Chp.12

advertisement
Operations
Management
Inventory Management
Chapter 12
Heizer/Render
Asli Sencer Erdem
1
Outline
 GLOBAL COMPANY PROFILE: AMAZON.COM
 FUNCTIONS OF INVENTORY

Types of Inventory
 INVENTORY MANAGEMENT
ABC Analysis
 Record Accuracy
 Cycle Counting
 Control of Service Inventories

Heizer/Render
Asli Sencer Erdem
2
Outline - Continued
 INVENTORY MODELS
Independent versus Dependent Demand
 Holding, Ordering, and Setup Costs

 INVENTORY MODELS FOR INDEPENDENT
DEMAND
Basic Economic Order Quantity (EOQ) Model
 Minimizing Costs
 Reorder Points
 Production Order Quantity Model
 Quantity Discount Models

Heizer/Render
Asli Sencer Erdem
3
Outline - Continued
 PROBABILISTIC MODELS WITH CONSTANT
LEAD TIME
 FIXED PERIOD (P) SYSTEMS
Heizer/Render
Asli Sencer Erdem
4
Learning Objectives
When you complete this chapter, you should be
able to :
Identify or Define:
ABC analysis
 Record accuracy
 Cycle counting
 Independent and dependent demand
 Holding, Ordering, and Setup Costs

Describe or Explain:

The functions of inventory and basic inventory
models
Heizer/Render
Asli Sencer Erdem
5
AMAZON.com
 Jeff Bezos, in 1995, started AMAZON.com as a
“virtual” retailer – no inventory, no warehouses, no
overhead; just a bunch of computers.
 Growth forced AMAZON.com to excel in inventory
management!
 AMAZON is now a worldwide leader in warehouse
management and automation.
Heizer/Render
Asli Sencer Erdem
6
Order Fulfillment at AMAZON
1. You order items;, computer assigns your order to
distribution center [closest facility that has the
product(s)]
2. Lights indicate products ordered to workers who
retrieve product and reset light.
3. Items placed in crate with items from other
orders, and crate is placed on conveyor. Bar
code on item is scanned 15 times – virtually
eliminating error.
Heizer/Render
Asli Sencer Erdem
7
Order Fulfillment at AMAZONContinued
4. Crates arrive at central point where items are
boxed and labeled with new bar code.
5. Gift wrapping done by hand (30 packages per
hour)
6. Box is packed, taped, weighed and labeled
before leaving warehouse in a truck.
7. Order appears on your doorstep within a week
Heizer/Render
Asli Sencer Erdem
8
What is Inventory?
 Stock of materials
 Stored capacity
 Examples
© 1995
Corel Corp.
© 1984-1994 T/Maker Co.
© 1984-1994 T/Maker Co.
© 1995 Corel Corp.
Heizer/Render
Asli Sencer Erdem
9
The Functions of Inventory
 To ”decouple” or separate various parts of the
production process
 To provide a stock of goods that will provide a
“selection” for customers
 To take advantage of quantity discounts
 To hedge against inflation and upward price
changes
Heizer/Render
Asli Sencer Erdem
10
Types of Inventory
 Raw material
 Work-in-progress
 Maintenance/repair/operating supply
 Finished goods
Heizer/Render
Asli Sencer Erdem
11
The Material Flow Cycle
Note the proportion of time material spends as inventory as
opposed to being actually worked on.
Effective inventory management and materials movement can
reduce overall cycle time significantly!
Heizer/Render
Asli Sencer Erdem
12
Disadvantages of Inventory
 Higher costs
Item cost (if purchased)
 Ordering (or setup) cost



Costs of forms, clerks’ wages etc.
Holding (or carrying) cost

Building lease, insurance, taxes etc.
 Difficult to control
 Hides production problems
Heizer/Render
Asli Sencer Erdem
13
ABC Analysis
 Divides on-hand inventory into 3 classes

A class, B class, C class
 Basis is usually annual $ volume

$ volume = Annual demand x Unit cost
 Policies based on ABC analysis
Develop class A suppliers more
 Give tighter physical control of A items
 Forecast A items more carefully

Heizer/Render
Asli Sencer Erdem
14
Classifying Items as ABC
Class
A
B
C
% Annual $ Usage
100
80
60
% $ Vol
80
15
5
% Items
15
30
55
A
40
B
20
C
0
0
50
100
% of Inventory Items
Heizer/Render
Asli Sencer Erdem
15
Cycle Counting
 Physically counting a sample of total inventory on
a regular basis
 Used often with ABC classification

A items counted most often (e.g., daily)
Heizer/Render
Asli Sencer Erdem
16
Advantages of Cycle Counting
 Eliminates shutdown and interruption of
production necessary for annual physical
inventories
 Eliminates annual inventory adjustments
 Provides trained personnel to audit the accuracy
of inventory
 Allows the cause of errors to be identified and
remedial action to be taken
 Maintains accurate inventory records
Heizer/Render
Asli Sencer Erdem
17
Why do we control inventory?
 Inventories represent a vast segment of total
economic activity.
 Even minor improvements can create large
savings.
How do we control inventory?
 Application of optimization techniques
 Information processing and retrieval techniques
Heizer/Render
Asli Sencer Erdem
18
Decisions of an inventory policy
 If there is no production, i.e., pure inventory system
How much to order? Order quantity
 When to order? Reorder quantity
Ex:Order Q=100 units when the inventory level drops to
ROP=15 units.

 If there is also production

When to start/stop production?
Heizer/Render
Asli Sencer Erdem
19
An inventory system
Heizer/Render
Asli Sencer Erdem
20
Elements of Inventory Decisions
 Costs:



Ordering and Procurement costs (or Setup costs in manufacturing
systems)
Inventory holding or carrying costs
Inventory shortage costs
 Demand structure

How does it vary? Certain, uncertain?
 Supply structure

Any capacity limitations, defectives, number of suppliers?
 Lead times:

Certain, uncertain?
Heizer/Render
Asli Sencer Erdem
21
Ordering and Procurement Costs
 Represent all expenses incurred in ordering or manufacturing items related
to
Acquisition
Transportation
 Collecting, sorting, placing the items in the storage
 Managerial and clerical costs associated with order placement.


 Ordering costs are fixed, independent of the order size.




Supplies
Forms
Order processing
Clerical support
 Procurement costs depend on the order size.
Heizer/Render
Asli Sencer Erdem
22
Set-up Costs
In a manufacturing system order costs are
realized in the form of “set-up” costs. These
include:
 Clean-up costs
 Re-tooling costs
 Adjustment costs
Heizer/Render
Asli Sencer Erdem
23
Inventory Holding or Carrying Costs
 Expenses incurred during the storage of items.
Physical Costs: Warehouse operation costs, insurence,
property taxes.
 Pilferage, spoilage, obsolescence
 Opportunity cost of investing in inventory rather than
investing somewhere else, ex. in a bank.

 Inventory costs are variable costs that depend on
the order size.
Heizer/Render
Asli Sencer Erdem
24
Inventory Holding Costs
(Approximate Ranges)
Category
Cost as a
% of Inventory Value
Housing costs (building rent, depreciation,
operating cost, taxes, insurance)
6%
(3 - 10%)
Material handling costs (equipment, lease
or depreciation, power, operating cost)
3%
(1 - 3.5%)
Labor cost from extra handling
3%
(3 - 5%)
Investment costs (borrowing costs, taxes,
and insurance on inventory)
11%
(6 - 24%)
Pilferage, scrap, and obsolescence
Overall carrying cost
Heizer/Render
Asli Sencer Erdem
3%
(2 - 5%)
26%
25
Shortage Costs
 Occur whenever the demand is not satisfied. Order is
either “backordered” or “lost”.
 Backordering Costs:
Fixed cost of extra managerial work.
 Loss of customer goodwill: Variable cost that depends on
duration of backorder.

 Lost Sales Costs:
Marginal profit that the item would have earned.
 Loss of customer goodwill.

Heizer/Render
Asli Sencer Erdem
26
Demand Structure
 Continuous versus discrete demand
Ex: Natural gas consumption in houses
Detergent consumption in houses
 Deterministic (certain) versus stochastic (uncertain)
demand
Ex: Order quantities for the next months are 20,30,10,50.
Order quantities in a month are normally distributed with mean 25
and variance 4.
 Constant versus dynamic demand
Ex: Demand quantities for the next months are 20, 21, 20, 19
Demand quantities for the next months are 20, 50, 10, 2
Heizer/Render
Asli Sencer Erdem
27
Supply Structure
 Any defectives?
If the received lot includes defective items this
brings uncertainty
 Any capacity limitations?
Do we fully receive what we order?
 Number of suppliers, fixed or variable?
Heizer/Render
Asli Sencer Erdem
28
Lead time
 Time elapsed between the order delivery and
order receipt.
 Can be constant or stochastic.
Ex: Lead time is 10 days.
Lead time is between 8-12 days.
Heizer/Render
Asli Sencer Erdem
29
Inventory Models
 Fixed order-quantity models
Help answer the
inventory planning
questions!
Economic order quantity
 Production order quantity
 Quantity discount

 Probabilistic models
 Fixed order-period models
© 1984-1994
T/Maker Co.
Heizer/Render
Asli Sencer Erdem
30
The Economic Order Quantity
EOQ-Model
 Decision variable:
Q = Order Quantity
 Parameters:
S = Fixed cost per order ($/order)
D = Annual number of items demanded (unit/year)
P = Unit cost of procuring an item ($/unit)
I = Annual cost of holding a dollar in inventory ($/$/year)
H= Annual cost of holding a unit item in inventory ($/unit/year)
H=IP
 Objective is to “minimize total annual cost”.
Total
Annual cost
Heizer/Render
Asli Sencer Erdem
=
Ordering
Cost
+
Holding
Cost
+
Procurement
Cost
31
EOQ Inventory Policy
Average Inv. Level
Heizer/Render
Asli Sencer Erdem
32
Assumptions of Classical EOQ Model
 Demand rate is constant or stable.
 There is infinite supply availability.
 Lead time is constant or zero.
 No quantity discounts are made.
 All demand is met on time, no backordering, no stockout.
 Only order (or setup) cost and holding cost
Heizer/Render
Asli Sencer Erdem
33
EOQ Model
How Much to Order?
Annual Cost
Minimum
total cost
Order (Setup) Cost Curve
Optimal
Order Quantity (Q*)
Heizer/Render
Asli Sencer Erdem
Order quantity
34
Costs of EOQ Model
 Total ordering cost is the number of orders times the cost
per order:
D
Annual ordering cost   S
Q
 Total holding cost is the cost per item held 1 year times the
average inventory:
Q
Annual holding cost  H  
2
 The annual procurement cost is the product of annual
demand and unit cost:
Procurement cost = PD
Heizer/Render
Asli Sencer Erdem
35
Annual Cost of EOQ-Model
D
Q
Total annual cost    S  H    PD
2
Q
 Here PD is not a relevant cost and thus dropped.
 Minimize Total Annual Inventory Cost:
D
Q
TC (Q)    S  H  
2
Q
Heizer/Render
Asli Sencer Erdem
36
Optimal Solution of EOQ
 Optimal solution is the economic order quantity
Q* 
2SD
H
 Optimal Total Cost
TC *  2SDH
Heizer/Render
Asli Sencer Erdem
37
Example:The House of Wines and
Liquors
Allex Mullen decides that the first task in utilizing inventory
models is to determine the value of model parameters:
 Annual demand 5200 cases of beer
 $10 telephone charge for ordering
 Purchase cost is $1.5/case beer +shipping cost $0.5/case
 10%bank interest, 5%state franchise tax, 5% theft
insurance rate
How many should he order, how often, and at what annual
relevant inventory cost?
Heizer/Render
Asli Sencer Erdem
38
Solution:
The economic order quantity is
Q* 
2 DS
25200 10 

 509.9 or 510
IP
.202 
 The inventory cycle duration is
T = Q/D = 510/5200 = 0.098 year or 36 days
 The total annual relevant inventory cost is:
 5200 
 510 
TC (510)  
10  .20(2)
  $101.96  102.00  $203.96 / year
510
2




Heizer/Render
Asli Sencer Erdem
39
Robustness of EOQ Model
 EOQ is a robust model with respect to the estimation
errors in D, S, P or I.
 Let Dactual=4 Destimated
Then EOQactual=2 Destimated
Since
2DactualS
2DestimatedS
EOQactual 
2
 2EOQestimated
H
H
Heizer/Render
Asli Sencer Erdem
40
Ex: The House of Wines and Liquors
 Alex Mullen applies EOQ to another product, a
particular variety of Chilean wine that sells 1000
cases annually. The cost is $20 per case. A
telephone call to Chile to place an order costs
$100. The holding costs are the same as for Tres
Equis Beer.
Heizer/Render
Asli Sencer Erdem
41
Ex: The House of Wines and Liquors
Q* 
2 DS
21000100

 223.6 or 224
IP
.2020
T = Q/D = 24/1000 = .224 year or 82 days
 1000 
 224 
TC (224)  
100  .20(20)
  $894.43/ye ar
 224 
 2 
Heizer/Render
Asli Sencer Erdem
42
The Reorder Point ROP
When to give orders?
Q*
Inventory level (units)
Slope = units/day = d
ROP
(Units)
ROP=d*LT
Time (days)
Lead time = LT
Heizer/Render
Asli Sencer Erdem
43
Optimal Inventory Policy
with Backordering
Orders placed during shortages are backordered.
Heizer/Render
Asli Sencer Erdem
44
Optimal Inventory Policy
with Backordering
Imax: Quantity on hand when a shipment arrives.
C: Cost of being one item short for a year
2
2
D
H I max
C Q  I max 
TC (Q, I max )    S 

2Q
2Q
Q
Optimal order quantity and order level
2DS C  H
Q* 
H
C
Heizer/Render
Asli Sencer Erdem
2 DS
C
I max * 
H CH
45
Example:The House of Wines and
Liquors-Backorders
The marketing department tells Alex that beer is a convenience product that can not be
backordered, so sale is lost! However some wine customers are connoisseurs who are
willing to order out-of-stock items. Nevertheless, the store owner will incur some penalty
cost if there is a shortage of wine.
Suppose that retailer suffers lost profit on future business equal to $0.01/unit each day
that a wine is on backorder. What should be the optimal ordering policy if backordering
is allowed?
Solution: The order quantity is computed:
p = $.01×365 = $3.65/unit/year.
Q* 
2 DS
IP
Heizer/Render
Asli Sencer Erdem
C  IP
21000100  3.65  .2020 

 324
C
.2020
3.65
46
Example: Solution
 The max. order level Imax is
I
*
max

2 DS
IP
C

C  IP
21000 100 
3.65
 154
.2020 
3.65  .2020 
 The relevant cost is
.2020154 3.65170
 1000 
TC (324,154)  

 617.82
100 




324
2
324
2
324
 why?
smaller than before,
2
Heizer/Render
Asli Sencer Erdem
2
47
Is backordering better?
 Fewer orders are placed when there is backordering.
 Average inventory level is smaller.
Backorders/cycle= Q* – Imax* =324 – 154 = 170 units/cycle.
Proportion of demand not satisfied on time
=(Q*- Imax* )/Q*=170/324= 52.5%
Service level = 1- Proportion of demand not satisfied on time
= 1-52.5%=47.5%
 The results suggest that:
Retailers will run short in each cycle.
But can they get away with it?
 So backordering must make sense!
Heizer/Render
Asli Sencer Erdem
48
Shortage Penalty Considerations
Shortage Cost
Shortage Cost
C
C
Time
Time
1 year
Theoretical Assumption
Heizer/Render
Asli Sencer Erdem
1 year
Case excluded by the model
49
Imputed Shortage Penalty
An alternative approach for establishing an inventory policy is based on
achieving a desired service level.
Service Level, L is the proportion of demand met on time
I *max
 L, so
*
Q
LQ*  I *max
Imputed shortage penalty
C =
Heizer/Render
Asli Sencer Erdem
HL
1L
50
As C increases EOQ is more
robust
D=1000 units/yr
S=$100/order
P=$20/unit
I= $0.20/$/year
L=47.5%
324
L=90%
Q*
236
EOQ with no
backordering
224
Imax*
212
154
C
$3.65
Heizer/Render
Asli Sencer Erdem
$36
imputed shortage penalty
51
Production Order Quantity Model
 Answers how much to order and when to order
 Allows partial receipt of material

Other EOQ assumptions apply
 Suited for production environment
Material produced, used immediately
 Provides production lot size

 Lower holding cost than EOQ model
Heizer/Render
Asli Sencer Erdem
52
Reasons for Variability in Production
Most variability is caused by waste or by poor
management. Specific causes include:
employees, machines, and suppliers produce units that do
not conform to standards, are late or are not the proper
quantity
 inaccurate engineering drawings or specifications
 production personnel try to produce before
drawings or
specifications are complete
 customer demands are unknown

Heizer/Render
Asli Sencer Erdem
53
Economic Production-Quantity
Model
The EOQ model may be extended to find the optimal production quantity,Q.
p: Daily production rate, d : Daily demand rate
1-d/p
Heizer/Render
Asli Sencer Erdem
55
POQ Model Equations
Maximum inventory level
Setup Cost
Holding Cost
Heizer/Render
Asli Sencer Erdem
=
D
Q
2*D*S
d
H* 1 p
= Q* =
p
Optimal Order Quantity
= Q*
(
1 -
( )
d
p
)
* S
= 0.5 * H * Q
( )
1-
d
p
D = Demand per year
S = Setup cost
H = Holding cost
d = Demand per day
p = Production per day
56
Quantity Discount Model
 Answers how much to order &
when to order
 Allows quantity discounts


Reduced price when item is purchased in larger
quantities
Other EOQ assumptions apply
 Trade-off is between lower price & increased
holding cost
Heizer/Render
Asli Sencer Erdem
57
Quantity Discount Schedule
Discount
Number
1
Discount
Quantity
0 to 999
Discount
(%)
No discount
Discount
Price (P)
$5.00
2
1,000 to 1,999
4
$4.80
3
2,000 and over
5
$4.75
Heizer/Render
Asli Sencer Erdem
58
Quantity Discount – How Much to
Order
Heizer/Render
Asli Sencer Erdem
59
Probabilistic Models
 Answer how much & when to order
 Allow demand to vary
Follows normal distribution
 Other EOQ assumptions apply

 Consider service level, L & safety stock
Service level = L =1 - Probability of stockout
 Higher service level means more safety stock


Heizer/Render
Asli Sencer Erdem
More safety stock means higher ROP
60
Probabilistic Models
When to Order?
Inventory Level
Frequency
Service
Level
P(Stockout)
Optimal
Order
Quantity
SS
X
ROP
Reorder
Point
(ROP)
Safety Stock (SS)
Place
order
Heizer/Render
Asli Sencer Erdem
Lead Time
Receive
order
Time
61
Fixed Period Model
Answers how much to order
Orders placed at fixed intervals
Inventory brought up to target amount
 Amount ordered varies

No continuous inventory count

Possibility of stockout between intervals
Useful when vendors visit routinely

Example: P&G representative calls every 2 weeks
Heizer/Render
Asli Sencer Erdem
62
Inventory Level in a Fixed Period
System
Various amounts (Qi) are ordered at regular time intervals
(p) based on the quantity necessary to bring inventory up
to target maximum
On-Hand Inventory
Target maximum
Q1
Q4
Q2
Q3
p
p
p
Time
Heizer/Render
Asli Sencer Erdem
63
Fixed Period Model
When to Order?
Inventory Level
Period
Heizer/Render
Asli Sencer Erdem
Target maximum
Period
Period
Time
64
Download