we will see shortly, this is nearly impossible to achieve. The stockout
percentage is equal to 100 minus the service level.
There are three ways to express service level:
1. Service level is the probability that all orders are filled from stock during the replenishment lead time of one reorder cycle.
2. Service level is the percentage of demand filled from stock during a particular period of
time (e.g., one year).
3. Service level is the percentage of time the system has stock on hand.
Each of these definitions of service level leads to slightly different reorder points. Furthermore, one must determine whether to count customers, units, or orders when applying
these definitions. In this text, for the sake of simplicity, only the first definition of service
level will be used.
The reorder point is based on the notion of a probability distribution of demand during
the lead time. As inventory is used (depleted), eventually the inventory manager places
1] | 304
Part Five
Inventory
FIGURE 14.8
Probability distribution
of demand during lead
time.
Service-level probability
Stockout
probability
Frequency
Demand during Lead Time
an order. But until the order arrives, the inventory system is exposed to the potential for a
stockout. Therefore, the only risk of a stockout is during the replenishment lead time.
Figure 14.8 shows a typical probability distribution of independent demand during the
lead time. We must know the statistical demand distribution during the lead time to estimate R (we use a reasonable assumption of a normal distribution here). The reorder point
R in the figure can be set to any desired service level.
The reorder point is defined as follows:
R=m+s
(14.3)
where
{
R = reorder point
m = mean (average) demand during the lead time
s = safety stock (or buffer stock)
We can express safety stock as:
where
z = Safety factor
o = standard deviation of demand during the lead time
= Vleadtime X (o single baie)
Then we have:
R=m+2z0
Thus, the reorder point is set equal to the average demand during the lead time (m) plus a
specified number (z) of standard deviations (0) to protect against stockouts. By deciding z,
the number of standard deviations, the firm is setting both the reorder point and the service
level. A high value of z results in a high reorder point and a high service level.
The values in Table 14.1 are from the normal distribution. These service levels represent
the probability that the demand during the lead time will be satisfied by using safety stock.
That is the same as saying that demand during the lead time will fall within the specitied
number of standard deviations (z) from the mean. When a firm decides what the service
level should be, the corresponding z from Table 14.1 is used to calculate the reorder point.
Chapter 14
TABLE 14.1
Normal Demand
Percentages
ics
0)
45)
Service Level (%)
ee
ad
doa
R2
1s;
1.4
ES)
1.6
86.4
"88:5
90.3
91.9
93.3
eyBl
Z
sere
1.8
1.9
96.4
97.1
2.0
2
:
1356
145)
9.7
8.1
6.7
DES)
ould
-
ae
4.5
i
3.6
2.9
SET
2.3--
98.2
1.8
22
26,
98.6
98.9
1.4
1.1
2.4
99.2
8
2a
2.6
99.4
99.5
6
Dies
IF,
Example
15.9
;
95.5
a
?
[305 || F
SOMOS
30.9
84.1
Pe s
Demand Inventory
_ Stockout (%)
oats ORO s
5 e094
1.0
Independent
99.6
(qe
2.8
2.9
99.7
99.8
3
2
30
99.9
1
An example will help cement these ideas. Suppose we are managing a warehouse that distributes a particular breakfast food to retailers. The breakfast food has the following characteristics:
Average demand =
Lead time =
Standard deviation of daily demand =
Desired service level =
S =
i=
200 cases per day
4 days to receive the order form the supplier
150 cases
95%
$20 per order
20 % per year
C = $10 per case
Assume that a continuous review system is used and that the warehouse is open
5 days per week, 50 weeks per year, or 250 days per year. Then average annual
demand = 250(200) = 50,000 cases per year.
The economic order quantity is calculated by using Equation (14.2):
o= 2eareocee)
) — 7,000,000 = 1000 cases
.2(10)
The average demand during the lead time is 200 cases per day for four days; therefore,
m = 4(200) = 800 cases. The standard deviation of daily demand is 150 cases, but we
need to calculate the standard deviation over the four-day lead time. This is done with a
simple conversion:
o = Vieadtime x (a single period ) =
V4 x (150) = 300 units
The 95 percent level requires a safety factor of z= 1.65 (see Table 14.1). Thus, we can
calculate the reorder point by using Equation (14.3):
R=m+Zo=
800 + 1.65(300) = 1295
Recall that in a Q system, there are just two things we need to know: when to order (R)
and how much to order (Q). The Q system we have designed here includes placing an order
for 1000 cases whenever the stock position drops to 1295 cases. On average, 50 orders
it
| | 306,
.
Part Five
Inventory
will be placed per year, and there will be an average of five working days between orders.
The actual time between orders will vary, however, with demand.
To complete this example, Table 14.2 simulates the operation of the Q system decision rule.
A series of random demands were generated on the basis of an average of 200 cases per
day and a standard deviation of 150 cases per day. It is assumed that 1100 units are on hand
at the beginning of the simulation and none are on order. An order for 1000 cases is placed
whenever the stock position reaches 1295 units, and so an order must be placed immediately
on day 1. The stock position is reviewed each day as demands occur. The result is that orders
are placed on days 1, 7, 10, and 15. The lowest inventory level is 285 units at the beginning of
day 10. Check some of the numbers in Table 14.2 to see if you can verify them.
TABLE 14.2
Q-System Example*
Day
Demand
Beginning
Period on
Hand
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
nid
PNT
334
124
)
Sil
135
208
315
0
440
N27
SiS
114
241
140
1100
989
V72
438
1314
1314
943
808
600
285
4285
845
718
1403
1289
1048
“
Beginning
Period on
Order
Beginning
Period Stock
Position
—
1000
1000
1000
—
—
_
1000
1000
1000
1000
1000
1000
_
1000
1100
1989
1772
1438
1314
1314
943
1808
1600
1285
2285
1845
1718
1403
1289
2048
Amount
Ordered
Amount
Received
1000
_
1000
1000
1000
_
_
1000
1000
1000
*For this table, we have used @ = 1000 and R = 1295.
14.7
PERIODIC REVIEW SYSTEM
Instead of reviewing the stock position on a continuous basis, an inventory management
system can be designed to review stock position periodically. Suppose a supplier makes
deliveries only at periodic intervals, for example, every two weeks. In this case, the stock
position is reviewed every two weeks and an order is placed if inventory is needed.
This inventory management system, like the Q system, is based on the EOQ model. In this
section, we assume that the stock position is reviewed periodically (on a fixed schedule) and
that the demand is random. All EOQ assumptions in Section 14.5 except constant demand
and no stockouts remain in effect.
In a periodic review system (also known as the fixed order period system, or the P
system), the stock position is reviewed at fixed intervals. When the review is performed,
the stock position is “ordered up” to a target inventory level. The target level is set to
cover demand until the next periodic review plus the delivery lead time. The order quantity
depends on how much is needed to bring the stock position up to the target level.
A formal definition of the decision rule embedded in the P system is as follows:
Review the stock position (on hand plus on order) atfixed periodic intervals P. An amount
equal to target inventory T minus the stock position is ordered at each review.
Chapter
14
Independent Demand Inventory
307 |||
FIGURE 14.9
A periodic review
system (P System).
Position
Stock
Time
A graph of this system is shown in Figure 14.9. The stock on hand drops on an irregular
basis as it is used to meet demand, until the end of the fixed periodic interval is reached. At
that time, a quantity is ordered to bring the stock position up to the target level. The order
arrives later, after lead time L, and then the cycle of usage, reorder, and order arrival repeats.
The P system is different from the Q system in several ways: (1) It does not have a reorder point but instead a target inventory level; (2) it does not have an EOQ since the order
quantity varies according to demand; and (3) in the P system, the order interval is fixed; in
a Q system, an order can be placed whenever inventory is needed.
The P system is determined by two parameters, P and T: Since P is the time between
orders, it is related to the EOQ as follows:
P= OUD
(14 .4)
where
Q = EOQ
Equation (14.4) provides an approximately optimal review interval P. We note that if
demand is highly uncertain, the approximation of P may be poor.
The target inventory level is set by specifying a service level. In this case, the target level
is set high enough to cover demand during the lead time (L) plus the periodic review interval (P). This coverage time (P + L), known also as the protection interval, is needed because
an order cannot be placed again until the end of the next review interval, and that order will
take the lead time to arrive. To achieve the specified service level, average demand must
be covered over the time P + L and the safety stock also must cover P + L. Thus, we have
T=m'+s'
where
T = target inventory level
m'’ = average demand over P + L
s’ = safety stock to cover P + L
For safety stock, we have
sou
where
z = safety factor
o’ = the standard deviation of demand over P + L
Just as in the Q system, z reflects the desired service level (see Table 14.1).
(14.5)
|] | 308, Part Five
Example
Inventory
To illustrate, we will use the breakfast food example from the previous section. Recall that
the EOQ was 1000 cases and the daily demand was 200 cases. The optimal review interval
is then
P = Q/D = 1000/200 = 5 days
In this case, m’ is the average demand over P +L = 5 + 4 = 9 days. Thus, we have
m' = 9(200) = 1800. The standard deviation is for the P + L period, or 9 days. Thus, with
the daily standard deviation = 150 and the coverage period of 9 days,
o’ = V9 x (150) = 450 cases
Therefore, with a 95 percent service level (z = 1.65):
T=
m'+zZo' = 1800 + 1.65(450) = 2542 cases
The P system is summarized as the following: The stock position is reviewed
days, and the order quantity is set to order up to a target of 2542 cases.
every five
It is interesting to note that the P system requires 1.65(450) = 742 units of safety stock,
whereas the same service level is provided by the Q system with only 1.65(300) = 495
units of safety stock. The P system always requires more safety stock than the Q system
for the same service level. This occurs because the P system must provide coverage over
a time of P + L, whereas the Q system must protect against stockout only over the lead
time L.
This example is completed by an example in Table 14.3, which uses the same demand
figures as Table 14.2. Here, however, the review is periodic instead of continuous. A review
is made in periods 1, 6, 11, and 16—that is, every five periods. The amounts ordered
are 1442, 786, 1029, and 1237. While the review period is fixed, the amount ordered is
allowed to vary. It is a good idea to practice calculating some of the numbers in the table to
see if you can verify them.
Beginning
Period on
Order
Beginning
Period
Stock
Position
Day
Demand
Beginning
Period on
Hand
1
2
&)
4
5
6
7
8
9
10
11
q2
13
14
15
111
Aa.
334
124
0
371
135
208
Bis
O
440
127
SiS
114
241
1100
989
772
438
1756
1756
1385
1250
1042
Novis
1513
1073
946
631
1546
_
1442
1442
1442
_
_
786
786
786
_
1029
1029
1029
_
1100
2431
2214
1880
1756
1756
2171
2036
1828
1 te
1513
2102
1975
1660
1546
16
140
1305
_
1305
*For this table, we have used P = 5 and JT = 2542.
Amount
Ordered
Amount
Received
1442
—
786
1442
—
1029
786
—
1029
Chapter 14
14.8
LO14.6
Independent
Demand Inventory
309 ||]
USING P AND Q SYSTEMS IN PRACTICE
Explain
In industry, both Q and P systems, as well as modifications of them, are used widely for
how continuous
independent demand inventory management. Examples of independent demand invento-
and periodic review
systems are used in
ries are in wholesale, retail, restaurants, hospitals, factory finished goods, and MRO (maintenance, repair, and operations) inventories. The choice between Q and P systems is not a
practice.
simple one and may be dictated by management practices as well as economics. However,
there are some conditions under which the P system may be preferred to the Q system:
1. The P system should be used when orders must be placed or delivered at specified intervals, for example, weekly or daily deliveries of food to grocery stores.
2. The P system should be used when multiple items are ordered from the same supplier
and delivered in the same shipment. In this case, the supplier prefers to consolidate the
items into a single order. For example, a large supplier such as Dole may consolidate
orders for a number of products when delivering to grocery warehouses.
3. The P system should be used for inexpensive items whose inventory level is monitored
only at specific intervals, not continuously. An example is the nuts or bolts used in a
manufacturing process. In this case, the bin size determines the target inventory level,
and the bin is filled at fixed intervals.
In sum, the P system provides the advantage of scheduled replenishment and less record
keeping. However, it requires a larger safety stock than does the Q system, as the previous example illustrates. Because of this larger safety stock, the Q system often is used for
expensive items where it is desirable to hold down the investment in safety stock inventory.
The choice between the Q and P systems should be made, therefore, on the basis of timing
of replenishment, the type of record-keeping system in use, and the cost of the item.
In practice, one can find hybrid systems that are mixtures of P and Q systems. One of
these systems is characterized by Min/Max decision rules and periodic review. In this case,
the system has both a reorder point (Min) and a target (Max). When the periodic review is
performed, no order is placed if the stock position is above the Min. If the stock position
is below the Min, an order is placed to raise the stock position to the Max level. See the
Operations Leader box for a successful example of a Min/Max inventory system at IKEA.
LO14.7
Describe
Service Level and Inventory Level
how inventory and
There is an important trade-off between the
service level are
service level and the inventory level. In man-
related.
aging independent demand inventories, one
of the key considerations is the level of customer service the firm wishes to maintain.
High customer service levels are clearly good
for customers (and perhaps for relationshipbuilding purposes), but they must be balanced
against the required investment in inventory,
since higher customer service levels generally
require higher inventory investments. The average inventory level / is given by
[=Q/2
+ zo
Q/2 units are carried
on
average
; when
ordering in lots of size Q, and zo units are
The P system is often used when multiple items
are ordered from a single supplier.
—Stockbyte/Getty Images
|
310, Part Five
J/nventory
sega ‘=
5am
ian
IKEA Uses Hybrid P and Q System
IKEA was founded in Sweden in 1943 and continues _ its in-store restaurants maintain both the thrifty and the
to expand its global reach today. With 424 stores in 52
Swedish themes.
Store managers use a “minimum/maximum” invencountries, it is still growing. Stores are stocked with nearly
tory replenishment system for stocking stores. The
10,000 functional and reasonably priced products. And
minimum is set to signal the need to order, with reorder
points set at the store level. The maximum is the most
of a particular product to order at one time. The logic
is to balance inventory stock within the fixed order
period, and this target is based on the number of items
that will fit in each product’s designated bin. This combined continuous
IKEA to meet
and periodic review system enables
customer demand and lower the likeli-
hood of lost sales.
Point-of-sale data help managers determine how
much inventory to order each day. Automation in warehouses further reduces manual counting and moving of
inventory. Their successful management of inventory has
made IKEA the world’s largest furniture retailer.
Tooykrub/Shutterstock
Source: www.tradegecko.com/blog, July 2, 2018.
carried on average in safety stock. (For the P system, use o’ in place of o.) Thus, the inventory level is the sum cycle stock (Q/2) and safety stock (zo).
If we fix Q, the inventory level is a function of z, which represents the service level.
Thus, we can vary z and plot the service level versus the average inventory required, as
shown in Figure 14.10.
The figure shows the increasing inventory level required to achieve higher service levels. Recall that the service level is the probability of being able to satisfy demand from
stock, thus avoiding lost sales or backorders. In order to achieve a service level close to
100 percent, very large inventories are required. This happens because, assuming normally
distributed demand during the lead time, the safety stock must be very large to cover very
unlikely events as the service level approaches 100 percent.
Due to the highly nonlinear relationship between service level and inventory level,
it is crucial for management to estimate the costs of a reasonable service level. The
selection of an arbitrary service level (“Let’s just set it at 99 percent!”) may be very
costly since the difference of a few percentage points in service level could increase the
required inventory level substantially. For example, in Figure 14.10 an increase in the
service level from 95 to 99 percent requires a 32 percent increase in inventory. Thus,
although there is often pressure from marketing to set service levels very high, it is the
inventory manager’s job to make sure that the firm recognizes and fully accounts for the
cost of the chosen service level.
The selection of the service level (and thus the related inventory level) helps determine the number of inventory turnovers. Inventory turnover indicates the number of
times (during a year) the inventory in stock is completely renewed, that is, the relationship between the average inventory on hand and the annual usage of inventory. Inventory
Chapter 14
Independent Demand Inventory
1311] |]
FIGURE 14.10
Service level versus
inventory level.
(Q = 100; o = 100.)
i
wm
—}
\=)
‘So
—}—)
(percent)
Leveli)nn
Service
150
200
250
300
Average Inventory Level
turnover (or inventory turns) can be calculated for single items or for the overall stock
of inventory:
Inventory turnover = Annual cost of goods sold/Average inventory level
In practice, firms have inventory turnover anywhere between | and 50 turns per year. Specialty shops may have turnover as low as one to two turns per year. Summit Brewery, a
regional brewer, has 12 to 18 turns per year, and it is quite rare for a firm to have more than
50 turns per year. To assess whether inventory is being well managed, it is often best to
compare the inventory turnover of the firm against the inventory turnover of the best firms
in the same industry. If inventory turnover is low, it could be explained by either higher
service levels or different ordering and holding costs. Management should look beyond
inventory turnover to the service level policy or the cost structure inherent in the inventory
system. Management may accept a lower inventory turnover than the industry norm in
favor of a higher service level.
Alternatively, management might focus on reducing Q or o, thereby reducing the inventory
required for a given service level. The lot size Q can be reduced by reducing setup time or ordering costs. The standard deviation of demand during the lead time can be reduced by decreasing
the daily variation in demand or by decreasing the lead time. Daily variation in demand can
be reduced by working with customers to smooth out demand and reduce uncertainty in their
ordering patterns. Lead time can be reduced by decreasing throughput times in the production
and distribution process. Another possible cause of low inventory turnover is that the firm has
too much inventory of slow moving items that should be reduced and written off.
14.9
VENDOR
LO14.8 Define
vendor managed
inventory (VMI) and
the ABC system.
MANAGED
INVENTORY
Many firms have adopted vendor managed inventory (VMI). VMI is a supply chain
management initiative that passes the responsibility for managing inventory stocks to vendors (or suppliers). To make this work, a vendor under VMI is given access to the firm’s
demand forecast and inventory records. The vendor is contractually tasked with maintaining the correct inventory level at the firm’s location. VMI requires collaboration between
the supplier (vendor) and the customer (firm) in terms of sharing data as well as access to
the firm and its stocking locations. For example, VMI is used in grocery stores for some
food items, with the supplier stocking the shelves in the store, and is used for some supplier
deliveries to manufacturing plants. As the payoff for the effort expended to collaborate,
both supply chain partners can benefit from greatly reduced ordering costs and often a
|
312, Part Five
Inventory
Manufacturer P&G, renowned for its sophisticated supply
chain, uses vendor managed inventory (VMI) practices to
manage inventory levels for its retailers and distributors.
P&G makes leading consumer brands in many product
segments,
including
Crest, among
inventory
many
needs
Tide,
Bounce,
others. Used
Gillette,
Dawn,
and
primarily to fulfill the
of its largest retail customers,
P&G
is
expanding VMI to provide broader opportunities for collaboration with its customers around the world.
P&G’s experience shows that VMI improves
performance in a variety of ways. Retail stockouts of P&G products are less frequent, even while lowering inventory
levels by about one-third. On-time delivery is better, and
delivery truck fill rates improved at least 5 percent with
Roberto Machado No@a/Contributor/Getty Images
VMI, wasting less truck space and, thus, less fuel.
P&G helps its retailers quickly grasp VMI best prac-
software flags critical inventory levels, so that an analyst
tices, while improved technology interfaces allow them
can step in when needed.
to increase their collaboration with both suppliers and
customers. VMI is relatively “auto-pilot,” but the system
Source: consumergoods.com, 2020; us.pg.com, 2020.
higher level of customer service. Check the Operations Leader box to see how P&G uses
VMI to hold down its investment in inventory.
Cloud-based VMI platforms can allow vendors and their business customers to safely
connect and share inventory data. Information systems are increasingly able to communicate with one another, making VMI more easily implemented. Most VMI partnerships
result in reduced inventory in the supply chain, on average about 30 percent.
14.10
ABC CLASSIFICATION OF INVENTORY
Pareto’s law states that, for many events, roughly 80 percent of the effects come from
20 percent of the causes. It is also true that a few products in a firm account for most of
the sales. The law of the significant few can be applied to inventory management as well.
With inventories, a few items usually account for most of the inventory value as measured by dollar usage (demand x cost). Thus, one can manage these few items intensively
and control most of the inventory value. In inventory management, items usually are
divided into three classes: A, B, and C. Class A typically contains about 20 percent of the
items and 80 percent of the dollar usage. It therefore represents the most significant few. At
the other extreme, class C contains 50 percent of the items and only 5 percent of the dollar
usage. These items account for very little of the dollar value of inventory. In the middle is
class B, with 30 percent of the items and 15 percent of the dollar usage. The classification
of inventory in this way is called ABC analysis or the 80-20 rule.
Table 14.4 shows an example of an inventory with 10 items. For each item (row), we
have multiplied the annual usage in units by the unit cost to determine the dollar usage
of that item. We also have calculated the percentage of total dollar usage by comparing
each item’s dollar usage to the entire inventory usage ($254,725). We see that items 3 and
Chapter 14
TABU ie
Annual Usage of
—
Item
Items by Dollar Value
Annual Usage
in Units |
©
Unit Cost
Dollar Usage
$ sik 7,500
1
= 8,000! +.
$ 1.50Z
2
1,500
8,00
3
4
5
6
U
1
VAO,OOO'
~
©,
6,000
7,500
6,000
5,000
Independent
Demand Inventory
10:50).
2.00
SO
13.60
5
12,000 _
Coe
5
Percentage of Total
Dollar Usage
;
ATs
105,000.
41.2
12,000
3,750
81,600
9/550)
4.7
5
32.0
Liss)
8
4,500°
1.25
Db 6255
E22
7,000
3,000
12250)
2.00
17,500
6,000
6.9
2.4
$254,725
|
2.9%
che
¢)
10
Total
1313} |]
pte 100.0%
6 account for a great deal of the dollar usage (73.2 percent) and are classified as A items.
Items 1, 5, 7, 8, and 10 are low in dollar usage (10.5 percent) and are considered C items.
The other items are considered B items.
The designation of three classes is arbitrary; there could be any number of classes. Also,
the exact percentage of items in each class will vary from one set of inventory to another.
The important factors are the two extremes: a few items that are significant and a large
number of items that are relatively insignificant.
Most of the dollar usage in inventory (80 percent) can be controlled by closely monitoring the A items (20 percent). For these items, a tight control system should be used, including continuous review of stock levels, less safety stock, and frequent resupply. Looser
control might be used for C items. A periodic review system probably would be used with
longer ordering cycles and lower reorder and transportation costs. The B items require an
intermediate level of attention and management control.
With computerized systems, a uniform level of control sometimes is used for all items.
Nevertheless, managing inventories still requires setting priorities, and ABC analysis is
useful in doing this. A items usually deserve additional attention and effort.
14.11
KEY POINTS AND TERMS
This chapter provides an overview of inventory management and specific methods for the
management of independent demand inventories. The major points include the following:
¢
Inventory management is a key operations management responsibility. Inventory management affects capital requirements, costs, and customer service.
¢
Inventory is a stock of materials used to facilitate production or satisfy customer
demands. Inventories include raw materials, work in process, and finished goods.
¢
Inventories are held for many purposes, including cycle inventory, safety stock, anticipation inventory, and pipeline inventory.
¢
Inventory decisions should account for several costs. There are four inventory costs to
consider: item cost, ordering (or setup) cost, carrying (or holding) cost, and stockout cost.
¢
The economic order quantity (EOQ) is a simple but powerful calculation for estimating
the best order size while balancing ordering and holding costs. It includes assumptions
of a constant demand rate, constant lead time, fixed setup time, no stockouts, lot ordering, no discounts, and a single independent product.
¢
A continuous review (Q) system provides one way to handle random demand. When the
stock position drops to reorder point R, a fixed quantity Q is ordered. The time between
Dy}| 314] Part Five
Inventory
orders varies depending on actual demand. The value of Q is set equal to the EOQ. The
value of R is based on the service level desired.
A periodic review (P) system provides another way to handle random demand. The
stock position is reviewed at fixed intervals P, and an amount is ordered equal to target
inventory T minus the stock position. The amount ordered at each review period varies
depending on actual demand. The value of P is determined by using the EOQ, and the
value of T is based on the service level desired.
The choice between P and Q systems should be based on the timing of replenishment,
type of record keeping, and cost of the item. The P system should be used when inventory orders must be on a regular schedule.
High service levels require high investment levels for a given order quantity (Q) and
standard deviation (co). Management should analyze the investment required for a set of
service levels before setting the desired level. Looking at inventory turnover alone does
not provide an adequate basis for decisions on inventory levels.
VMI, vendor managed inventory, passes the responsibility for monitoring and replenishing inventory stocks from the buying firm to its vendors. Vendors must have access
to demand forecast and inventory levels. Collaboration between vendors and the buying
firm is crucial.
ABC analysis classifies inventory items into A, B, and C categories. ABC analysis is
based on the law of the significant few and the insignificant many. A items should be
closely managed. Less effort and cost should be expended on B and C items.
Key Terms
Bar coding 293
Point of sale data 293
Radio-frequency
identification 293.
™
Inventory 293
Safety stock 295
Cycle inventory 295
Anticipation inventory 295
Pipeline inventory 295
Item cost 296
Ordering (setup) cost 296
Carrying (holding) cost 296
Stockout cost 296
Independent demand 297
Dependent demand 297
Replenishment
philosophy 298
Requirements philosophy 298
Economic order quantity 298
Lead time 298
Total cost 300
Stock position 302
Continuous review system 302
Reorder point 302
Stockout probability 303
Service level 303
Periodic review system 306
Target level 306
Inventory turnover 310
Vendor managed
inventory 311
ABC analysis 312
ee
LEARNING
ENRICHMENT
Variety of Inventory Topics
www.effectiveinventory.com/articles
Website
(for self-study or
Inventory Management Software
www.cissltd.com
Website
Using Inventory Technology in Small Business
https://youtu.be/1d0O83MAMyAM
Video
5:03
ABC and Inventory Turnover
https://youtu.be/-19rIMekpfs
Video
ere
Inventory Technology at Walmart
https://youtu.be/KRJV 1SPYpIE
Video
6:20
instructor assignments)
eee
en
SSS
SSS
Sassi
SSS
sens
ll Sls sSsSssssssssseneGssunssscussnesss
Chapter 14
SOLVED PROBLEMS
Problem
Independent Demand Inventory
1315] |
:
1. EOQ_ Ina hardware warehouse, the independent demand for a commonly used bolt is
500 units per month. The ordering cost is $30 per order placed. The carrying cost is
25 percent per year, and each unit costs $.50.
a. According to the EOQ formula, what lot size should this product have?
b. How often should this product be purchased?
c. A quality team has found a way to reduce ordering costs to $5. How will that change
the lot size and the frequency of purchasing for this product?
Solution
First, convert demand to the same time units as the carrying cost. In this case, the carrying
cost is in years and demand is in months. The annual demand is 500 x 12 = 6000 units.
250)
a.
Q= (=
2 x 30 x 6000
ey 795 50
360,000
“wiVneies
= 1697.06 — 1697 units
b. The annual frequency of procurement = D/Q or 6000/1697 = 3.54 times per year. To
convert to months, divide 12 months per year by the annual frequency of procurement.
tesceee Every 3.39 months
3.54
re
3.54
Every 14.69 weeks
= = Every 103.11 days
2SD
_
(2x5 x 6000
se
A
___
/60, 000
1
Viont5
EG)
= 692.8 — 693 units
The frequency of procurement = DQ or 6000/693 = 8.66 times per year.
Problem
2. Q System
Part number XB-2001 is a spare part with annual independent demand
of 4000 units, a setup cost of $100, a carrying cost of 30 percent per year, and an item
cost of $266.67. The production facility 1s open 5 days per week and 50 weeks per year,
making a total of 250 productive days per year. The lead time for this product is nine
days, and the standard deviation of demand is two units per day. The firm wants to have
a 95 percent service level for this spare part.
H} | 316] Part Five
Inventory
Compute Q, using the EOQ formula.
. Compute R.
If the firm is using a Q system of inventory control (continuous review), interpret the
results of your computations.
Solution
_
[2x 100 x 4000
is
3 X 266.67
10,000
= 100
. Solving this part of the problem correctly requires two steps. First, the daily demand
must be calculated. This is done by dividing the annual rate of demand by the number of working days per year—4000/250 = 16 units per day. Thus, the average demand during lead time is 16 units per day for 9 days, or 9 X 16 = 144 units. Second,
the standard deviation of demand during the lead time must be calculated. This is
determined by taking the standard deviation of daily demand (two units) and multiplying by the square root of the number of days of lead time (the square root of 9).
R=m+2z
(9x 16)+ 1.65
x (2x V9)
1444+ 9.9
153.9 > 154 units
. Order 100 units‘when inventory (on hand plus on order) gets down to 154 units. On
the average, 9.9 units of safety stock will be on hand when the order arrives. In
5 percent of the cycles, there will be a stockout before the order arrives.
Problem
. P System
Consider the product described in solved problem 2 when answering the
following questions:
a.
How often should orders be placed for this product if they are placed at regular intervals using a periodic review system?
. Compute the target inventory level.
. State the specific decision rule for this product by using the information you have
calculated so far.
. Assume it is time for a periodic review. A check of the inventory level for this product reveals that there are 60 units on hand and 110 units on order. What should be
done?
Solution
iPS «(use Q and D from the previous problem )
100
~ 4000
II 025 year
= 1.25 working weeks (.025 year x 50 working weeks per year)
= 6.25 days (.025 year xX 250 woring days per year)
= 6 days (round off)
Chapter 14
p. L=m'
=m
Independent
Demand
Inventory
317] |
+s’
+zo'
= (average demand over P + L) + z(s.d of demand during P + L)
= 16x (6+ 9)+ 1.65
x (2x V9+4+6)
= 240 + 12.8
= 252.8 > 253
c. Review stock (on hand and on order) every six days and order up to a target level of
253 units.
d. Order up to the target level. The target level is 253 units. The amount of inventory on
hand and on order is 60 + 110 units, or 170 units total. The difference between the
target level and the inventory on hand and on order is the quantity which should be
ordered for delivery in nine days, 253 — (60 + 110) = 83. Order 83 units for delivery
in nine days.
Discussion Questions
il. Identify the different types of inventories (raw materials, work in process, and finished goods) carried in the
following organizations: gas station, hamburger stand,
clothing store, and machine shop. What functions
(purposes) do these inventories perform?
. Why are stockout costs difficult to determine? Suggest
an approach that might be used to estimate them.
. What is the difference between a requirements philosophy and a replenishment philosophy of inventory
management? Why is this difference important?
. Compare and contrast the management of finished
goods inventory in a manufacturing firm with that in a
retail or wholesale firm.
For a given service level, why does a P system require
a larger inventory investment than a Q system? What
factors affect the magnitude of the difference?
Suppose you are managing the Speedy Hardware Store.
Give examples of items that might be managed by a
P system and other items for which a Q system might
be used. How do these items differ?
How should a manager decide the appropriate service
level for finished goods items? Should some items have
a 100 percent service level?
What is the appropriate role of inventory turnover as a
measure for evaluating the management of inventory?
Under what circumstances is high turnover detrimental
to a firm?
. Suppose you are managing a chain of retail department
stores. As a top manager, how would you measure the
overall inventory management performance of each
store? How would you use this information in your
relationship with the individual store managers?
Problems
Five Excel
spreadsheets
are provided
on Connect
for
assistance in solving the chapter problems.
ile The Always Fresh Grocery Store carries a particular
brand of tea that has the following characteristics:
De The Grinell Machine Shop makes a line of metal
tables. Some of these tables are carried in finished
goods inventory. A particular table has the following
characteristics:
Sales = 8 cases per week
Sales = 300 per year
Ordering cost = $10 per order
Carrying charge = 20 percent per year
Setup cost = $1200 per setup (this includes machine
setup for all the different parts in the table)
Item cost = $80 per case
Carrying cost = 20 percent per year
a. How many cases should be ordered at a time?
b. How often will tea be ordered?
Item cost = $25
a. How many of these tables should be made in a
production lot?
c. What is the annual cost of ordering and carrying
tea?
b. How often will production be scheduled?
d. What factors might cause the firm to order a larger
or smaller amount than the EOQ?
c. What factors might cause the firm to schedule a lot
size different from the one you have computed?
7 | | 313]
Part Five
Inventory
3. The local Toyota dealer has to decide how many spare
shock absorbers of a particular type to order for repairing Toyota automobiles. This shock absorber has a
demand of four units per month and costs $25 each.
The carrying charge is 30 percent per year, and the
ordering cost is $15 per order.
a. What is the EOQ for this item?
b. How often will the dealer reorder this part?
c. What is the annual cost of ordering and carrying
this part?
4. What is the effect on EOQ and total cost of
eXcel the following types of ahr for the data in
problem 1?
a. A 40 percent increase in demand.
b. A 20 percent increase in carrying charge.
c. Use a spreadsheet to study the relationship between
lot size and carrying cost.
. The famous Widget Company sells widgets at the rate
of 80,000 units per year. Each widget sells for $100, and
it costs 30 percent to carry widgets in inventory for a
year. The process of widget production has been automated over the years, and it now costs $1000 to change
over the widget production line to other products that
are made on the same line.
a. What is the economical lot size for the production of
widgets?
b. How many lots will be produced each year?
c. What are the annual cost of carrying widgets and the
annual cost of changeover?
d. What factors or changes in assumptions might cause
the Widget Company to produce a larger lot than the
economic lot size calculated in part a?
. The Harvard Co-op orders sweatshirts with the
Harvard University emblem on them which costs them
$50 each. During a typical month, 900 sweatshirts are
sold (this includes all styles and sizes ordered from a
particular supplier). It costs $25 to place an order
(for multiple sizes and styles) and 25 percent to carry
sweatshirts in inventory for a year.
a. How many sweatshirts should the Co-op order at
one time?
b. The supplier would like to deliver sweatshirts once
a week. How much will this cost the Co-op per year?
Under what conditions would you agree to the
supplier’s proposal?
c. Suppose that sales increase to 1500 sweatshirts per
month but you decide to keep the lot size the same
as in part a. How much will this decision cost the
Co-op per year?
. The Co-op in problem 6 has discovered that it should
establish a safety stock for its sweatshirts. It wants
to use a reorder point system with a two-week lead
time. The demand over a two-week interval has an
average of 450 units and a standard deviation of
250 units.
a. What reorder point should the Co-op establish to
ensure a 95 percent service level?
b. What reorder point should be established to ensure
that no more than one stockout occurs in the course
of a year?
c. How much average inventory will the Co-op carry
for part b? Include both cycle inventory and safety
stock in your answer.
d. How often will the Co-op turn over its inventory,
using the results from part c?
eXcel
8. An electronics retailer carries a
particular
cell phone with the following characteristics:
Average monthly sales = 120 units
Ordering cost = $25 per order
Carrying cost = 35 percent per year
Item cost = $300 per unit
Lead time = 4 days
Standard deviation of daily demand = .2 unit
Working days per year = 250
a. Determine the EOQ.
b. Calculate the reorder point for a 92 percent service
level, assuming normally distributed demand.
c. Design a Q system for this item.
d. What happens to the reorder point when the lead
time changes? What happens to the reorder point
when the standard deviation of demand changes?
eXce | 9. For the data given in problem 8:
a. Design a P system for this phone with a
92 percent service level.
b. Compare the inventory investments required for the
P and Q systems (from problem 8) for a 92 percent
service level and other various values of service
level.
c. Why does the P system require a higher inventory
investment?
10. The Toyota dealer from problem 3 is considering
installing either a Q or a P system for inventory control.
The standard deviation of demand has been 4 units per
month, and the replenishment lead time is two months.
A 95 percent service level is desired.
a. If a continuous review system is used, what is the
value of Q and R that should be used?
b. Ifa periodic review system is used, what is the value
of P and T that would be applicable?
c. What are the pros and cons of using the P system
compared with using the Q system for this part?
Chapter 14
11. The Suregrip Tire Company carries a certain type of
tire with the following characteristics:
Average annual sales = 600 tires
Ordering cost = $40 per order
4
Carrying cost = 25 percent per year
Item cost = $50
P
per tire
1319] |)
Assume that the items are to be ordered together
from the same supplier at an ordering cost of
$20 per order and an annual carrying cost of
20 percent. Also assume 300 working days in a
year.
,
:
:
:
a. If aP system is used, what is the optimal ordering
:
:
interval in days?
Lead time = 4
iat
days
Standard deviation of daily demand = | tire
b. How much of each type of fabric would be ordered
when a combined order is placed?
a. Calculate the EOQ.
b. For a Q system of inventory control, calculate the
safety stock required for service levels of 85, 90, 95,
c. What is the effect on the ordering interval of
changing the carrying cost to 25, 30, and
35 percent?
97, and 99 percent.
c. Construct a plot of total inventory investment versus
d. Can these fabrics be ordered by using a Q system?
Explain.
service level.
e. Classify the four items above as A, B, or C inventory
d. What service level would you establish on the basis
of the graph in part c? Discuss.
12. For the data in problem 11:
a. Calculate the annual turnover as a function of service level.
eXce/
130 The :Gover-up
Drapery
Gonipany
ee the following
eee
four We types a
of fabric with
Carri
characteristics:
Type
items.
14. Suppose you are the supplier of the Cover-up Drapery
Company described in problem 13. It costs $2000 each
time you change over your fabric-producing machine
from one type to another (1, 2, 3, or 4). Assume that
b. If sales were to increase by 50 percent, what would
happen to the turnover at a 95 percent service level?
1
2
3
4
=Independent Demand Inventory
Annual Demand
300
250
100
200
(yards)
Item Cost per Yard
$20
$18
$12
$ 8
YORS carrying cost is 30 percent and the other data are
as given in problem 13.
a. What lot sizes would the supplier of fabric prefer to
make for items 1,DECK
2, 3, and 4?
b. How would you reconcile the lot sizes that the
supplier would like to produce and those that the
Cover-up Drapery Company would like to buy?
Describe several ways in which these two differing
lot sizes can be reconciled.
VEG
Part Five
Inventory
Supplement
Advanced Models
S-LO14.9 Solve
advanced inventory
problems.
This supplement presents two additional models that are useful for managing independent
demand inventory. The first model applies to outside procurement in which price discounts
are given; the second applies to a gradual fill of inventory when the lot arrives uniformly
over time rather than all at once.
PRICE BREAKS
Outside suppliers often offer price discounts for large purchases. These discounts may
be given at different procurement levels, and they may apply either to the whole order or
to only the increment purchased. In this supplement, we assume that the price discounts
apply to the entire order. For example, the procurement price may be $2 per unit for 1 to
99 units and $1.50 per unit for 100 units and up. The cost of the units thus exhibits a jump
or discontinuity at 100 units. For 99 units, the cost of the procurement order is $198, and
for 100 units the cost is $150.
To solve for the EOQ, the procedure is to first calculate the EOQ for each different
procurement price. Some of these EOQs may not be feasible because the EOQ falls outside
the range of the price used to compute it. The infeasible EOQs are eliminated from further
consideration. The total procurement and inventory operating cost for each feasible EOQ
and each price-break quantity is then computed. The feasible EOQ or price break that
results in the lowest total cost then is selected as the order quantity.
Consider the following example:
D = 1000 units per year
i = 20 percent per year
S = $10)per order
C, = $5 per unit for 1 to 199 units
C, = $4.50 per unit for 200 to 499 units
C, = $4.25 per unit for 500 units or more
First, we calculate the three EOQs corresponding to the three values of C,. We obtain
Q, = 141, Q, = 149, and Q, = 153. In this case, Q, and Q, are infeasible, and they are
eliminated from further consideration. We then compute the total cost of procurement
and inventory at the remaining EOQ and at the two price breaks. These total costs are as
follows:*
D
ING = (2) + ic(§) + CD
a)
2
TC(141) = 1019041) + .2(5) (1414) + 5(1000) = $5141
TC(200) = 10(190%/99) + .2(4.5) (2004) + 4.5(1000) = $4640
TC(S00) = 10(1990/%00) + .2(4.25) (5004) + 4.25(1000) = $4482
Since TC(500) is the lowest annual cost,-500 units should be ordered.
“Note that the annual cost CD of buying the units is included in the total cost equation since this cost
is
affected by the discount.
Chapter 14
Independent Demand Inventory
321] 1
FIGURE S14.1
Inventory cost with
price breaks.
200
500
The cost behavior for the example is shown in Figure $14.1. Notice that at each price
break, the total cost is reduced. Therefore, in this example, the quantity at the highest price
break is selected.
It is not always necessary to calculate all the EOQs and the cost at each price break.
A more efficient procedure follows:
1. Calculate the EOQ for the lowest cost per unit (the largest price-break quantity). If this
EOQ is feasible, that is, above the price break, then this is the most economic
order
quantity.
2. If the EOQ is not feasible, use the next lowest price and continue calculating EOQs
until a feasible EOQ is found or until all prices have been used.
3. Next, calculate the total cost of the EOQ and the total cost at all the higher price breaks.
4. The minimum of these total costs indicates the most economic order quantity.
In the above example, this procedure yields the same result as in the calculations earlier.
By coincidence, both methods require the same number of calculations for this particular
example.
UNIFORM LOT DELIVERY
In some cases, the entire lot is not placed in inventory at one time but is delivered gradually. An example is a manufacturer that builds inventory at a constant production rate.
Another example is a retailer that accepts delivery of an order in several shipments over a
period of time.
The effect of this delivery condition on inventory is shown in Figure $14.2. The inventory level builds up gradually as both production and consumption occur. Then the inventory level is depleted as only consumption takes place.
The effect of gradual delivery is to reduce the maximum and average inventory level
over that obtained in the simple EOQ case when the entire lot is accepted at one time. Suppose units are produced at a rate of p units per year and consumed at a rate of D units per
year (where p > D). Then the average inventory level is
27)
This formula can be derived with the use of geometry by reference to Figure $14.2.
i] | 322,
Part Five
Inventory
FIGURE $14.2
Uniform lot delivery.
(1
D/p)Q
Level
Inventory
—
~~
Q/D ——
Time
The above expression for average inventory is used in place of Q/2 in Equation (14.1).
When the resulting expression for TC is minimized, the following EOQ formula is obtained:
2SD
2=Wcq— Dip)
The EOQ in this case is always somewhat larger than the ordinary EOQ because the factor (1 — D/p) is less than 1. As p approaches D, the EOQ becomes very large, which means
that production is continuous. When p is very large, the above EOQ formula approaches
the ordinary EOQ. In deriving the ordinary EOQ, we assumed that the entire lot arrived in
inventory at once, which is equivalent to an infinite production rate p.
Supplement Problems
eXce | 1. Suppose that for problem | in the chapter,
the Always Fresh Grocery Store is offered
a discount by its supplier if more than SO cases of
tea are ordered at one time. The unit item costs are
$80 per case for 0 to 49 cases and $76 per case for
50 cases or more. This price of $76 applies to the
entire order.
a. Should the grocery store take the discount offer?
b. What discount is required for the store to be indifferent between taking the discount and ordering the
EOQ?
. A supplier has come to you and offered the following
deal. If you buy 29 or fewer cases of cleaning solution,
the cost will be $25 for each case. If you buy 30 or
more cases, the cost will be $20 per case. Assume your
cost of carrying inventory is 15 percent a year, it costs
would negotiate for and why those quantities and
prices are selected.
3 . For problem 2 in the chapter, suppose the Grinell
Machine Shop produces its tables at a rate of two per
day (250 working days per year).
a. What is the optimal lot size?
bisDiaii 2 Sapo
Fon handanventoryoyerm stim,
SY HE SS eee eae
te OAC
. A producer of electronic parts wants to take account
of both production rate and demand rate in deciding on
its lot sizes. A particular $50 part can be produced at
arate of 1000 units per month, and the demand rate is
200 units per month. The firm uses a carrying charge
of 24 percent a year, and the setup cost is $200 each
time the part is produced.
$20 for you to order the material, and you use 50 cases
a. What lot size should be produced?
per year.
b. If the production rate is ignored, what would the lot
a. How many cases should you order?
b. Would you negotiate with this supplier for a further
discount? Explain the quantities and prices that you
size be? How much does this smaller lot size cost
the firm on an annual basis?
c. Draw a graph of on-hand inventory versus time.
Car
Aw Pais
Materials Requirements
Planning and ERP
BUEABABRABRHREAEBRAR
EE B&B
Contrast and compare MRP vs. order-point systems.
Describe in detail each element of an MRP system.
Explain the five requirements for a successful MRP system.
Wia
Describe what an ERP sy: tem does.
A |
Hit
i
The typical manufacturing firm has thousands of products and parts to manage, constantly
shifting priorities, and unpredictable demand. It is possible to manage this complex situation through the use of a computerized planning and control system called materials
requirements planning (MRP).
Consider a company that is making severai different types of skateboards. Each skateboard requires a deck (top), four wheels, and two axle units to attach the wheels. The com-
pany assembles the skateboards from parts that are purchased. For each skateboard type,
the company needs to plan ahead to order enough decks, wheels, and axle units to arrive in
time for final assembly. Assuming there is already some inventory of these parts on hand
or scheduled to arrive and the lead times for ordering new parts are known, how does the
company plan the orders and assembly so all the parts arrive just when needed to make
°
a
>
o
o
°o
°
o
°
o
sia
7
°
ray
323|
i
Ta
H] | 324| Part Five
Inventory
a batch of, say 100, new finished skateboards three weeks from now? This planning and
scheduling can be done by an MRP system.
Service organizations also can benefit from the use of MRP systems. Services such
as restaurants, hospitals, and electric-power companies need facilitating goods to support
their service delivery systems. Delivery of these goods can also be scheduled by an MRP
system to arrive just when they are needed to support the scheduled service offerings.
An MRP system is driven by the master schedule, which specifies the production timing of
final products (often referred to as end items) or output of the production function. All future
demands for purchased parts or shop orders to make the parts are dependent on the master
schedule and derived mathematically by the MRP system from the master schedule. Remember
it takes exactly one deck, four wheels, and two axle units to make a skateboard. Given the num-
ber of skateboards in the master schedule, the number of decks, wheels, and axle units needed
to meet the master schedule is then known. Since conditions are usually changing, the master
schedule is a far better basis than past demand for planning future material requirements.
MRP software “explodes” the master schedule into purchase orders and shop orders for
scheduling the factory. For example, if the product in the master schedule is a hand-held calculator, the process of parts explosion will determine all the parts and components needed
to make a specified number of calculator units. This process of parts explosion requires a
detailed bill of materials that lists each of the parts needed to manufacture any particular
end item in the master schedule. The required parts may include assemblies, subassemblies,
manufactured parts, and purchased parts. Parts explosion thus results in a complete list of
the parts that must be ordered and the shop schedule that is required for internal production.
In the process of parts explosion, it is necessary to consider inventories of parts that are
already on hand or on order. For example, an order for 100 end items may require a new
order of only 20 units of a particular part because 50 units are already in stock and 30 units
are on order.
Another adjustment made during parts explosion is for production and purchasing lead
times. Starting from the master schedule, each manufactured or purchased part is offset
(i.e., ordered earlier) by the amount of time it takes to get the part (the lead time). This
procedure ensures that each component will be available in time to support the master
schedule. If sufficient manufacturing and supplier capacity is available to meet the orders
resulting from parts explosion, the MRP system will produce a valid plan for procurement
and manufacturing actions. If sufficient capacity is not available, it will be necessary to
replan the master schedule or change the capacity.
15.1
LO15.1
THE MRP SYSTEM
Define the
elements, inputs, and
outputs of an MRP
system.
A typical MRP system, along with inputs and outputs, is illustrated in Figure 15.1. The
MRP system begins with the master schedule, which is determined by customers’ orders,
aggregate production planning, and forecasts of future demand. The parts-explosion process, at the center of the system, is driven by three types of information: master schedule,
bill of materials, and inventory records. The result of the parts-explosion process is two
types of output: purchase orders that go to suppliers and shop orders that go to the factory.
Before shop orders are sent to the factory, however, materials planners check on whether
sufficient capacity is available to produce the parts required. If capacity is available, the
shop orders are placed under control of the shop-floor control system. If capacity is not
available, a change must be made by the planners in the capacity or in the master schedule
through the feedback loop shown. Once the shop orders are under the shop-floor control
system, the progress of these orders is managed through the shop to make sure
that they
are completed on time.
Chapter 15
FIGURE 15.1
Materials Requirements Planning and ERP
325 ||
A closed-loop MRP system.
S&OP
(aggregate
- production —
plan)
/ Firm orders:
©
A
from customers
_ Forecast
TOF
|
_ or from finished-
demand —
goods inventory
_ Master
x
Bi
ey
go
gn
changes
Pi es
Teeny
materials
schedule
Parts
_ explosion
:
BRCCSERERRERRRERTRRRARRERERRRREERE
REP
eee eee ee
nf Inventory ui
records
Capacity
Suppliers
PRODUCTION
CONTROL. This
production control
employee uses a
computer to track the
flow of materials as
part of an MRP system.
Cultura Exclusive/Getty
Images
Raw
aterial:
iy ties a
planning
Operations
:
BER
RRR
Inventory |
transactions
ERR R BPE e eee
=P Product
eee
pEUUEUCUORUCUCROC
TET
OOO
TT
P| | 326, Part Five
Inventory
ea
|
OPERATIONS LEADER
|
RR
ne pa
ha i Wil
mn an osu
|
|
a
3M Canada Uses JDA Software
3M
Canada
is one
of the first subsidiaries
Headquartered
in London,
established
also could consider capacity and frequently reschedule
Ontario,
its factories.
Using JDA Company (formerly i2) software, 3M integrated collaborative forecasting, material planning, and
by 3M
in 1951.
3M
Canada
now has sales offices in the provinces of Alberta,
British Columbia, Ontario, and Quebec, as well as five man-
ufacturing facilities—four in Ontario and one in Manitoba.
shop-floor execution with complete visibility across the
supply chain. 3M implemented two applications, JDA
Factory Planner and JDA Supply Chain Planner, across
all factories in Canada. 3M could replan every day on
the basis of data from the previous shifts and new customer orders. Supply Chain Planner created optimized
purchase requirements and manufacturing orders. The
previous system could plan only on the basis of material
availability, but Factory Planner also considered machine
capacity and other resources. As a result, 3M Canada
Brigham/123RF
Paul
3M Canada used to operate a legacy MRP system that
considered
only material
availability.
By incorporating
finite capacity scheduling and constraint management, it
was able to grow sales through better on-time delivery,
reduce inventories by 23 percent, and improve cash flow.
Source: www.3m.com,
2020.
Figure 15.1 represents MRP as an information system used to plan and control inventories
and capacity. Information is processed through the various parts of the system to support management decisions. If the information is accurate and timely, management can use the system
to control inventories, deliver customer orders on time, and control costs for the firm. In this
way, the materials are managed continually in a dynamic and changing environment.
15.2
LO15.2
MRP VERSUS ORDER-POINT SYSTEMS
Contrast
and compare MRP
vs. Order-point
systems.
TABLE 15.1
Comparison of MRP.
and Order-Point
Systems
An MRP system calls into question many of the traditional concepts used to manage inventories. Order-point systems do not work well for the management of inventories subject to
dependent demand. Dependent demand is determined by demand for another item and
not by the market. When a product is manufactured, all demand for components, assemblies and parts is dependent on demand for the final product in the master schedule.
Some of the key distinctions between MRP and order-point systems are summarized in
Table 15.1. One distinction is the requirements philosophy used in MRP systems versus a
replenishment philosophy used in order-point systems. A replenishment philosophy indicates
that material should be replenished when it runs low. An MRP system does not do this. More
Demand
Order philosophy
Forecast
Control concept
Objectives
Lot sizing
Demand pattern
Types of inventory
MRP
Order Point
Dependent
Requirements
Based on master schedule
Control all items
Meet manufacturing needs
Discrete
Lumpy but predictable
Work-in-process and raw materials
Independent
Replenishment
Based on past demand
ABC
Meet customer needs
EOQ
Random
Finished goods and spare parts
Chapter 15
Materials Requirements Planning and ERP
327] |]
material is ordered only when it is needed to meet production specified on the master schedule. If there are no
future manufacturing requirements for a particular part,
it will not be replenished even though the inventory level
is low. This requirements concept is particularly important in manufacturing because demand for component
parts is “lumpy.” When a lot is scheduled, the component
parts are needed for that lot, but demand is then zero until
another lot is scheduled. If order-point systems are used
for this type of lumpy demand pattern, material will be
carried on hand during long periods of zero demand.
Another distinction between the two systems is
the use of forecasting. For order-point systems, future
demand is forecast on the basis of the past history of
demand. These forecasts are used to replenish the stock
levels to serve customers. In MRP systems, past demand
for component parts is irrelevant for forecasting future
needs. Future needs are derived mathematically from
These auto parts are subject to dependent demand.
Barry Willis/Getty Images
the master production schedule.
The ABC classification of inventory also does not
work well for MRP systems. In manufacturing a product,
components that are C items are just as important as A items. For example, an automobile
cannot be shipped if it lacks a fuel line or radiator cap even though these items are relatively
inexpensive C items. A meal cannot be served by a restaurant if a key ingredient is missing.
Therefore, it is necessary to control all parts, even the C items.
The traditional EOQ is not useful in MRP systems, although modified lot-sizing formulas are available. The assumptions used to derive the traditional EOQ are badly violated
by the lumpy demand patterns for component parts. Lot sizing in MRP systems should be
based on discrete requirements. For example, suppose that the demand for a particular part
by week over the next six weeks is 0, 30, 10, 0, 0, and 15. Further assume that the EOQ is
calculated to be 25 parts. With the EOQ or multiples of the EOQ, we could not match the
requirements exactly and therefore would end up with remnants in inventory. These remnants from the EOQ cause unnecessary inventory carrying costs. It would be far better to
base lot sizes on the discrete demand observed. For example, with a lot-for-lot policy, we
could order 30 units for the second week, 10 for the third week, and 15 for the sixth week,
resulting in three orders and no carrying costs. We could also order 40 units for the second
and third weeks combined, thereby saving one order but incurring a small carrying cost.
With MRP systems, various discrete lot sizes need to be examined.
The objective in managing independent demand inventories with reorder-point rules is
to provide a high customer service level at low inventory operating costs. This objective
is oriented toward the customer. In contrast, the objective in managing dependent demand
inventories with MRP is to support the master production schedule. This objective is manufacturing oriented; it focuses inward rather than outward.
15.3
LO15.3
PARTS EXPLOSION: HOW AN MRP SYSTEM WORKS
Construct
a materials plan
given the gross
requirements.
The easiest way to understand MRP is to focus on the parts-explosion process itself. Suppose that tables of the type shown in Figure 15.2 are being manufactured. The finished
table consists of a top and a leg assembly. The leg assembly in turn consists of four legs,
two short rails, and two long rails. In this particular example, leg assemblies are built in
i
|
328] Part Five
Inventory
FIGURE 15.2
Table example.
Top
Leg
assembly
Short rail
Long rail
advance and stored in inventory. This procedure permits the table to be produced faster, as
orders are received, than it could be if the table were completely assembled from individ-
ual parts. It is common practice in manufacturing firms to build assemblies for inventory to
reduce total production lead time and save setup costs.
The bill of materials (BOM) for this table is shown schematically in Figure 15.3. The
finished table is at the first level of the bill, and is called the end item or finished product. The
leg assembly and tabletop are at the second level, since these parts are assembled together
to produce a finished table. The pieces that go into the leg assembly are all listed at the third
level. We are assuming that the parts for this table are purchased from outside; otherwise
there would be a fourth level in the BOM for the wood used to make the legs, rails, and top.
Another piece of information needed before parts explosion is the planned lead times
for manufactured and purchased parts, as shown in Table 15.2. For planning purposes, it
takes one week to assemble the finished table from the leg assembly and top. This planned
lead time includes average waiting time due to interference from other jobs, which is usually much longer than the actual working time. Similarly, Table 15.2 shows that two weeks
are planned for purchase of a tabletop from the time the order is placed until the tabletop
is in the factory. One week of lead time is required to purchase the table legs and rails, and
one week needed for leg assembly.
It is now possible, using parts explosion, to construct a materials plan for the finished
tables and all parts. The resulting materials plan is shown in Table 15.3. Now, we will walk
through the line-by-line calculations for finished tables shown in the first panel at the top
of Table 15.3.
FIGURE 15.3
Bill of materials
(quantity per unit
shown in parentheses).
Leg assembly
(1)
Short rails
Long rails
(2)
(2)
Chapter 15
TABLE 15.2
2
:
Planned Lead Times
Materials Requirements Planning and ERP
|329} |]
4
occ
Assemble table*
_ Finished leg assembly’ |
Purchase legs
Purchase short rails
Purchase long rails
Purchase top
Ss
fee
Ry)
ese
a
*Assume the tabletop and complete leg assembly are available.
+ Assume the legs, short rails, and long rails are available.
Line 1: Gross requirement is the demand for finished tables: 200 in week 4, 150 in
week 5, and 100 in week 6.
Line 2: The scheduled receipts for tables are zero for all weeks in this case. Scheduled
receipts consist of tables currently being made and expected to be completed from past
shop orders.
Line 3: There are currently 50 tables on hand. Projected Ending Inventory is 50 for
each of the first three weeks, since we are not planning to use any inventory during
these weeks.
Line 4: Net requirement is just the gross requirement minus the projected ending
inventory at the end of the previous week. In this case in week 4 the net requirement 1s
200 — 50 = 150. We use the on-hand inventory from week 3 to meet some of the gross
requirement in week 4; the rest of the gross requirement is now listed as a net requirement. Since there is no on-hand inventory carried over for week 5 or 6, the net requirement equals the gross requirement in those two weeks.
Line 5: Planned order receipt for 150 tables is entered in week 4 and is the same as the
net requirement in that week since we are assuming lot-for-lot (L4L) production. L4L
means every net requirement is met by a planned order receipt of the same amount in
that period. Likewise, the planned order receipts for week 5 are 150 and for week 6 are
100, the same as the net requirements in those weeks.
Line 6: Planned order releases are offset by one period for the table lead time (from
Table 15.2). This means we will release the order so it arrives one week later as a
planned order receipt. In this case, orders are planned to be released using the one
week offset for 150 tables in week 3, 150 in week 4, and 100 in week 5.
The weeks in the materials plan are such that the current week (in other words, today) is
always within week |. This means that after the current week has passed, what is currently
listed as week 2 will become week 1, and a new planning week is added at the far end
of the schedule. Note, all quantities are assumed to occur at the beginning of each week,
except projected ending inventory that occurs at the end of the week.
Now, step back and look at the first panel in Table 15.3. You can see that the 200 tables
required in week 4 are translated into a net requirement of 150 tables, since there are 50
finished tables in inventory. Using L4L logic, the planned order receipts in week 4 are also
150 tables. Using the one-week offset in lead time we get a planned order release of 150
tables in week 3. In a similar way, the gross requirements for weeks 5 and 6 are netted for
zero projected inventory and then entered as planned order receipts and offset by a oneweek lead time to ultimately get planned order releases of 150 in week 4 and 100 in week 5.
This is the logic of MRP.
Next, the planned order releases for tables are used to calculate gross requirements for
tops and leg assemblies at the next level down in the BOM. The planned order releases for
F} | 330]
Part Five
Inventory
TABLE 15.3
Materials Plan—
Parts Explosion
Week
1
On hand = 50
2
3
Tables
Gross requirement
LT = 1 wk
Scheduled receipts
Lot size: L4L
Projected ending inventory
Safety stock = 0
Net requirement
50
50
Planned order receipts
Planned order releases
Tops
4
=
6
200
150
100
50
1502
< 150.7
100
He ae eis 150
150
150
150
100
100
150
15OR
OO
On hand = 50
Gross requirement
LT = 2 wk
Lot size: L4L
Safety stock = 0
Scheduled receipts
Projected ending inventory
Net requirement
50
50
150
100
Planned order receipts
(Sit 3:/aus Sea
50
150
100
50
100
150
100
150
150
100
100
100
On hand = 100
LT = 1 wk
Lot size: L4L
Safety stock = 0
On hand = 150
LT = 1 wk
Lot size: L4L
Safety stock = O
On hand = 50
LT = 1 wk
Lot size: L4L
Safety stock = 0
On hand = O
LT = 1 wk
Lot size: L4L
Safety stock = O
Planned order releases
Leg assembly
Gross requirement
Scheduled receipts
Projected ending inventory
50
100
150
150
100
Net requirement
Planned order receipts
Planned order releases
Leg
Gross requirement
Scheduled receipts
Projected ending inventory
400
150
Net requirement
Planned order receipts
Planned order releases
Short rail
Gross requirement
‘Scheduled receipts
Projected ending inventory
Net requirement
Planned order receipts
Planned order releases
Long rail
Gross requirement
Scheduled receipts
Projected ending inventory
Net requirement
400
400
200
50
200
200
reap
50
200
100
300
200
Planned order receipts
eee ea 100
300
200
Planned order releases
100
200
300
tables (level 1) are transferred to gross requirements for tops and leg assemblies (both at
level 2) on a one-for-one basis, since it takes one tabletop and one leg assembly to make
a table (see arrows in Table 15.3). The projected ending inventory and scheduled receipts
for table tops and leg assemblies are subtracted from the gross requirements to arrive at
net requirements. The net requirements are then entered as the same planned order receipts
using L4L logic. The planned order receipts are offset by lead time to arrive at planned
order releases for tabletops and leg assemblies. Notice that in Table 15.3, 50 tabletops are
currently on hand and 50 are scheduled to arrive in week 2. These arrivals increase the
projected ending inventory in week 2 to 100 units total. Since we have 100 tops available
in inventory, we only have a net requirement of 50 more tops to meet the gross requirement
of 150 tops in week 3. Using L4L logic and a two-week offset (for the lead time) leads to a
planned order release of 50 tops in week 1.
Chapter
15
Materials Requirements Planning and ERP
334] |] |
Once we have completed the calculations in Table 15.3 for tops and leg assemblies, we
are ready to finish the material plan. The planned order releases for leg assemblies are used
to compute gross requirements for legs, short rails, and long rails. The planned orders for
leg assemblies are multiplied by 4 for legs, by 2 for short rails, and by 2 for long rails to
arrive at gross requirements. A gross requirement at any level is the amount of material
required to support planned order releases at the next-highest level. The gross-to-net calculation and the offset for lead time are then performed for each of the three remaining parts
to arrive at planned order releases. This completes the parts explosion.
Table 15.3 has been constructed from the master schedule down, one level at a time,
through the BOM. The materials plan for each level in the BOM was completed before
moving down to the next level. For each part, the gross requirements have been reduced by
projected ending inventory and scheduled receipts to arrive at net requirements. The net
requirements have been entered as planned order receipts and then offset (planned earlier)
by the lead time to arrive at planned order releases. By the process of netting and offsetting, the master schedule is converted to planned order releases for each part required.
What does the materials plan in Table 15.3 tell us? While most of the plan helps with
record keeping, the planned order releases will require action. Currently, only orders in
week | must be dealt with. First, we should immediately release purchase orders to our suppliers for 50 tops, 50 short rails, and 100 long rails since these are the planned order releases
at the beginning of week 1. The materials plan also gives us the planned order releases for
each week in the future. If the master schedule and all other conditions remain constant,
the planned orders will be released when the time comes. For example, in week 2 we plan
to release an order to the shop to complete 50 leg assemblies. If the materials arrive as planned,
we will have on hand the legs and rails needed for this shop order: 200 legs, 100 short rails,
and 100 long rails. In addition to the shop order for 50 leg assemblies, we plan to release, in
week 2, purchase orders for 150 tops, 550 legs, 300 short rails, and 300 long rails. These materials will be needed to support future leg-assembly and table-assembly shop orders.
This example illustrates the construction of a time-phased materials plan. All purchase
orders and shop orders are interrelated to provide materials when needed. If the actual
lead times can be managed to meet the planned lead times, there will be no unnecessary
inventory accumulations or wasted time waiting for materials in the shop, and the orders
for delivery of finished tables will be shipped on time.
As a matter of fact, once any initial inventories are depleted, no finished goods inventories will be planned by the MRP system using the L4L approach. This is the case because we
have planned production to just equal final demand, after adjusting for available inventories.
Likewise, no purchased-parts inventories are planned after initial inventories and scheduled receipts are depleted. Unless safety stocks are added for uncertainties or economic lot
sizes are used to smooth out production levels, no inventory will be planned except work
in process required for assembly or fabrication. The logic embedded into the MRP system
assumes parts are available exactly when they are needed to support the production plan.
Lot sizing is very important in MRP to achieve economical production lots and purchasing orders. As a result, a fixed or calculated lot size may be used for each level in the material plan rather than an L4L approach. For illustration purposes, Table 15.4 shows a fixed
lot size for the finished tables and table tops. We assume the economic lot size for finished
tables is 200 tables and the economic lot size for ordering tops is 300 tops. We also assume
that a safety stock of 50 units is planned for the table tops, but it won’t be used until
needed. This safety stock provides protection for late deliveries by the table top supplier.
The revised material plan is shown in Table 15.4. In period 4 we have a gross requirement of 200 tables and a net requirement of 150 tables, as before. However, the lot size is
200 tables, so the planned order receipts to assemble tables will be 200, leaving a projected
|
332)
Part Five
Inventory
TABLE 15.4
Material Plan with
Week
1
Lot Sizing
On hand = 50
LT = 1 wk
Lot size = 200
Safety stock = 0
2
4
5
6
200
150
100
50
50
150
200
100
100
200
200
200
200
200
150
150
300
250
50
300
Tables
Gross requirement
Scheduled receipts
Projected ending inventory
Net requirement
Planned order receipts
50
50
Planned order releases
On hand = 50
LT = 2 wk
Lot size = 300
Safety stock = 50
3
Tops
Gross requirement
Scheduled receipts
Projected ending inventory
Net requirement
Planned order receipts
Planned order releases
50
50
300
250
250
300
ending inventory of 50 tables at the end of period 4. In period 5 the gross requirement is
for 150 tables, leaving a net requirement of 100 tables after using the 50 tables in inventory. Using the lot size of 200 we must assemble 200 tables, which leaves a projected ending inventory of 100 tables at the end of period 5. This inventory of 100 tables just meets
the gross requirement for period 6, so there are no additional planned order receipts. The
planned order receipts are now offset by a one-period lead time to yield planned order
releases of 200 in period 3 and 200 in period 4. These planned order releases now become
gross requirements for the tops (see arrow in Table 15.4).
This changes the gross requirements for tops from the prior L4L calculations. Proceeding ahead, we subtract the safety stock of 50 tops from the on-hand inventory, to give a
projected ending inventory of zero tops in period |. This simply takes safety stock out of
the calculations going forward. Then 50 tops are scheduled receipts in period 2 leaving a
projected ending inventory of 50 at the end of period 2. We then complete the calculations
for purchasing tops in Table 15.4 by netting and offsetting. Notice, by ordering the fixed
quantity of 300 units at a time, we end up with planned order releases of 300 tables in periods | and 2 instead of the smaller planned orders we had with L4L lot sizing.
15.4
LO15.4
MRP SYSTEM ELEMENTS
Describe in
detail each element
of an MRP system.
Master
Scheduling
Although parts explosion is the heart of the MRP system, it takes a good deal more to
make an MRP system work. An MRP system needs several elements to be successful. We
describe these elements more formally next.
Master scheduling drives the entire materials planning process. The master schedule has been
described as “top management's handle on the business.” By controlling the master schedule,
top management can control customer service, inventory levels, and manufacturing costs. Top
managers cannot perform the master scheduling task by themselves because there are too many
details; therefore, they often delegate the task to a cross-functional team. However, top managers can set master scheduling policy, thereby controlling the materials planning function.
Top management’s primary interface with manufacturing is through the aggregate production plan (or S&OP) shown at the top of Figure 15.1. The aggregate production plan deals
with families of products or product lines, not specific products, models, or options that are
Chapter 15
:
X
Cloud computing facilitates MRP
systems.
ImageFlow/Shutterstock
Materials Requirements Planning and ERP
/333 ||]
in the master schedule. For example, if a manufacturer such as John Deere makes
a variety of tractors, the aggregate production plan might contain various types of
tractors but not the particular size of engine, hydraulics options, or other features
that the customer can select. Thus, the master scheduling process accounts for the
overall aggregate production plan, which has already been established, or seeks to
modify this plan if it is discovered, for example, that capacity is not available.
The master schedule might extend into the future for a year or more. It must
extend at least beyond the longest cumulative production lead time to ensure that
sufficient time is available to order all parts and make the finished product. Generally speaking, the master schedule should be frozen inside the cumulative production lead time to prevent unnecessary scrap and expediting due to changes
during the production cycle.
/
Rarely is the master schedule a reflection of future demand forecasts. Rather,
the master schedule is a forecast of what will be produced. It is a “build” schedule. Finished goods inventory is a buffer between the master schedule and final customer
demand, smoothing out workloads and providing fast customer service.
Bill of Materials
(BOM)
The BOM is a structured list of all the materials or parts needed to produce a particu-
lar finished product, assembly, subassembly, manufactured part, or purchased part. The
BOM serves the same function as a recipe used for cooking: It lists all the ingredients. It
would be foolish to allow errors to creep into your favorite cooking recipes. The same is
true for a BOM. If there are errors in the BOM, the proper materials will not be ordered
and the product cannot be assembled and shipped on time. As a result, the other parts
that are available will wait in inventory while the missing parts are expedited. Management must therefore insist that all BOMs are 100 percent accurate. Experience has
shown that it is not too costly to have 100 percent accuracy; rather, it is too costly to
tolerate imperfect BOMs.
Some firms have several BOMs for the same product. Engineering has one BOM, manufacturing has a different version, and cost accounting has still another. An MRP system
requires a single BOM for each product to be used by all individuals in the firm. The
BOM in the MRP system must be the correct one, and it must represent how the product is
manufactured. In firms where the BOM has been used as a reference document and not a
materials-planning tool, this concept of a single bill is very difficult to implement.
BOMs are constantly undergoing change as products are redesigned. Thus, an effective
engineering-change-order (ECO) system is needed to keep the BOMs up to date. Usually
an ECO coordinator must be appointed and charged with the responsibility for coordinating all engineering changes with the various departments involved.
Inventory
A typical computerized inventory record includes the following data segments. The item
Records
master data segment contains the part number, which is the unique item identifier, and
other information: lead time, standard cost, and so on. The inventory status segment contains a complete materials plan for each item over time. Finally, the subsidiary data segment contains information concerning outstanding orders, requested changes, detailed
demand history, and the like.
In practice, constant effort is required to keep inventory records accurate. Traditionally,
inventory accuracy has been ensured by an annual physical inventory count, where the
plant is shut down for a day or two and everything is counted from wall to wall. Because
inexperienced people are often doing the counting, as many errors are introduced by this
procedure as are corrected. After the inventory is taken, the total inventory in dollars is
accurate for financial purposes because the plus and minus errors cancel out. But the
Fy| 334] Part Five
Inventory
counts of individual items are usually not accurate enough for material planning. As a result, cycle counting has been developed as a
substitute for the annual physical inventory.
With cycle counting, a small percentage of the items are counted
each day by storeroom personnel. Errors are corrected in the records,
and an attempt is made to find and correct the procedure that caused
them. By developing a high regard for accuracy and adopting daily
cycle counting, firms can eliminate most errors in inventory records.
The result is so reliable that many auditors no longer require an
annual physical inventory when an effective cycle-counting system
Cycle counting ensures accuracy of inventory
is in place.!
records.
Alistair Berg/Digital Vision/Getty Images
Capacity
Planning
The purpose of capacity planning is to aid management in checking
on the validity of the master schedule. There are two ways this can
be done: shop loading or finite capacity scheduling. When shop loading is used, a full
parts explosion is run before capacity planning. The resulting shop orders are then loaded
against work centers through the use of detailed parts-routing data. As a result, workforce
and machine hours for each work center are projected into the future. If sufficient capacity
is not available, management should adjust either capacity or the master schedule until it is
feasible. At this point, a valid materials plan is available.
An alternative to shop loading is scheduling to finite capacity. Finite capacity scheduling establishes a work schedule to be produced in a certain time period, considering relevant limitations of resources. Software has been developed to implement methods for
forward scheduling to finite capacity. An MRP system can, therefore, be modified to start
with a feasible finite capacity schedule, and materials are planned to arrive in time to support the feasible schedule.
Purchasing
The purchasing function in the firm is helped by the use of an MRP system. Past-due
orders are largely eliminated because an MRP system generates valid due dates and keeps
them up to date.
By developing and executing a valid materials plan, management can eliminate much of
the order expediting that usually is done by purchasing. This allows the purchasing managers to concentrate on their primary function: qualifying suppliers, looking for alternative
sources of supply, and working with suppliers to ensure delivery of quality parts, on time,
at low cost.
With an MRP system, it is possible to provide suppliers with reports of planned future
orders. This gives them time to plan capacity before actual orders are placed. The practice
of giving suppliers planned orders interlocks them more closely with the firm’s own
materials plan and helps to coordinate the supply chain. Many firms have gone so far as
to insist that their suppliers also install MRP systems so that their delivery reliability
can be more readily ensured.
Also, electronic data interchange (EDI) and cloud-based
systems are being used to transmit planned order releases by an MRP system directly to
suppliers.
Shop-Floor
Control
The purpose of the shop-floor control subsystem is to release orders to the shop floor and
manage the orders on their way through the factory to make sure they are completed on
' Accounting students should take note of this concept, since they are likely to encounter cycle counting
in practice.
Chapter 15
Materials Requirements Planning and ERP
335 ||]
time. The shop-floor control system helps management adjust to all the day-to-day things
that go wrong in manufacturing: absenteeism among workers, machine breakdowns, loss of
materials, and so on. When these unplanned complications arise, decisions must be made
about what to do next. Good decision making requires information on job priorities from the
shop-floor control system, also called a manufacturing execution system (MES).
Job priorities are calculated by dispatching rules that determine the sequence of work to be
performed. Dispatching rules (e.g., first come, first served) are particularly useful when there
is more than one job waiting to be processed by a resource. When these dispatching rules are
used as part of the shop-floor control system, it is possible to adjust to changing conditions
and still get the work out on time. Through the use of dispatching rules, a job’s production lead
time can be drastically cut or increased as it goes through the shop. This is possible because
a job normally spends as much as 90 percent of its time waiting in queues. If a job is behind
schedule, its priority can be increased until it gets back on schedule. Similarly, a job can be
slowed down if it is ahead of schedule. It is the function of the shop-floor control system to
provide information to managers so that they can manage production lead time dynamically.
Lead times can be managed by expanding or contracting them on the basis of priority.
This concept has been popularized by the old saying, “Lead time is what you say it is.”
This is a very difficult concept to accept when managers are used to thinking in terms of
fixed lead time or lead times as random variables.
It is possible through a shop-floor control system to de-expedite orders—that is, to slow
them down. This is not done in normal manufacturing, where orders are expedited but
never de-expedited. Orders should be slowed down when the master schedule is changed
or when other parts will not be available on time. This results in the minimum inventory
consistent with MRP timing requirements.
15.5
OPERATING
LO15.5 Discuss
DRP and different
ways to deal with
uncertain demand.
AN MRP SYSTEM
There is much more to the MRP system than just installing the proper software modules.
Management must operate the system in an intelligent and effective way.
One of the decisions management should make is how much safety stock to carry. To
the surprise of many managers, little safety stock is needed if the MRP system is properly
used. This is due to the concept of lead time management, where both purchasing and shop
lead times are effectively controlled within small variances. In purchasing, this is done by
developing relationships with suppliers who provide reliable deliveries. In the shop, lead
times can be managed by a shop-floor control system as described above. Once the uncertainty in lead time is reduced, there is much less need for safety stock.
If safety stock were carried at the component-part level, a great deal of it would be
needed to be effective. Suppose, for example, that 10 parts are required to make an assembly and each part has a 90 percent service level. The probability of having all 10 parts on
hand when needed is only 35 percent.’ It is much better, therefore, to plan and control the
timing of the 10 parts than to cover all contingencies with safety stock. When safety stock
is carried, it often is added at the master schedule level. This ensures that matched sets of
components, not simply an assortment of various parts, are available for final products.
The purpose of safety stock at the master schedule level is to provide flexibility to meet
changing customer requirements.
Safety lead time is a concept that should be considered for component parts. If a supplier is unreliable and the situation cannot be remedied, the planned lead time can be
2 Probability = (.9)'° = .35, assuming parts availabilities are independent events.
H} |336,
Part Five
Inventory
lengthened by adding safety lead time. This will add to overall inventory levels, however,
when the supplier delivers within the original lead time.
A third way of handling uncertainty is to plan for safety capacity. This approach means
planning production levels below the actual capacity. The extra capacity is used to respond
to uncertainties such as late supplier delivery or incorrect production order size. The problem with safety stock is that it is frequently available for the wrong parts—too much of one
part and too little of another. Thus, serious consideration should be given to safety capacity
as an alternative to safety stock; this has not been widely done in industry. Rather, safety
stock (inventory) has been considered an asset, even if it is never used, and high capacity
utilization is a desirable goal, even if excess inventories result.
Management must also decide on the scope of its MRP application. MRP can be
extended in the supply chain through distribution all the way to the final customer. In this
case the application is called distribution requirements planning (DRP). When DRP is
implemented, for example in retail, it can start with the retailer who makes a forecast of
future demand and then does a time-phased plan by netting inventory and offsetting lead
times to pass planned orders on to the wholesaler. The wholesaler in turn aggregates all
planned orders from various retailers as its gross requirement and then constructs a timephased plan by netting inventory and offsetting lead times to arrive at planned orders for
the manufacturer. In this way the entire downstream supply chain is linked together.
The same logic can be applied to upstream suppliers. The manufacturer can provide timephased planned orders to its first tier suppliers to give them visibility about future demands.
The first tier supplier can use these planned orders as gross requirements together with planned
orders from other customers in its MRP system. The first tier supplier then constructs a timephased material plan by netting inventory and offsetting lead time to provide planned orders
to second tier suppliers and so on. This links the manufacturer to the upstream supply chain.
There must be sufficient safety stock in the system to prevent a tightly linked system
from becoming “neryous.” Every small change in future downstream planned orders
should not trigger an immediate change in the master schedules of the suppliers. This
would cause too many small changes in the supply chain, only to be adjusted later when
conditions change once again.
15.6
LO15.6
THE SUCCESSFUL MRP SYSTEM
Explain the
five requirements
for a successful MRP
system.
The logic embedded in an MRP system is relatively simple and straightforward. But, some
firms have tried to implement and deploy an MRP system without success. For the MRP
system to be successfully implemented and consequently deployed, at least five issues
must be addressed.
Implementation planning must, first and foremost, be a prerequisite to any effort to
implement and deploy an MRP system. Unfortunately, too many firms jump in and start
implementing MRP without adequate preparation. Advanced planning and problemprevention efforts can help smooth out implementation efforts.
Second, there must be appropriate and adequate IT support available. An adequate computer system is probably one of the easiest elements of MRP to implement. Today, there
are approximately 100 MRP software packages on the market. Most firms use these standard packages rather than writing their own software.
Third, an MRP system requires accurate data. Some firms are accustomed to lax record
keeping in manufacturing because they have always been managed by informal systems. But
accurate data are required when decisions are made from information supplied by the MRP
system. Accurate data include inventory records, bills of materials, and the master schedule.
Chapter 15
Materials Requirements Planning and ERP
337 ||)
MRP systems can
be used for services
too. In renovating
hotel rooms, Marriott
develops a bill of
materials and a bill
of labor for each
room type and then
“explodes” the bill
throughout the hotel
facility to summarize
its furniture and
decorating needs.
Onoky/SuperStock
Fourth, the importance of management support to the successful MRP system cannot be
overemphasized. But management support requires more than lip service and passive support on the managers’ part. “Management participation” or “leadership” would be a better
phrase. The ultimate change required by management at all levels is to use the system, not
to override it by management edicts and arbitrary decisions.
The fifth and final requirement for the successful MRP system is user knowledge at
all levels of the firm. An MRP system requires an entirely new approach to manufacturing. All employees must understand how they will be affected and grasp their new
roles and responsibilities. All supervisors, middle managers, and top managers need to
understand the MRP system including those inside and outside of manufacturing. As
the MRP system is broadened in scope, the extent of training within the firm must be
broadened too.
There is tremendous room for the application of MRP system elements in the service
industry. If the bill of materials is replaced by a bill of labor or a bill of activities, one
can explode the master schedule of output into all the activities and personnel required
to deliver a particular mix of services. Some service operations will also require a bill of
materials when materials are an important part of the goods-services bundle.
As an example, one electric utility has been using an MRP system for several years
in the electric hookup part of its business. When a new customer requests electrical service, a planner enters the request into a computer system for the type of service required.
The computer then explodes this service request into detailed labor, material, and work
activities. Each of these requirements is time-phased and accumulated over all jobs to
determine whether sufficient capacity is available. When the time comes, the utility
hookup crews are given work orders from the system output, and completed work is
entered back to the system. The MRP system then drives billing, labor-reporting, and
other accounting systems.
15.7
LO15.7
ENTERPRISE RESOURCE
Describe
what an ERP system
does.
PLANNING SYSTEMS
So far, we have been discussing the use of MRP systems in manufacturing and service operations. While MRP is the base for operations planning and control, it can also be
extended into all other business functions through the use of an enterprise resource planning
(ERP) system. For example, MRP transactions can be fed directly to the accounting and
finance system. Accounting transactions can be seen as putting MRP transactions into
|] | 338)
Part Five
Inventory
dollars-and-cents terms. Accounting control in dollars is closely related to control of the
physical units and the physical flow of inventory in operations.
Likewise, an ERP system is useful for linking marketing to operations. Marketing and
sales transactions should be closely integrated as inputs to the MRP through order entry.
Often, marketing and sales systems are developed and designed as separate systems that
are not fully integrated with the MRP system in operations. As a result, the functions of
marketing and operations are isolated from an information systems point of view.
Finally, transactions in a manufacturing or service firm should be integrated with its
human resource system. This takes place not only through payroll transactions but also
through recruiting and selection activities as well. For example, when operations makes a
decision to expand capacity by hiring more people, this decision should be fed directly to
the HR system and the hiring process should be tracked through to completion. The payroll
system should pick up the new people hired by operations, and these employees should be
followed through their lifetime of employment.
When operations, finance/accounting, marketing/sales, and HR systems are integrated
through a common database, the ERP system is completed. The ERP system will track
transactions from their origin at the customer, to order entry, through operations and
accounting until the transaction is completed. Also, all decisions made in one function
will be apparent to other functions and reflected in their information systems. No longer
will the various functional information systems be isolated; rather, they will be integrated
through a common database.
For example, the type of ERP system we have been discussing has been developed
by SAP, a German firm. SAP has hundreds of thousands of customers across 180 coun-
tries and serves numerous industries with software and cloud services. One advantage
of the SAP system is that it can be tailored to different firms and industries. Standard
processes, including order entry, payroll, MRP, inventory management, accounts payable, and accounts receivable, have been designed into the SAP system. Each firm can
select those processes that fit its needs and customize them to its particular business
(see Table 15.5).
TABLE 15.5
ERP: CrossFunctional
Integration Through
Data Sharing
An ERP system integrates data across functions in the firm. This list shows some of the many
functions supported by SAP’s ERP package.
Financial Systems
Accounts receivable and payable
Asset accounting
Cash management and forecasting
Cost-element and cost-centered accounting
Executive information system
Financial consolidation
General ledger
Product-cost accounting
Profitability analysis
Profit-center accounting
Standard and period-related costing
Human Resources
Human resources accounting
Payroll
Personnel planning
Travel expenses
Operations and Logistics
Inventory management
Materials requirements planning
Materials management
Plant management
Production planning
Project management
Purchasing
Quality management
Routing management
Shipping
Vendor evaluation
Sales and Marketing
Order management
Pricing
Sales management
Sales planning
Chapter 15
Materials Requirements Planning and ERP
339 ||]
Operations Leader
LG Electronics ne
Global
giant LG
subsidiaries
and
Electronics,
based
in Korea,
more
82,000
employees
than
has
114
across
40 countries. In the past, data management challenges
made it difficult for LG to operate as a global company.
They continue to develop and expand their ERP system
to meet their needs.
LG’s past challenges included
local information
sys-
tems leading to unclear reporting, local processes lacking transparency and automation, limited sharing of best
practices, difficulties with decision
gaged employees.
making,
and disen-
Their ERP solution has supported the following benefits:
Kobby Dagan/123RF
centrally managed system with minimal maintenance costs,
transparency in processes, easy sharing of best practices
across locations, and real-time reporting for upper man-
LG continues to expand their system, as needed.
Building a single centralized system that integrates available
modules for uniform use at a global level replaces their
previous dependency on local systems.
agement for informed decision making. Employee morale,
productivity, and engagement have improved.
Their ERP system enables LG to mitigate the challenges posed by their numerous global locations. Today,
Source: www.lg.com/us, 2020.
Oracle is another ERP software vendor. Cleveland Clinic, a leader in health care deliv-
ery, uses Oracle systems in their domestic and international locations. Systems developed
specifically for the health care industry are designed to benefit back-office personnel,
medical professionals, and patients. The systems are used to enhance financial forecasting, revenue management, and data integration across functions. For operations, the Oracle modules support supply procurement and inventory management. See the Operations
Leader box about use of ERP at LG Electronics.
ERP systems are the basis for cross-functional integration. When all functions share
information through a common database the functional silos are minimized and functions
can communicate effectively with each other. Once ERP systems are fully operating inside
a firm they can be integrated with suppliers and customers. A variety of information can be
exchanged to help coordinate decisions along the supply chain. Supply chain coordination
can take the form of Collaborative Planning and Forecasting and also planning for new
product introductions with suppliers and customers.
Many firms have determined that their existing information systems, which have
grown up separately, can no longer meet the needs of the business and must be integrated through an ERP system. ERP system implementation, however, is expensive,
time-consuming, risky, and nontrivial. ERP system implementation failures are not
uncommon. For example, Avon abandoned its ERP implementation in 2013 after having
spent over $125 million, and Waste Management ended up in court while suing (and being
counter-sued by) SAP from a failed implementation. ERP implementation is a worthwhile
endeavor with great benefits; realizing these benefits requires careful planning, significant resources, and patience in execution.
)} | 340| Part Five
15.8
Inventory
KEY POINTS AND TERMS
This chapter describes MRP and ERP systems for managing dependent demand inventory.
The major points include the following:
An MRP system is an information system used to plan and control manufacturing of
dependent demand inventory items.
The parts-explosion process has three principal inputs: master schedule, bill of materials (BOM), and inventory records. There are two principal outputs: purchase orders and
shop orders. Parts explosion is the heart of the MRP system.
An MRP system follows a requirements philosophy—parts are ordered only as required
by the master schedule. Past demand for parts is irrelevant, and component inventories
are not replenished when they reach a low level.
Master schedules should be based on both marketing and production considerations.
They should represent a realistic build plan within factory capacity. Management
should use the S&OP and master schedule to plan and control the business through a
cross-functional planning team.
The BOM contains the list of parts used to make the product. To maintain the accuracy
of the BOM, an engineering-change-order system is needed.
The accuracy of the inventory record system should be maintained through cycle counting. Daily cycle counting can be used in place of annual physical inventory count.
Shop-floor control is used to control the flow of materials through the factory. This is
done by managing lead times dynamically as the product is manufactured. If lead times
are properly managed, much safety stock can be eliminated.
A successful MRP system requires (1) implementation planning, (2) adequate IT
support, (3) accurate data, (4) management support, and (5) user knowledge. Both
system and people problems must be solved to use MRP successfully. When this is
done, benefits include reduced inventory, increased customer service, and improved
efficiency.
Enterprise resource planning systems integrate a basic MRP system with information from marketing, sales, finance/accounting, and human resources through a common database. ERP systems can be the basis for cross-functional and supply chain
integration.
Key Terms
Materials requirements
planning (MRP) 323
Master schedule 324
Parts explosion 324
Dependent demand 326
Bill of materials 328
Planned lead times 328
Materials plan 328
Gross requirements 329
Net requirement
329
Planned order receipt 329
Lot-for-lot (L4L)
production 329
Planned order releases 329
Time-phased materials
plan 331
Purchase orders 331
Shop orders 331
Engineering-changeorder
333
Cycle counting 334
Shop loading 334
Manufacturing execution
system 335
Safety lead time 335
Safety capacity 336
Distribution requirements
planning (DRP) 336
Enterprise resource planning
(ERP)
337
Chapter
15
Materials Requirements Planning and ERP
LEARNING
ENRICHMENT
SAP ERP System
https://www.sap.com/products/enterprise-management-erp.html
Website
(for self-study or
Oracle ERP System
Website
pote
ores Gnine nts)
341] |]
https://www.oracle.com/applications/erp/what-is-erp.html
Example of Using an MRP System
https://youtu.be/BItcluapE6g
Video
6:57
Elements in a Bill of Materials
https://www.thebalancesmb.com/bill-of-materials-222 1363
Website
ERP Solutions for Manufacturing
https://youtu.be/YA68o0lpZbk8
Video
2:09
SOLVED PROBLEM
Problem
1. MRP
a. Using the information given below, develop a complete MRP material plan.
End item
(lead time = 1)
Component B
(lead time = 2)
Component C
(lead time = 1)
It takes one unit of B and two units of C to make one unit of A. At the beginning of
time period 1, the following information is available:
Item ID
Quantity on Hand
Lead Time
A
100
1
B
150
2)
Cc
80
1
The gross requirements of item A are 200 units for period 4 and 250 units for period
5. Use L4L lot size planning.
b. If the lead time for item A increases by one week and the lead time for item C also
increases by one week, what will the revised materials plan look like? Are there any
problems that need immediate attention?
c. Returning to the material plan for part a, what is the effect of using lot sizes of 200
for all items A, B, and C? You many use multiples of 200 when needed to satisfy
planned order receipts.
Py |342,
Part Five
Inventory
a. The MRP materials plan is as follows:
Week
1
2
S
4
5
200
250
Item A
On hand = 100
Gross Requirement
LT = 1 wk
Lot size: L4L
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Planned order releases
Safety Stock = O
On hand = 150
LT = 2 wk
Lot size: L4L
Safety Stock = 0
100
Item B
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Item C
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Planned order releases
100
100250
250
OORE
100937250
1007
EXO)
TO)
3250
50
200
200
Planned order releases
On hand = 80
LT = 1 wk
Lot size: L4L
Safety Stock = O
100
200
2OORmS90
80
80
1 ZOOM
120
2 ORD OO
SOO
500
b. The revised MRP materials plan is as follows:
Week
1
On hand = 100
LT = 2 wk
Lot size: L4L
Safety Stock =O
On hand = 150
LT = 2 wk
Lot size: L4L
Safety Stock = O
On hand = 80
LT = 2 wk
Lot size: L4L
Safety Stock = 0
Item A
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Planned order releases
Item B
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Planned order releases
Item C
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
LOOR
150
100m
3
1@©)
Ase
NOOR
250
50
200
200
500
80
IZ0==500
120
500
4
5
200
250
OOM
1OOR
256
S250
eRaOO
200
200
Planned order receipts
Planned order releases
2
500
Chapter
15
Materials Requirements Planning and ERP
343 ||)
Yes, there is a problem that needs immediate attention. Item C is behind schedule
and will need to be expedited immediately to ensure that 120 units of C are available in week 2. Alternatively, the master schedule could be revised to accommo-
date the availability of item C. Combinations of these two alternatives could also be
considered.
c. The revised materials plan is as follows. Note, although the lot size is 200, we need
to order two lots for item C of 200 each to exceed net requirements for this product.
Week
1
On hand = 100
ae leWwi<
Lot size = 200
Safety Stock = O
On hand = 150
LT = 2 wk
Lot size = 200
Safety Stock = 0
Item A
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Planned order releases
Item B
Gross Requirement
Scheduled Receipts
Projected Ending Inventory
Net Requirement
Planned order receipts
Planned order releases
Item C
On hand = 80
Gross Requirement
LT = 1 wk
Scheduled Receipts
Lot size = 200 x 2.
Projected Ending Inventory
Safety Stock = O
Net Requirement
Planned order receipts
Planned order releases
OOP
2
80
4
5
ZOO
M250
50
P1150
200
200
100
TOOR
200
200
200
200
tal5O
50
200
a>Oneat 5
50
200
400
400
80
S205
400
400
80
SAO)
400
POO TS 100k
TSO ma SOV
200
3
200
80
400
80
Discussion Questions
1. In what ways do independent demand inventories differ
from dependent demand inventories?
7. Is it possible to control financial totals without physical
control of materials in manufacturing?
2. Why is demand history irrelevant for the management
of raw materials and work-in-process inventories?
8. A company president said his firm is too small to afford
an MRP system. Discuss.
3. With regard to inventory management, discuss the
difference between a replenishment philosophy and a
requirements philosophy.
9. A materials manager said that her firm needs only a
replenishment inventory management system. What
would you tell her about the additional capabilities of
4. Can ABC inventory classification be applied to manufacturing component inventories? Discuss.
5. How much safety stock should be carried in an MRP
system? What is the role of safety stock in MRP systems? Where should safety stock be carried?
6. Describe the advantages of cycle counting over an
annual physical inventory count.
an MRP system?
10. Describe how MRP concepts could be used for the following service operations:
a. Hotel
b. Legal office
11. How are MRP and ERP related?
| | 344
Part Five
Inventory
Problems
Three Excel spreadsheets are provided on Connect for
assistance in solving the chapter problems
The firm would like to produce 600 chairs in week 5
and 300 chairs in week 6.
1. The following information is given for a particular part.
Using a lead time of two weeks, complete the table. Use
L4L with On hand = 80 and SS = 0.
a. Develop a materials plan for all the parts using L4L.
b. What actions should be taken now?
c.
Week
1
Gross requirement
Scheduled receipts
Projected ending
inventory
Net requirement
Planned order receipts
Planned order
2
3
4
5
100
400
300
50
Assume it takes one hour to assemble backs, one
hour to assemble legs, and two hours to finish completed chairs. Total assembly time for all three types
of assembly is limited to 1000 hours per week. Will
this capacity constraint cause a bottleneck in assembly? If it does, what can be done?
d. What is the effect of changing the master schedule
to 300 chairs in week 5 and 400 chairs in week 6?
80
. Product A consists of subassemblies B and C. Subas-
sembly B requires two parts of D and one part of E.
Subassembly C requires one part of D and one
part of F.
releases
a. Draw a product structure tree (BOM) for this
2. Using the original information in problem 1, complete
the table again, but this time assuming that the supplier
requires a fixed lot size of 500 units.
3. Again using the original information in problem | and
L4L, complete the table with a requirement of safety
stock of 100 units.
eXce/
4. The Old Hickory Furniture Company
manufactures chairs on the basis of the BOM
shown below. At the present time, the inventories of
parts and lead times are as follows:
product.
b. How many parts are needed to make 300 units of
finished product?
eXce/
6. The BOM for product A is given below:
Part
On Hand
Weeks of Lead Time
A
B
(E
D
75
150
50
100
1
2
1
2
A
|
Leg
assembly
Back
assembly
rer
Legs
Top
D(2)
B
Assume the master schedule calls for 200 units of
Spindles
(4)
C(2)
D
product A in week 5 and 250 units in week 6.
(4)
Wsevi4ie
a. Develop a materials plan for this product.
b. What actions should be taken immediately?
Chairs
Leg assembly
Back assembly
Seat
Rails
Legs
Top
Spindles
On Hand
Weeks of Lead Time
100
50
25
40
100
150
30
80
1
2
1
iS}
1
1
2
2
c. Project the inventory ahead for each part.
d. If you were suddenly notified that part D will take
three weeks to get instead of two weeks, what
actions would you take?
. The master scheduler in the ABC Widget Company is
in the process of revising the master schedule. At the
present time, he has scheduled 400 widgets for week 5
and is considering changing this to 500 widgets.
a. What information would you need to decide whether
you should make this change?
Chapter 15
Materials Requirements Planning and ERP
345 ||)
b. Suppose that each widget takes one hour of press
a. Assume the master schedule calls for 200 scissors to
time and three hours of assembly time three weeks
before delivery. Can the additional 100 widgets be
be shipped in week 4 and 500 in week 5; work out a
complete materials plan using L4L.
made in view of the following shop loadings?
Press Hours
500
b. Suppose the supplier of right-hand sides calls to say
that deliveries of the 200 parts on order will be one
_ Maximum available
week late. What effect will this have on the materials plan?
400)
c. If demand for scissors is uncertain and has a standard deviation of 50 units, what would you recommend the firm do to maintain a 95 percent service
300
level for scissors?
d. If the delivery of the scissors parts is unreliable
and the standard deviation of delivery lead time
is one week for each of the parts, what would you
recommend the firm do to maintain its production
schedule?
Shop
Week
by
Press
Hours
eXce!
9. A lamp consists of a frame assembly and
a shade, as shown below in the sketch and
the bill of materials. The frame is made from a neck, a
socket, and a base, which are assembled together from
Assembly Hours
Maximum available
.
purchased parts. A shade is added to the frame assembly to make the finished lamp. The number of parts on
hand, the parts scheduled to arrive, and the lead times
to obtain more parts are shown below.
Hours
Assembly
Week
by
c. If the additional widgets cannot be made in part b
above, what actions might be taken to make it possible to produce the required widgets?
Neck
Socket
Base
8. A firm makes a basic scissors consisting of three parts:
the left side, the right side, and the screw that holds the
sides together. At the present time, the firm has the fol-
Weeks of
lowing numbers of parts on hand and on order. The lead
times for reorder of each part are also shown along with
a BOM and a sketch of the scissors.
OnHand
Lamp
Frame
Neck
Scissors
|
Right
side
Scheduled
LeadTime _ Receipts
200
100
1
2
bie
ae
—
O
1
Socket
300
1
oak
Base
Shade
200
400
1
S
=
_
a. Assuming 1000 finished lamps are required in
week 5 and 1500 in week 6, construct a complete
Seeare
er
Right side
ein
materials plan for the lamp using L4L. What actions
should be taken immediately to implement the plan?
Weeks of
Scheduled
On Hand
100
50
Lead Time
1
>
Receipts
4100 In waa?
75
5
SOON Week2
total assembly time is required in each week? What
300
1
SOON Weeke
can be done if insufficient time is available in any
given week?
b. If it takes 15 minutes of assembly time to assemble
the parts into the frame and 5 minutes to assemble
the shade and frame into a finished lamp, how much
| |346| Part Five
Inventory
c. If the lead time for assembly of lamps is extended
from one week to two weeks, what changes will
be needed in the materials plan to adjust for this
change?
10. A telephone is assembled from a handset and a base.
The handset in turn is assembled from a handle and a
Mt. A small toy robot is assembled from six parts: a body, a
head, two arms, and two legs. The firm uses a one-level
bill of materials to assemble this product. The number
of parts on hand and the lead times (weeks) to obtain
more parts are shown below. There are no parts
on order.
cord, and the base is assembled from a case, a circuit
board, and a face plate. A BOM and a sketch of the
phone, along with the numbers of parts on hand and
lead times, are shown as follows.
Handset
Handle
Toy robot
Cord
Case
Circuit
Face
board
plate
On Hand
Weeks of Lead Time
Phone
200
1
Handset
Handle
Cord
Base
Case
300
250
Us
250
225
|
Z
2
1
2
Circuit board
150
‘Eiad
Face plate
300
2
a. Management would like to start assembling phones
as soon as possible. How many phones can be made
from the available parts, and when can they be
delivered? Construct a materials plan to show your
answer.
Body
Head
Arm
Leg
On Hand
Lead Time
Ds.
50
60
80
2
1
2
1
a. Assume that an order for 200 robots is received now
for delivery at the beginning of week 4 and that it
takes one week to assemble the parts once they are
all available. Construct a complete materials plan for
the robots using L4L. What actions should be taken
immediately to implement the plan?
b. The customer has called and asked if he could
receive a portion of the 200 robots as soon as
possible. How many robots can be assembled and
delivered to him ASAP, and when would they
arrive? What are the implications of this action?
b. For the materials plan constructed in part a, develop
an inventory projection of parts and finished goods
on a week-by-week basis using L4L.
c. The supplier of heads has just sent an e-mail that
said it will take two weeks to deliver the heads
instead of one week. What effect will this have on
your materials plan from part a?
c. Suppose another 100 circuit boards can be obtained
within one week. What effect will this have on your
answer for part a?
. Using the information from problem 11, complete the
plan for the robot parts assuming that a fixed lot size of
400 units is required for all parts.
PART
Supply Chain Decisions
Zuma
www
&
16.
Supply Chain Management
17.
Sourcing
18.
Global Logistics
This part deals with external supply chain decisions that connect operations to its
suppliers and customers. The first chapter addresses general supply chain issues,
followed by sourcing and global logistics chapters. Sourcing connects operations
with its service and manufacturing suppliers. Logistics deals with moving materials,
either from operations to customers or suppliers to operations.
@
Coin
AIP IP WS Is
Supply Chain
Management
LO16.2
LO16.4
You probably never thought about the everyday impact and importance of supply chains.
Americans eat an average of five pounds of food every day. A small city of 100,000 people
requires 500,000 pounds of food shipped every day, assuming no inventory buildup or
reduction. This is a huge amount of food that must be transported from the farm all the way
to restaurants or tables in homes. The fresh fruit, meats, and vegetables are refrigerated
and have a short shelf life. They require special and fast transportation from the farm to
packaging plants, to deconsolidation warehouses and then finally to the retail stores. Other
products, such as packaged foods, travel by train, ocean, and truck through manufacturing,
distribution centers, and retail stores until they reach your table. The global food supply
chain must be reliable, low cost, and resilient to disruptions (e.g., natural and human disas-
24s
ters). This is no small task and supply chain managers make it happen every day.
Chapter 16
Supply Chain Management
1349 ||)
In this chapter, we introduce the subject of supply chain management, which has generated a great deal of interest recently in industry and academia. There are several reasons
for this sudden interest. First, the total time for materials to travel through the entire supply
chain can be six months to a year or more. Since the materials spend so much time waiting in
inventory, there is a great opportunity to reduce the total supply chain cycle time, leading to
a corresponding reduction in inventory, increased flexibility, reduced costs, and better deliveries. Also, many companies have improved their internal operations dramatically and now
find it necessary to consider relations with external customers and suppliers in the supply
chain to gain further improvements in operations. Finally, supply chain thinking is an application of systems thinking and provides a basis for understanding processes that cut across
a company’s internal departments and processes that extend outside the company as well.
16.1
SUPPLY CHAIN AND SUPPLY CHAIN MANAGEMENT
LO16.1 Define
supply chain and
supply chain
management.
FIGURE
16.1
Upstream or Backward Materials and Information Flow
A typical supply chain
from the perspective of
dias
Supply chain management is an essential aspect of business today. To understand what supply chain management entails, we first provide a formal description of a supply chain.
Although a supply chain generally can be depicted without specifying a vantage point, it is
often more useful to show a supply chain from the perspective of a firm, a factory, a service
delivery unit for a firm, a product family (e.g., automobiles), or even a type of service (e.g.,
outpatient surgeries). Figure 16.1, for example, shows the entire supply chain for an organization called F, the focal entity. The organization F, in this case, can be the entire firm or a
specific factory or hotel property. The various nodes or ovals represent other facilities through
which materials and requisite information flow for the product that F sells to end customers.
The flow of materials from upstream nodes into the focal entity generally is referred to
as physical supply and the flow of materials from F through downstream nodes toward
Me
BOC a CRY
abe
ep lanamienG
6
Fe
teisetier
r
Is
'
5
Distribution
@
#
rJ
&®
a
‘7
Suppliers
.
5
.
#
Centersand
:
Warehouses
& _ Retailers
8;
'
a
‘
5
a
a
2
a
a
rT
a
a
Suppliers
a
Downstream or Forward Materials and Information Flow
i
H] | 350, Part Six
Supply Chain Decisions
CC
anyC
Kena
1a
ii 8
i
OPERATIONS LEADER
beha Pl
|
\
}
ERI
= at
|
ea
bi
mli
SNARES
Many high-tech consumer products are not manufactured by the
companies
whose
brands
they carry, but instead
are manufac-
tured by a host of other companies that act as 1st-tier, 2nd-tier,
and 3rd-tier suppliers. The iPhone design, software development, chip design, and marketing are done in the U.S. by Apple.
The manufacturing is outsourced to Taiwan, China, Korea, Japan,
and Singapore along with other countries. The iPhone is being
assembled by the millions of units in China by FoxConn, a huge
Taiwan-based electronics company. Taiwan Semiconductor Manufacturing Company (TSMC), provides Apple’s custom-designed
chip—the most important hardware component of the phone.
The digital camera modules, internal circuitry, Bluetooth chipsets,
screen
controllers,
and
other
components
are
provided
by 2nd- and 3rd-tier suppliers. This illustrates the complex and
global supply chain used by Apple.
adrianhancu/123RF
Sources: “The Global Supply Chain behind the iPhone,” Beta News,
2015; and Apple A12 processor, wikipedia.com, 2020.
the end customers is referred to as physical distribution. Distribution channel, a term
frequently used in marketing, is a specific route from a producer (in this case, F) forward
through the various nodes (e.g., distributors and wholesalers) to the end customer and is
therefore only part of the supply chain for F. Notice also that the physical supply for the
focal entity can be ségmented further into tiers such that Ist-tier suppliers have a direct
linkage (represented by an arrow) to F, 2nd-tier suppliers have a linkage to F through Isttier suppliers, and so on. Similarly, the focal entity has linkages to downstream entities
(distributors and wholesalers) as part of its physical distribution. Each entity in Figure 16.1
ideally plays a value-added role in transforming materials and services into the desired
final product for the customer while passing along relevant information. The Operations
Leader box titled “Apple’s iPhone” describes the integrative roles that different players
assume in bringing forth today’s increasingly complex products.
A large company will have several supply chains. For example, large companies such as
Procter & Gamble and General Electric may use 20 to 40 different supply chains to bring
their products to market. Some of these supply chains use distribution through companyowned warehouses, some use direct distribution, some use outside manufacturing, and
some use in-house manufacturing sites. The elements in a supply chain can be arranged in
many different ways.
A company can identify its supply chains by first selecting a particular product group or
product family. Then it should trace the flow of materials and information from the final
customer (end user) backward through the distribution system to the manufacturer and
then to the suppliers and the sources of raw materials. This entire chain of activities and
processes constitutes the supply chain for that product group.
Armed with this understanding of a supply chain, a useful definition of supply chain
management (SCM) taken from the Institute for Supply Management is as follows:
Supply chain management is the design and management ofseamless, value-added processes
across organizational boundaries to meet the real needs of the end customer.
Chapter 16
Supply Chain Management
351] |]
SCM, by this definition, involves a sequence of value-added processes that not only cut
across organizational boundaries but also must be tightly integrated. To be integrated, the
processes must be appropriately designed and systematically managed to allow information to flow and be deployed within and across them. The design and management of
these processes therefore require that decisions be made to implement strategies and solve
problems to ensure an effective and efficient flow of materials and requisite information
across the entire supply chain. These strategies and problem resolutions are intended to
reduce uncertainty across the entire supply chain. Taking a systems perspective is therefore
paramount so that strategies and problem resolutions to reduce uncertainty at one node do
not end up negatively affecting another node. In Section 16.3, we demonstrate more clearly
the system dynamics inherent in supply chains and explain why supply chains and their
management must be viewed in a holistic fashion.
The SCOR or Supply Chain Operations Reference model is described in the Operations
Leader box. This model defines the processes of plan, source, make, deliver and return that
apply to each organization in the supply chain. By providing a common reference model,
supply chain designers in each organization can refer to the processes needed in a uniform
manner that allows for integration of the supply chain. The SCOR model is an essential
tool used for supply chain design across organizations.
Besides defining SCM from a process or decision-making orientation, many scholars and managers define SCM as the integration of three traditionally separate functions:
sourcing (purchasing), operations, and logistics. With respect to our example, sourcing is
the function that deals with the physical supply to ensure the flow of materials and information into the focal entity, operations is the function that produces the product or service,
and logistics is the function that designs and manages the physical transportation of materials both inbound into the focal entity and outbound from it.
What has to be emphasized is that in recent years the three functions of sourcing, operations, and logistics have evolved and taken on responsibilities for materials and information flow. As a result, there has been a blurring of what duties, decisions, and problems
belong to sourcing or operations or logistics. For example, what used to be known as the
Council of Logistics Management has been renamed the Council of Supply Chain Management Professionals to reflect a broadened definition of logistics management that is
identical to that of SCM. Similarly, the National Association of Purchasing Managers has
changed its name to the Institute for Supply Management (ISM).
16.2
MEASURING SUPPLY CHAIN PERFORMANCE
LO16.2 Review key
measures of supply
chain performance.
While it is useful to think of a supply chain when one is considering the set of activities
needed to manufacture a product or deliver a service, measuring its performance cannot be
done in isolation from the various entities that make up the supply chain for a particular
product or service. Measuring the performance of a supply chain therefore must be done
one company at a time from each company’s individual perspective, along with a few measures that are applicable to the entire supply chain. From the standpoint of supply chain
measurement, each company can derive its own performance measures, which in turn are
affected by its supply chain partners. Finally, the performance of a company within a sup-
ply chain can affect the performance of the other companies as well.
For example, suppose a manufacturing company has three kinds of inventory: raw
materials, work in process, and finished goods. The level of raw-materials inventory is
a function of the suppliers’ lead times and the safety stock needed to handle variance in
lead times and demand. Therefore, raw-materials inventory depends on the suppliers in the
F] | 352| Part Six
Supply Chain Decisions
i
=| OPERATIONS LEADER |# 5
SCOR Model
The SCOR, or “Supply Chain Operations Reference,”
model represents a cross-industry process framework
and standard for defining what SCM entails. In this
1
i
materials into a finished product. The Deliver process
is the set of activities
involved
in order entry, materi-
als handling, and transporting of goods and services to
Customer
Supplier
Suppliers’ |
! Customer’s
eRe A ene i tear
lorextcnal
Internal or External
'Customer
Your Company
respect, it seeks to play the same role that ISO 9000
or the Baldrige Award plays with respect to total quality
management.
%
SCOR was introduced in November 1996 by the SupplyChain Council consisting of 69 firms. The 69 member
firms worked for six months to define common supply
chain management processes, best practices for those
processes, and benchmark performance data.
The current version of the SCOR model is represented as a four-level pyramid, with Level 1 (Top
Level)
defining
five distinct
SCM