AN APPROACH FOR DEVELOPING
COST ESTIMATING
A PRELIMINARY
METHODOLOGY FOR USCG VESSELS
by
MARK JAMES GRAY
BaSC Mech.
Eng., University
of Water 1oo
(1978)
MaSC Mech.
Eng., University
of
(1980)
Water1oo
SUBMITTED TO THE DEPARTMENT OF
OCEAN ENGINEERING
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
MASTER OF SCIENCE
IN
NAVAL ARCHITECTURE AND MARINE ENGINEERING
at the
MASSACHUSETTS
INSTITUTE
OF TECHNOLOGY
June 1987
0 Mark
J.
Gray,
1987
The author hereby grants to M.I.T. and to the U.S. Government
permission tc reproduce and to distribute publicly copies of this
thesis document in whole or in part.
Signature
of Author:
Departmen
f
Ocean Engineering
May 8, 1987
Certified
by:-
Ceifi
--
e d.
Henry
.
S. Marcus
Associate Professor, Marine Systems
Thesis Supervisor
Accepted byl-----___+__
Ae yDouglas
Carmicheal
Chairman, Ocean Engino ring Departmental Committee
Ir
I
AN APPROACH FOR DEVELOPING A PRELIMINARY COST ESTIMATING
METHODOLOGY FOR USCG VESSELS
by
MARK JAMES GRAY
Submitted to the Department of Ocean Engineering
on May 8, 1987 in partial
fulfillment
of the
requirements for the Degree of Master of Science in
Naval Architecture and Marine Engineering
ABSTRACT
A study was done on methods used to estimate new ship
construction costs during the feasibility/preliminary design
phase. Costs were functionally related to SWBS weight groups and
design and fabrication complexities. Other factors affecting
shipyard construction costs include inflation, production
learning and scheduling.
Cost estimating relationships (CERs) are derived from historical
data and are normalized to reflect technical cost trends in ship
construction. The accuracy of the CERs is dependent upon the
quality
and range of the data. Cost risk or uncertainty
can be
quantified provided the probability distributions of the input
data and CER regression coefficients are known.
Five cost models are examined and found to be quite similar in
their approach. Differences can be attributed to the different
needs of the organizations that support their use. The models are
applicable to a wide range of ship types and displacements,
depending upon the data available during their development. Cost
estimates during the feasibility design phase are thought to be
accurate
to within
10-20% of actual costs.
In order for cost analysis to have a measurable impact during
ship design, costing must be an integral part of the decisionmaking process. Before this can occur, management must have
confidence in the costing team's ability to produce high quality
esti mates.
Thesis Supervisors Dr. Henry S. Marcus
Title: Associate Professor of Marine Systems
2
ACKNOWLEDGEMENTS
I would like to express my appreciation to the many people who
assisted me in this research endeavor.
Thanks to my thesis advisor, Professor Henry Marcus, not only for
his guidance and direction, but also for allowing me the freedom
to explore
and learn on my own.
Special thanks to Dennis Clark, Mike Jeffers and Bob Jones of the
Costing and Design Systems Office, Code 1204 at David Taylor
Naval Ship Research and Development Center (DTNSRDC), for their
support and invaluable assistance in the research and for sharing
with me their philisophical understanding of cost engineering in
particular, and life in general.
The assistance
of Jim Herd,
Tzee-Nan
Lo and especially
Ron
Schnepper in the Ships Cost Analysis Division, NCA-2 at the Naval
Center for Cost Analysis was greatly appreciated. Thanks also to
Mike Hammes and Bob Simpson in the Cost Estimating & Analysis
Division, SEA 017 at the Naval Sea Systems Command. These people
gave freely of their time and effort during this study and
supplied invaluable information.
Several individuals in the United States Coast Guard deserve
special mention. Thanks to Lt John Tuttle, Office of Acquisition
and Lt Shawn Smith, Office of Research & Development for their
continuing interest in this work, and to Richard Rounsevell of
the Planning and Estimating Section in the Office of Engineering
for his helpful suggestions.
To all the people who spent the time talking and explaining the
seemingly infinite aspects o cost estimating to me, thank you
all. I hope that I have managed to include all of the useful
insights that were given to me.
I have saved my final note of appreciation and thanks for my wife
Louise, who provided endless encouragement and love throughout
this study.
This thesis work was funded by the Office of Research &
Development of the United States Coast Guard.
TABLE OF CONTENTS
Page
A
MT
rl
w
I
I I
ACKNI
_A
WLEDGEMENTS .
TABLIE
...........
.
OF CONTENTS ....................
LIST OF FIGURES
2
.....
a
.
.
3
.
.
4
..............
.
.
.
.
.................
LIST OF TABLES ...................
.
.
.
.... .....................
8
.................
CHAP'TER 1 - INTRODUCTION
.
...............................
14
.
1. 1
SHIP DESIGN PROCESS ............................
1.2
SWBS WEIGHT GROUPS .............................
1.3
ESTIMATE QUALITY .................
1.4
EARLY SHIP DEFINITION ..........................
.
20
.
24
.
27
...........
30
.
1.5
1I
.
1.4.1
Effect of Design & Construction Standards
34
1.4.2
Incorporating New Technology ............
35
........................... 36
SHIP PROGRAM COSTS ..
1.5.1
Reporting Ship Costs .
...................40
43
CHAPITER 2 - OVERVIEW OF SHIP COST ESTIMATING .............
2.1
ESTIMATING BASIC CONSTRUCTION COSTS ............
47
2.1.1
General ................................
47
2.1.2
Should Cost versus Actual Cost ..........
2.1.3
Learning Rate .....................
2.1.4
Scheduling ......
2.1.5
Inflation and Escalation .................
62
Calculating ROM Escalation ........
68
2.1.5.1
2.2
..
Reliability of CERs ........................
4
.
51
53
........................ 58
CER DEVELOPMENT ................................
2.2. 1
I.
70
74
TABLE OF CONTENTS
Page
2.3
RISK ANALYSIS .....................................
2.3.1
General ....................................
2.3.2
Analytical Methods .........................
82
90
2.3.2.1
Cost Element Probability
Distribution Functions .............. 90
2.3.2.1.1 Beta FDFs .................... 93
2.3.2.1.2 Gaussian PDFs ................ 96
2.3.2.1.
Triangular PDFs .............. 96
2.3.2.1.4 Uniform PDFs ................
99
99
2.3.2.2 Method of Moments ...................
2.3.3
Monte Carlo Methods ........................
103
CHAPTER 3 - DESCRIPTION OF SPECIFIC SHIP COST MODELS ........ 105
3.1
ACQUISITION & LIFE--CYCLECOST ESTIMATING ..........
3.1.1
NAVSEA
3.1. 1.1
3.1.1.2
3.1.1.3
3.1.1.4
107
Code (:017Cost Model .................
General ............................. 1 07
11 C)
Cost Estimating Relationships .......
130
Input Information ...................
Information .................. 132
Output
............................137
NCA Cost Model
3.1.2
3.1.2.1 General .............................
3.1.2.2 Cost Estimating Relationships .......
3.1.2.2.1 One-Digit Model ..............
3.1.2.2.2 Two-Digit Model ..............
3.1.2.3 Input Information ...................
3.1.2.4 Output Information ..................
3.1.3
3.1.3.2
3.1.3.3
3. 1.3.4
3.1.4
137
139
141
146
153
155
156
ASSET Cost Model ................
3. 1.3.1
1C07
General...................
*. . . . .
156
.
Cost Estimating Relationshi ps .......
Input Information .........
*
Output Information ........
. . . .. . .
melm......
161
m..
.......
.
.
.
.
.
.
.
.
.
183
189
USCG Cost Model ............................ 194
3.1.4.1
3.1.4.2
3.1.4.3
3.1.4.4
.............................
General
Cost Estimating Relationships .......
Input Information ...................
Output Information ..................
5
194
196
21 0
213
TABLE OF CONTENTS
Page
3.1.5
3.1.6
RCA PRICE Hardware Model ................... 216
3.1.5.1
General
3.1.5.2
3.1.5.3
3.1.5.4
Cost Estimating Relationships ....... 219
Input Information ................... 233
Output Information ................
234
216
Other Cost Models .......................... 238
3.1.6.1
3.1.6.2
3.1.6.3
3.2
............................
FAST-E ........................
.. 238
239
Merchant Ship Cost Models .
Private Contractor Models ........... 240
COST MODEL COMPARISONS ............................ 241
3.2.1
General
3.2.2
Range of
3.2.3
Model Characteristics
..........
241
pplicability ..................... 243
........ 243
3.2.2.1
Vessel Type and Displacement
3.2.2.2
Ship Design Phase ................... 245
Model Input/Output Comparisons ............. 246
3.2.3.1
Minimum Input Requirements .......... 247
3.2.3.2 Cost Output Sensitivities ........... 252
3.2.3.2.1 Marginal Cost Factors ....... 252
3.2.3. 2.2 Program Cost Factors ......... 254
254
3.2.3.3 Cost Output Accuracy ................
CHAPTER 4 - DEVELOPING A SHIP COST ESTIMATING CAPABILITY ....
4. 1
COST MODEL DEVELOPMENT ...........................
4.1.1
Data Gathering ....
4. 1.1.1
4.2
259
262
264
.........................
Establishing User Needs ............. 267
4.1.2
Resource Requirements ......................
4.1.3
Organizational Considerations .............. 271
COST MODEL SELECTION ..............................
270
274
CHAPTER 5 - CONCLUSIONS AND RECOMMENDATIONS ................. 276
5.1
CONCLUSIONS
.................
6
...............
276
TABLE OF CONTENTS
Page
5.2
RECOMMENDATIONS ..................................
281
5.2.1
General Recommendations ....................
282
5.2.2
USCG Recommendations .......................
284
.................................................
REFER ENCES
AFFEN IDIX A -
SWBS WEIGHT
287
GROUP KEYS FOR COSTED
MILITARY PAYLOAD .............. 292
APPENIDIX B - ASSET K N FACTORS ...............................
AF'PEN IDIX C
294
- COST ESTIMATING SURVEY .......................
304
APPEN DIX D - SWBS GROUP BREAKDOWN FOR THE
..........
309
APPENDIX E - ANVCE SHIP SAMPLE AND SWBS GROUP CERS ..........
320
NCA TWO-DIGIT COST MODEL ...
APPENDIX F - PRICE H INPUT GLOSSARY AND DATA SHEET
7
.......... 332
LIST OF FIGURES
Figure
1.1
No.
Title
F'age
Cost Impact of Decisions
a Design Process
during
16
Cost Savings versus Cumulative
Program Costs
17
1.3
The Ship Design
22
1.4
Exploratory
1.2
Process
Cost Model Levels and
Their Purpose
23
1.5
SWBS Group Descriptions
25
1.6
Typical NAVSEA Ship Acquisition
Time Line
29
1.7
Systems Influence on Ship Design
33
1.8
Representative Life Cycle Cost
Time Line
37
1.9
Inflation Rates
41
2.1
Weight Based Unit Production Prices
49
2.2
Unit and Cumulative Learning Curves
55
2.3
Disruption of the Learning Curve
57
2.4
Schedule Length versus Optimum
Program Cost
59
2.5
Program Cost Comparisons
61
2.6
BLS Shipbuilding Composite Index
63
2.7
Cumulative Manpower Curves
65
2.8
CER Development Methodology
71
2.9
Confidence Interval about a
Regression Line
8C)
2.10
The General Cost Risk
8
nalysis
pproach
83
LIST OF FIGURES
Title
Figure No.
Page
2.11
Uncertainty Measures Using Probability
Density and Cumulative Density Functions
2.12
Cumulative Density Functions for
Dependent and Independent Input Variables
89
The Normalized Beta Probability
Density Function
94
2.14
Shape Variations using Beta PDFs
95
2.15
The Gaussian Probability Density
Function
2.16
The Triangular Probability Density
Function
2.13
98
Additive Moments for Method of Moment
Calculations
1 )2
Decrease in Variance Error with
Increasing Run Size for Monte Carlo
Methods
104
Acquisition Review Milestones, Technical
Definition and Cost Estimate Quality
109
3.2
Categories
111
3.3
Shipbuilding Inflation Factors for 1980
140
3.4
SIW Total Ship Labor Factors
142
3.5
One-Digit Propulsion Systems Material Cost
143
3.6
ASSET Applicability Within the Ship
Design Process
157
3.7
ASSET Computational Modules
158
3.8
ASSET Cost Module Calculative Sequence
162
3.9
ASSET Life Cycle Cost Elements
168
3.10
Basis of USCGMaterial Cost to Non-Cost
2.17
2.18
3.1
of a Total End Cost Estimate
Relationship
for 1980
197
9
LIST OF FIGURES
Figure No.
Title
3.11
PRICE Parametric Modeling Systems
3.12
Typical MCPLXS Factors for Mechanical
F age
217
Assembl ies
222
3. 13
MCFLXS Factors for Selected USN Vessels
224
3.14
Effect of Technology Improvement on
Costs
225
3. 15
Influence of Schedule on Costs
226
3.16
The PRICE Calibration Process
228
3.17
FPRICEH Ship Calibration Data
230
3.18
Frigates & Destroyer MCPLXS vs PSTART
Trend
231
4. 1
Cost Estimating
Functions
for the
Ship Design Process
4.2
261
Factors Affecting the Cost Database and
Cost FPredictions
265
4.3
BehaviouLral Information Processing Model
268
4.4
Point Estimate with Cost Reserve
F'robab
5.
1
273
1i ty
Cost Estimating Building Blocks
1 C)
278
LIST OF TABLES
Title
Tabl e No.
1.1
Example of Increasing
F'age
Level of
Technical Definition
21
1.2
SWBS One-Digit Weight Groups
24
1.
Examples
of Increasing
SWBS Level of
Technical Definition
26
1.4
NAVSEA Ship Cost Estimate Classifications
27
1.5
Common Cost Driver Classifications
30
1.6
Minimum Lead Ship Construction Cost Data
52
1.7
Minimum O&S Cost Data
32
1.8
Navy Frigate Life Cycle Cost Breakdown
38
2. 1
Cost Estimating Methodologies
44
2.2
BLS Material
66
2.3
SWBS Group Escalation Weighting Factors
67
2.4
ROM Escalation Factors
69
2.5
Percentage of Basic Construction Cost
75
2.6
Percentage Deviation for SWBS Group CERs
77
2.7
Percentage Deviation for Total Ship
Construction Costs
78
2.8
Index Breakdown
Relationship between Confidence Internal
and Standard Deviation for a Gaussian
Distri bution
87
2.9
Comparison of FDF Characteristics
91
3.1
Percentages to Calculate Program
Manager Cost Reserve
123
3.2
VAMOSC-Ships Cost Categories
128
3.3
NAVSEA Code 017 Ship Acquisition
Cost Parameters
1 (:)
11
LIST OF TABLES
Page
Title
Table No.
NAVSEA Code 017 Fleet Escalation/Discounted
Cost Parameters
132
3.54
NAVSEA Code 017 SWBS Output Summary
133
3.6
NAVSEA Code 017 P-8 Output
3.7
NCA Cost Model Ship Database
138
3.8
One-Digit SWES Group Cost Drivers
145
3.9
Two-Digit
Cost Model Structure
147
3.10
Two-Digit
Cost Model Cost Drivers
148
3.11
of Lightship
Percentage Distribution
for Two-Digit Model Groups
3.4
Summary
136
Weight
Cost Parameters
151
3.12
NCA Lead Ship Shipyard
3.13
NCA Lead Ship Output
3.14
ASSET Cost Module Ship Database
160
3.15
NAVSEAProgram Cost Percentages
166
3.16
& Support Data for
Operations
Selected USN Ships
174
3.17
Summary
153
155
ASSET USN Ship Construction Escalation
Index>es
182
3.18
ASSET Sailaway Ship Cost Parameters
183
3.19
ASSET Life Cycle Cost Parameters
185
3.20
ASSET Parameter Default Values
187
3.21
ASSET Ship Sailaway Output Summary
189
35.22
ASSET Life Cycle Cost Output
191
3.23
Follow Ship Reduction in Total
Labor Manhours
205
12
LIST
Table
OF TABLES
Title
No.
Page
3,.24
USCGCost Model Categories
208
3.25
USCG Ship Acquisition Cost Parameters
210
3.26
USCG Fleet
Escalation/Discounted
Cost Parameters
212
3.27
USCG Ship Output Summary
213
3.28
PRICE H Cost Output Categories
219
3.29
Fundamental Cost Drivers
in the PRICE H
Model
220
PRICE H Development & Production Cost
Parameters
233
3.31
Description of PRICE H Cost Elements
235
3.32
Relating PRICE and SWBS-based Ship Costs
236
5. 233
Cost Model
244
3.34
Representative Lead Ship Construction
Cost Parameters
248
Representative Multi-Ship Construction
Cost Parameters
249
Representative Life-Cycle Cost
Parameters
251
Naval Surface Combatant Marginal Cost
Factors
253
3.38
Labor/Program Cost Sensitivity
255
3. 39
Material/Program Cost Sensitivity
256
4. 1
Organizational Cost Estimating Functions
260
4.2
Criteria for Parametric Contract Pricing
262
4.3
Cost Model User Features
269
3.30
3.36
3.37
Vessel Applicability
13
CHAPTER
1
INTRODUCTION
For
question;
"How will the
system
answering
one
recently,
due to the impositionof budgetary
managers
agencies,
government
mainly about
ship design managers had to worry
years,
perform?"
within
constraints
must also ask
More
another question;
"Howmuchwill it cost?"
led
This increasing concern with system acquisition costs
the
of design-to-cost
implementation
1970s (e.g., see Ref.
of
to
in the early
(DTC) policies
45). DTC required the early establishment
realistic cost goals and a continuing effort to maintain them
throughout the design and production period.
This type of design
strategy has led to the expanded use of synthesis models for cost
and
on
trade-offs and has put more emphasis
performance
early
definition of the complete ship system.
Synthesis models (e.g., the US Navy's
possible
merit,
to
any particular
qLuantify
and assess the technical
the
ship's systems.
how
well -the
effectiveness).
capability
and cost impacts
The figure of merit
ship accomplishes
its
make it
ASSET program)
or
figur-e of
of its effect on
provides a measure of
mission
(e.g.,
military
The iterative nature of trade-off studies
allows a design to evolve which maximizes
synthesis
models
minimizes
for a cost figure of merit) the figure of merit
upon
the ship's design
and subject to
funding constraints.
14
technical,
schedule
using
(or
based
and
The
importance
of establishing a credible cost
estimating
capability early in the design phase cannot be overlooked. Figure
1.1
gives
a
qualitative
indication
of
the
cost
impact
decisions
throughout
important
to note the major impact of decisions in the
(i.e.,
a
generic
feasibility/conceptual)
acquisition
phase
(e.g.,
process.
see
of
It
is
planning
Ref.
3).
The
trend indicates that the maximum leverage for costs occurs during
the
early
design phases.
Therefore,
the ability to perform
large number of performance/system versus cost trade-offs
this
early
a
during
design stage will greatly assist in minimizing
cost
uncertainty during later design phases.
The
advantages of incorporating
performance/system
versus
cost trade-offs for the USN's DDG 51 is documented in Ref.
5. The
approach
was described as a closed loop feedback control process
involving
the
established
design
review
budget,
of
and
the
design
for
conformance
then making decisions
to
to
change
an
the
in the areas where the budget constraints were exceeded.
Cost
analysis
during
early
design
is
in
the
enviable
position of applying maximum effect for minimal cost (relative to
the
total
versus
program
result
The inverse effect
for
cost
cumulative program cost is illustrated in Fig.
relationship
advanced
cost).
is particularly applicable to the
technologies
from
the
development program.
into ship design.
introduction
of
The
impact
1.2.
introduction
greatest
innovations
early
The
of
savings
in
the
Attempts to incorporate innovation at later
phases, when the design is committed to alternative technologies,
can become prohibitively expensive.
15
t4
COST
IMPACT
OF
DECISION
NUMBER
OF
DECISION
t4
I
PHASE
PLANNINGOTOOLINO
MANUFACIURINQ
O(SIGN HiASE
-
--- TIME
Cost Impact of Decisions during a Design Process
(Ref. 60)
Figure 1.1
16
CL
COST
($)
Cost Savings versus Cumulative Program Costs
(Ref. 15)
Figure 1.2
17
Having established that potential cost savings are a primary
motivation for early ship design definition,
timely
and
stage
credible estimation of ship costs during this
help to:
would
operating costs;
cost
drivers,
"reduce ship acquisition and
and;
cycle
improve the naval engineering awareness
The purpose of this study is
and;
life
early
analyze cost drivers and technologies impact on
cost early in the ship design process" (Ref.
techniques
it follows that the
of
26).
twofold;
to
document the
used to cost ships during these early design
to address the major areas involved with the
phases,
establishment
of cost estimating capabilities. Although these topics are listed
sequentially,
between
there
costing
is
a
high
degree
of
techniques and the needs and
interrelationship
resources
of
an
organization that influence its costing requirements.
The
first
step
reference
for
that
used,
are
each level,
in this study is to describe
the different levels of cost
the technical
detail
a
frame
estimating
of
quality
that is commensurate
with
and where in the ship design process these different
estimate qualities can be found.
Ship
costing
overlapping
management,
is
is
aspects
a
multi-faceted
of
discipl ine,
engineering,
economics,
involving
business
statistics and human resources. The aim of Chapter 2
to provide a general understanding of the ship costing
and
its
terminology.
relationships
and
their
The
development
ncertainty,
of
cost
crucial to
field
estimating
early
stage
estimating, is discussed in somedetail.
A description
found
in Chapter
of specific
3.
cost estimating
The methodologies
18
techniques can
be
from five models used
by
various US Navy agencies and the US Coast Guard are included. The
is
presented
information
of
amount
upon
dependent
the
documentation available in the open literature or volunteered.
fairly qualitative comparison of the models was
A
by
proprietary nature of the material.
the
models
estimating
cost
cost
a
as
level.
the
chapters have concentrated on
previous
The
ship
Current
typically relate ship weights to
of technology
function
dictated
technical
aspects of cost estimating. Chapter 4 examines the management and
human
resources
factors
must be
that
considered
before
the
technical cost estimating capability can be integrated within the
organizational
a
entity,
operating
cost
analysis
environment.
As
with
group must receive
corporate
any
support
all
from
levels of management if it is to contribute effectively.
The
chapter
final
recommendations
of
presents
the
Any discussion of
the study.
and
conclusions
must
osting
address the effects of the following areas on the credibility
of
the estimates received. These are: (1) the experience of the cost
engineering
estimating
group;
(2)
algorithms,
the
and ;
development of
appropriate
cost-
(3) the range and applicability
of
the database.
The all-encompassing recommendation for implementing a
any
within
organization
estimating
capability
maintaining
a long-term commitment to its
in
the
areas of personnel levels,
collection functions.
19
is
operation;
that
cost
of
primarily
training standards and
data
SHIP DESIGN PROCESS
1.1
There are three principal divisions in the NAVSEA (Naval Sea
Systems Command) ship design process;
(1) exploratory design
(2)
acquisition design
(3)
service life design
this
Since
estimating,
cost
are
phases
of
study
with
stage
early
the exploratory and
only
The
concern.
studies,
feasibility
deals
four
preliminary
design
acquisition
are
phases
acquisition
design,
ship
new
contract design
and
detail design.
Fig.
Sollows,
the
1.3
(e.g.,
that
the
The
design.
the Navy or Coast Gard)
dotted
feasibility design phase,
for
order
the
design
process
from
starting from a statement of mission requirements
customer
detail
indicates
this study.
box
in
Fig.
1.3
with
ending
and
encloses
the
which is the phase of primary interest
Note the overlap with both exploratory
studies
and preliminary design.
these
Typically,
phases
can be differentiated
other by the increase in technical definition of the ship
a
reduction
progresses
in the
from
technical
uncertainty)
exploratory through to detail
each
from
(i.e.,
as the design
design.
At
any
given stage in the ship's design, all of the ship systems will be
defined
to the same level of detail.
Table 1.1 illustrates
increase in technical definition for a propulsion plant.
20
the
Level
Technical Definition
0
Whole Ship
I
Propulsion Plant
2
Propulsion Units
3
Gas Turbines
4
Engine Starter
System
Engine Starter
5
Example of Increasing Level of Technical Definition
(Ref.
26)
Table
1.1
In an R&Denvironment, the technical
to
a level commensurate with detail
remain in
levels
the
definition
can increase
design and yet the ship will
the exploratory studies phase. Fig.
1.4
links
the
in'Table 1.1 with cost modelpurposefor ship designs in
exploratory
definition
is
phase.
For example,
required for
a high degree of technical
costing
detailed
technology
assessments.
Fig.
1.4 also
illustrates
the relationships
number of systems examined (breadth),
between the
the number of parts
added together for each system (depth), and the relative
21
being
cost. If
-
b
-
MISSION
REQUIREMENTS
I
__
-
i
I
_
EXPLORATORY
STUDIES
--
I
I
I
I
I
I
I
- I1I
II
I
I
-
IL
FEASIBILITY
STUDIES
!
i
HI
I
i
I
II
-
II
CONTRACT
I
DESIGN
I
DETAIL
DESIGN
The Ship Design Process
(Ref. 60)
Figure 1.3
22
MODEL
LEVEL
SWBS
LEVEL
COST MODEL PURPOSE
*0
WHOLE SHIP
EAR LIEST-STAGE, CONCEPT DEVELOPMENT;
PROGRAM COSTS, SCHEDULING EFFECTS
1
ONE-DIGIT
(GROUP)
PROGRAM MANAGERS, RESOURCE
APPLICATION AND APPORTIONING
TWO-DIGIT
I
2
(SUB-GROUP)
3
THREE-DIGIT
(ELEMENT)
DESIGN ANALYSIS, TECHNOLOGY ASSESSMENT,
INTEGRATION ANALYSIS, ALTERNATIVE
SYSTEMS ANALYSIS
* ·
BREADTH
. . ,am
~
-'--
'
D
E
P
LE'
T
ASSE
H
c
LEVEL
I
ASSESSMENT
LEVEL II
<$10K
ASSESSMENT
<$100K
Exploratory Cost Model Levels and Their Purpose
(Ref. 26)
Figure 1.4
23
the
definition is at the whole ship level,
one
part
needs
inexpensive.
propulsion,
parts
As
to
be considered
electrical,
that each system i
(and the cost of analysis)
1.2
the
analysis
cost
systems increases
number o
the
and
only one system with
hull,
(e.g.
etcr.) the breadth decreases, and as the
broken down into increases
the
depth
increases.
SWBS WEIGHT GROUPS
The
ship
common
Ship Work Breakdown Structure (SWBS) provides a
means of communicating the level of technical definition
the
is
designer,
shipyard
and
cost
estimator.
between
The
major
elements of the SWBS system that are of interest for ship costing
are listed in Table 1.2.
Examples of some of the items that make
Description
SWBSGroup
100
Hul 1 Structure
200
Propulsion Plant
300
Electric Plant
400
Command
500
Auxiliary Systems
600
Outfit & Furnishings
700
Armament
800
Design & Engineering Services
900
Construction Services
Surveillance
Weight Groups
SWBSOne-Digit
Table 1.2
24
Definition
Groop Oescriptlon
I
mull structure
Shell plating or planklng;lonIqtudnl ad transversefraingqlInnerbttu plating; platform ard flats belao
lowrist continus deck; fourth dieck;third deck;secoddeck; maindeckor harar dck forecastle de (irclrdlng
platform, flats, and deckSbetween
min aridgallery deck);gallery decl flight deck;ladin platform ad pecial
purpoe decks Drve weatherdeck (includes catapult traugh)l superstructucas foldationsr for min propelling
dchinerylfo.vdationsfor auxiliarles andother ilpmsnt!structural blUade trunksad enclosures;tructural
spnsors; aror; aircraft fuel saddletankstructure; structuralcosting, forgings,and quivalentwldents; s
chests;ballast nd buoyancyunits; doors ndclosures, qscial purpose;
doos, hatches, maholes, rd scuttles
nronballistic;doors htches, mnoles, aridscuttles, ballistic; ests ad king posts;omprtmnt tsting.
2
Propusion
rs ad air eectorsl shafting, bearilngs,
ad propellers;
Boilers ad energyconverters; propulsionunits; maincondea
cibustion air supplysystem;uptakesandMke pipes;propulsionontrol qulpment;mainstem systma feeduater
md
cwxensate systm; circulating and coolingwater ystem;fuel oil servicesysta; lue oil systma.
3
Electric plant
Electric pr
geration; per distrlbution switchbords; p
distribution ad fixtures.
4
Cwsuicatlon
aId control
ad
armanrt mtrol systemscountermeasure
Kavigatioralwytamard equllpent; interior onwications syti;
ships' protective syta (exceptelectronic) electronic yrstm includingelectronic ounterasurms.
distrlbtion systms(cable)j lighting syst -
S
allilary
sytm
Heatingsystmrmventilation system;air-conditioningsyste; refrigerating pas, plant and rqulpnts gas, HWA,
cargopipinq, orgen-itrogen, aviation lub oil systemsplumbingIrstallatics; firmin, flushing.ad prilrer
system;fire etingulshing system;dralnage,trlming, heeling,and bllast syst ; freshwtersystm sppr
ad deckdrains; fuel arnddesel oil fillingr, venting, tog, aid trarsfer ystm; t
heatingsyste;
premaed-air systms auillary stem exmustst am,ad ta drainsi boyancy trol Wystm(floodingandventingurlrinesl); lcellanemaspiping systemidistilling plat; steeringgearsystnm;ruddcr vindes, capstns cranea,
aid acor-andlinq ystm; elevastors oving suirways, andcargao
hadlirq
imL t; operatinggearfor retractable
elevatingunitsl aircraft elevators; aircraft arrestinggearberrilrs madbarrice
ctapults andjet blast defloctors, hydrofoilsad lift system.
6
Outfit d
furnishnq
Hill fittings; bots, bott sto,
aid hadllrng; rigqgingd carv ; ladders d gratings, nnstructural bulkhad
for
andnonstructul doors; painting; deckcovering;hull Insulatlon;storeron, stoaages, ad lockers;equipment
utility paes; equilpent for rkshops;equipent for galley, pantry, cullery, ad cmlssry outfit; furnishings
for living s ; furnishingsfor offloe spaces, electronic, ad radar; furnishinqgfor medicalad dentalqpaces.
7
Armmt
Guns,uwnts, ad lar
s
Designad
Designo d erglrertingq srvice.
_gineering
hddl Ing.
nlrqdevlc;
ma iintlon-handllngsystem amuniton stoaqe; pcial weapnsoage aad
services
9
Construction
service
Staging,scaffolding,andcritbling lanchin; trals
handlingandrval;
cleairg ship services.
d dckirng tporary utilitle
~~_
...........
SWBSGroup Descriptions
(Ref. 54)
Figure 1.5
25
_
andservices; materLal
make
up
each
of one-digit
group
groups
can
be
found in Fig.
700 is
100 through
1.5. The summation
equal to the weight
of the
whole ship less load items.
The
SWBS
classification
system
specified at any of three levels;
Each
higher
definition,
The
level
as
can
three-digit
definition.
definition
indicates
Fig.
be
SWBS
1.4
allows
one-,
a
the
two-,
higher
degree
shows
the
the
SWBS
to
be
and three-digit.
seen from the examples
level represents
ship
of
in
Table
highest
levels
technical
of
1.3.
level
of
technical
as they apply to costing during the exploratory phase
of ship design.
SWBS Level
Breakdown
Technical Description
Whole Ship
1-Digit Weight
Hull Structure - Group 100
Electric Plant - Group 300
2-Digit Weight
Hull Decks - Group 130
Lighting Systems - Gr-oup330
3-Digit Weight
Second Deck - Group 152
Lighting Fixtures - Group
32
Examples of Increasing SWBS Level of Technical Definition
Table 1.3
All
of the ship costing techniques presented in this
use the SWBS weight groups as the means to classify weights.
26
study
1.3
ESTIMATE QUALITY
Estimate
quality
is related to a variety of
majority of which are programmatic in nature
factors,
(i.e.,
the
acquisition
strategy plans). In this section, the estimate quality as related
to
technical definition only is discussed.
NAVSEA uses a cost estimate classification system which uses
letters
the
of
increasing
alphabet
level of design
to
designate
definition
estimate
(i.e.,
quality.
decreasing
uncertainty) these are ROM (Rough Order of Magnitude),
D,
C.
Table
1.4
shows
the SWBS level of technical
In
level of
Class
F,
definition
appropriate for each estimate classification.
Estimate
Classification
SWBS
Technical
Definition
NAVSEA
Cost Phase
ROM
Less than
Feasibility Study
Planning
F
Feasibility Study
1-Digit Weights
Planning/
Programming
D
Preliminary Design
2-/ 3- Digit Weights
Programming
(maybe Budget)
C
End Preliminary Design
3-Digit Weights
Budget
NAVSEA Ship Cost Estimate Classifications
(Ref.
16)
Table 1.4
27
the
around
is concerned
study
This
estimate.
feasibility
The
is
estimate
in
with ship cost estimating
corresponding
phase,
a
to
one-digit SWBS group.
Therefore
class
wei gr; -or
ar, approximate
listed in Table
The
1.2 and Fig.
overlap
SWBS
one-digit
each of the 100-700
will
groups
SWBS
1.5.
means
that occurs between design phases
weights are available in the
exploratory (i.e., conceptual) studies,
latter
that
stages
of
all through feasibility
design.
stages of preliminary
and on into the initial
studies,
of
primary
the
technical input to the estimator for this degree of quality
be
F
Class
technical level of definition for this
the
and
Of
course, one-digit SWBS weights can be calculated by simply adding
up
become
weights of higher level components whenever they
the
available.
The NAVSEA cost phases referred to in Table 1.4 are
into
the
estimates
planning,
are
programming
generated
1.6
shows
budget
in both the planning
how
phases.
Fig.
picture"
of ship acquisition.
is
and
these phases fit
In Fig.
phase.
and
divided
Class
programming
into
the
to supply guidance to the various Navy agencies
in the acquisition process.
28
"big
1.6, the initial estimate
a ROM estimate and the POM (Program Objective Memorandum)
issued
F
is
involved
)(
Complete
Contract Plans
Contract
and Specif1cat1ons Award
Class "C"
Congressional i .: .. B 'asic'::.'"
IContractors
budqet
ruction:: Submtlt
*'Const
e Date::
Bids
:.'::
Bas,
Initial
Estimate
.".....
:.:::/
i
I
Progr amm1in
....
:'7"[
Budgeting
Start
of
Constructl1on
.. · · ('
Planning
l
'.*:
POM
8 Months
Launch
Delivery
~~~~~~~~~~~~~~~I
2 to 5 Yedrs
-
I .<<-{
I-
Approxlmdtely
2 Years
I
11 to 18 Months
I
-
-I
Typical NAVSEAShip Acquisition Time Line
(Ref. 16)
Figure 1.6
29
I
!
1.4
EARLY SHIP DEFINITION
In
order to estimate acquisition costs
estimator
requires
categories
1.5
a -ertain amount of information
which relate to the ship's
definition,
realistically,
in
configuration,
several
technical
and the design and fabricating specifications. Table
lists and describes these
categories as they are
in this study.
The inputs for each of the cost models
in
are
Chapter
the
3
classified
according
to
the
addressed
discussed
cost
driver
categories listed in Table 1.5.
Description
Cost Driver
Technology
technology available, stateof-the-art design influences
Ship Design
weights, margins, design
standards, component selection
Manufacturing
fabrication techniques, degree
of automation, learning
Programmatic
type of contract, procurement
strategy
Economic
escalation, inflation &
interest rates, discounting
Operations & Support
maintenance, personnel, fuel,
spares
Common Cost Driver Classifications
(Ref. 19)
Table 1.5
30
As
of
a means for providing
the
the necessary
information
for each
cost driver categories in Table 1.5, various
models
have
provide
a
been developed.
Synthesis models are
large number of feasible ship
consistently,
synthesis
intended
designs,
quickly
and
in accordance with a set of design specifications.
Detailed discussions of synthesis models for Coast Guard
and
patrol boats can be found in Refs.
The
US
Navy
CONcept
to
introduced the CONFORM
FORMulation)
7 and
8,
cutters
respectively.
(slrface ship
continuing
to provide consistent feasibility
designs
and cost estimates (e.g., see Ref. 46).
Fig.
1.7
illustrates
the
systems
development
of a feasible ship design.
within
constraints
the
manpower
levels,
imposed by
maintenance
and
that
influence
The design must
the
R&D
supply
the
operate
community,
naval
capabilities,
and
manufacturing practises.
An indication of the minimum amount of information
to
estimate
1.6.
The
GFM
exclusively
comprise
the lead ship construction
of
cost is given
(Government Furnished Material)
those
items of ordnance
the combat system.
and
required
in
Table
consists
almost
electronics
that
Follow ship costs can be related to
lead ship -costs through the application of learning theory.
A
similar
list for O&S (Operating and
Support)
costs
is
given in Table 1.7. This list represents the information required
by
the
USCG's
unpublished
C
CASHWHARS
patrol
financial
analysis
program.
boat feasibility studies have
Recent
used
ASSET synthesis model as a front-end to the CASHWHARS program.
the
Ship Type
Hul11 material and Fabrication methods
Listing of major GFNI items
Seven SWBS Group Weights
Propulsion Plant and SHP
Electrical Plant Capacity and Number
Crew Size
Special Equipments
(e.g., active fin stabilizers,
etc.)
Minimum Lead Ship Construction Cost Data
(Ref. 6)
Table 1.6
Interest & Inflation Rates
Fuel Costs
Vessel Maintenance Schedules
Operating Hours
Mission Profile
Crewing
Vessel Life
Minimum O&S Cost Data
(Ref. 34)
Table
1.7
_1··-·
y--
--
/
NIXATA
Systems Influence on ShiS Design
(Ref. 47)
Figure 1.7
33
1.4.1
Effect of Design & Construction Standards
Kehoe et al (Ref. 49) looked at the effect
country
design
practices on the size
cost of a typical antisubmarine
of different
and total ship
warfare
(ASW) frigate.
investment cost included the basic shipbuilders
NATO
investment
The
cost,
GFM,
total
and
other related costs (e.g., standard allowances for change orders,
electronics
escalation,
and weapons cost growth and cost growth
for future characteristics changes).
The
seven
major
design
listed
practises
below
were
examined;
(1) design & construction margins
(2)
hull forms
(3)
design displacement
(4)
sustainability
(5)
survi vabi 1 i ty
(6)
in-service margins
(7)
habitability
Sustainability
and design margins were found to be the most
contributing factors for increased ship size and cost.
Ship
construction
significantly
example,
the
higher
use
costs
can
for different
of
military
also
be
specification
instead
specifications will result in increased costs.
to
be
levels.
For
expected
of
commercial
1.4.2
Incrporating
Typically,
New Technology
uncertain cost predictions are not the
greatest
hurdle to be overcome when considering the introduction of a
technology (Ref.
political,
influences.
22).
Far greater impact can be expected
new
with
national security, environmental, energy and societal
However,
only
the cost issue is addressed in
this
secti on.
When judging whether the use of a new material or technology
to be incorporated into a ship design is cost effective, the cost
analysis
should
be based on its
LCC,
including
R&D,
design,
manufacture and O&S. The high degree of uncertainty expected with
such an analysis must be quantifiable to allow for risk trade-off
studies. Cost risk analysis is discussed in Chapter 2.
For
shipyard
applications,
the decision to implement
new
technologies is primarily based on the economic benefits expected
from
predicted
productivity and
efficiency
gains.
Since
the
services and skills of the shipbuilder only account for about 20%
of
the
total
productivity
amount
ship
from
program
new
of the total costs
cost,
major
technologies affect a
(Ref. 24).
55
gains
in
shipyard
relatively
small
1.5
SHIP PROGRAM COSTS
Program
costs are often referred to as the life cycle
(LCC) for the particular system.
into
three
areasi
investment,
and;
(1)
LCCs are generally categorized
research
operations
cost
and
and
development
support
(O&S).
(R&D);
(2)
Fig.
1.8
illustrates a representative time line of LCC expenditures.
Note
that there is generally some degree of overlap between each
cost
category.
R&D
costs
technology
are
for the work associated
making
items or innovation design features available to
ship at key commitment points.
associated
with
high
the
There is often developmental work
with combat system installations.
R&D costs are
not
O&S costs for conventional vessels are
the
specifically addressed in this study.
Investment
and
major components of LCCs (Ref. 36). Ship investment costs include
construction,
government furnished material
(GFM), outfitting,
post delivery and other miscellaneous costs.
The term investment
costs is used interchangeably with acquisition costs.
Ship
O&S costs include direct/indirect military
direct/indirect
personnel,
ship operations and maintenance and direct
36
ship
H-INVESTMENT
.. ___
COST OPERATING AND
ISUPPORT
PROGRAM
PROGRAM
PECULIAR
COSTS
R&D COSTS
--
· ··
0
·
.
-
TIME
Representative Life Cycle Cost Time Line
(Ref. 60)
Figure 1.8
37
COST
,I
moderization
costs.
development,
frigate.
Table
acquisition
Based
on
a
30
1.8
and
presents
O&S
LCCs
year ship life,
Category
the
breakdown
for a
360
ton
the
O&S
costs
Percentage Breakdown
Operating & Support (*)
Indirect Support
2
Maintenance
10
Moderni zati on
18
Operations
15
Personnel
12
57
Acqui sition
Combat systems
17
Machinery
18
Hu 1l
4
39
4
Development
(*) based
on a 30 year ship life
Navy Frigate Life Cycle Cost Breakdown
(Ref. 36)
Table
1.8
of
Navy
are
significantly greater. A lower percentage for acquisition costs
be
can
reduced.
R&D US
if the combat capabililty of
expected
The
following
vessels
Management
vessel
is
percentage breakdown for the FFG-7 isi
1.61 Investment - 35.2, and; O&S - 63.2.
Navy
the
are recorded by
of O&S Costs) program,
the
VAMOSC
The O&S costs for
(Visibility and
a brief description of
which
can be found in Section 3.1.1.2.
There
design
are
a number of things that can be done during
to reduce the main elements of the O&S costs and
the
decrease
total LCC
(e.g.,
see
Ref.
36).
ship
thereby
However,
these
improvements generally result in a higher acquisition cost (Refs.
5 and
50).
Although,
design
the
process,
decisions.
There
importance of LCCs are not overlooked in the
acquisition
are
costs tend to dominate the
three main reasons
for
the
influence of acquisition costs over LCCs (see Ref.
is
design
controlling
5): (1) there
a more immediate impact for the acquisition'costs;
(2) there
is considerably less uncertainty associated with acquisition cost
estimates,
andl
(3) there is greater flexibility in the
future
funding of O&S costs.
The
in
emphasis on minimum procurement cost has been reflected
fixed displacement and hull length constraints or some
design
criterion
dimensions
which
constrains
(e.g., see Refs.
hull
47 and 48).
.9
size
and
other
principal
1.
1
.ReportingShip
Observed
Costs
in ship costs,
differences
if
can
explainable,
reflect differences in the subsystems of the ships as a
function
of technology changes, inflation, or productivity differences. If
and
remaining
differences
major
the
data,
the
should reflect the technology
level
and
productivity
inflation
characteristics
of
are backed out
the ship
of
subsystems,
thus
enabling
analysts to establish specific cost trends in the data.
in
Variations
productivity
a
are very much
function
location
stability in the shipyard's workload and its geographic
(Ref. 2).
is
often difficult to remove
the
productivity
that account for different levels of shipyard
shifts
(see
It
.1.2).
section
However,
the effects
efficiency
of inflation
removed by expressing ship costs in so-called base year
Other
dollar terms used are then year dollars,
dollar
amount for
a previous fiscal year (FY),
of
can
be
dollars.
referring to the
and futre
year
dollars for future costs.
Cost f igures
required
can be converted
inflation
rates
variation
in inflation
inflation
rates
consistently
for
to any base year provided
are available.
rates
Fig.
1.9
for several years.
Department of Defense
(DoD)
shows
the
Note that
equipment
economy.
higher than that for the general
40
the
is
_
_
~
..
_
_
ii
,
.
DoD- All
US
Economy
__
~~~~~~~
|~~
__
_
_
__
l
_
Equipment
~~
~~~
_
_
_
_
_
_
-
9.3
9.8
10.6
1976
5.2
8.1
9.0
1977
5.8
10.2
1978
7.4
9.4
1979
8.5
1980
9.0
8.1
10.6
13.2
10.4
11.4
10.3
11.0
__I
Source:
10.9
9.4
_
~~~~
_
i
_
US Department of Commerce, Bureau of
Economic Analysis
Inflation Rates
(Ref. 17)
Figure 1.9
41
~
Aircraft
1975
1981
-
-
For
the calculation of total LCCs,
discounted
technique
O&S costs are generally
on a year by year basis using the net
(e.g., see Refs.
present
1 and 34). If, for simplicity,
costs are assumed equal to the same amount, A (base
value
the O&S
year $),
each
year throughout the vessel 's life, then the net present value for
the O&S costs, NPV (base year $),
NPV
A ( (+i)/(l+d)
is given by the expression;
+ (
(l+i)/(+d) )2 +
(1.
+ ( (l+i)/(l+d) )L
...
where
1)
)
i = yearly escalation rate
d yearly discount rate
L = ship life in years
The escalation
at
rate is used to equate a given sum of
the present time (i.e.,
money
base year) with another sum of money
at any future time.
Discounting
inflation,
to the
to
is
the reverse of the
compounding
effect
of
in that the discount rate movesfrom the future back
present.
To explain
in other terms, escalation is required
account for inflationary trends which decrease the
present
day
because
of
money in the future,
the
whereas discounting
time value of money from
expected
value
is
returns
of
used
on
investment. Obviously, the two effects are opposite. The discount
rate is often taken as 10% (e.g., Refs.
42
and 34).
CHAPTER 2
OVERVIEW OF SHIP COST ESTIMATING
Cost
and
estimating is concerned with all aspects of production
economics that influence the development,
operation
(including
construction
portion
retirement)
of any
of
government
a
construction
ship.
agency's
Within
and
the
shipbuilding
budget, there are three primary elements of cost:
(1)
costs incurred for management of the acquisition
program
The
(2)
costs incurred for the procurement of GFM
(3)
shipyard costs for construction of vessels
costs
associated
with
the
management
of
the
ship
acquisition are programmatic in nature. These include the effects
of
inflation,
profit
and
other economic factors
orders and various cost contingency margins.
For instance, a
two
change in total program costs after programmatic effects
ncluded
This
estimating
construction
(e.g.,
chapter
the
that
see Ref.
deals
5).
with
the
traditional
costs involved with GFM purchases
services,
(BCC). Direct
funds
costs
dollar change in production costs translates to about a
dollar
are
change
Programmatic
significantly impact on the total program costs.
one
plus
labor
methods
and
shipyard
referred to as basic construction
costs
and material BCCs make up only 25% of
must be appropriated
for each new
the
(Ref.
23).
Table 2.1 lists the various methodologies generally used to
come
up with cost estimates in this area.
43
ship
of
Delphi
Analogy
Engineering
Cost Estimating Relationships (CER)
Parametric
Cost Estimating Methodologies
Table 2.1
The Delphi
technique
of
experience
individuals
iterative
one,
involved.
The
working
from
approach for this method
the
towards a consensus among
method
Delphi
upon
rely
estimates
and often vary widely
best
the
Therefore,
another.
Delphi
suLch,
As
opinion.
expert
as the solicitation
can be described
is frequently
used to
of
the
one
to
is
an
persons
extrapolate
outside of the range of existing cost data.
Analogy
the comparison
of
new
existing
with
This technique requires
historical
on similar existing systems for it to be feasible.
Analogy
are the most economical to produce of all the
methods
estimates
listed
on
to determine costs.
systems
data
rlies
in Table 2.1.
The
engineering
estimating
and
method
is
referred
is the "accepted" method of
as
to
obtaining
bottoms-up
reliable
estimates (e.g., budget quality). This technique is characterized
by
separate
material
and
labor
44
costing
on
a
component
by
component
structure
the
typically
basis,
(e.g.,
the SWBS groups
aggregated (e.g.,
work
for Navy ships).
definition available,
technical
highly
using some form of
breakdown
Depending
component can
upon
range
whole ship) to very detailed
from
(5-digit
SWBS). Bottoms-up estimating is generally the most time consuming
and costly of all the methodologies
NAVSEA
The
Code
017
for the same level of detail.
model described in
Chapter
3
uses
the
traditional engineering approach for cost estimating.
The
previous
relationships,
such
as
method can be described as using
whereby costs
labor hours.
estimating,
cost-to-cost
are estimated from elements of cost
Parametric and CERs
use
cost-to-noncost
which involves the use of inputs such as performance
(e.g., SHP), weight, etc. to arrive at a system cost. Parametric
and
CERs
are derived using multi-variable and
single
variable
regression techniques, respectively, on existing data. Therefore,
these methodologies represent averaged system cost trends and are
called
top-down
economical
estimates.
Parametric
and CERs are
the
most
techniques to use as they give quick answers to
what
if questions, provided these changes can be properly reflected in
the noncost variable(s).
CERs can only be used to predict costs if the new ships have
systems similar to those in past ships.
models
found
in
construction costs.
as
Chapter
3
use
The NCA,
CERs
to
USCG and ASSET
calculate
basic
Increased system definition can be expressed
a separate CER whenever there is sufficient cost data at that
level to define a unique trend. The more design features that can
be accounted for by using different trends,
there is in the analysis.
45
the more flexibility
Parametric
detailed
models
generally
estimates (i.e.,
serve as a
cross
using the engineering
method).
have proven useful for evaluating future technology
and can influence
long term priorities.
check
for
They
assessments,
The RCA PRICE model is a
well-known parametric model.
Cost-to-noncost
into
for
computerized models.
estimates
Ref.
relationships
6).
to
is
requirements,
an
Although
be
one
effects
win
time
see
high
approval
Low estimates can result from unreal
difficulty
in
guestimating
unknown
and/or over-optimism
manhours).
can
the general estimating philosophy has always been
conservatism
(Ref.
51),
which serves
to
minimize
the
of unknowns that always arise as ship design (technical)
definition
database
the
the factors that contribute to a low estimate
important,
of
(e.g.,
important factor given
variables or predicting technology impacts,
(e.g., underestimating
incorporated
of the results
accept low estimates as a means to
against a competing system.
performance
typically
This leads to quick turn around
as well as consistency
Consistency
motivation
are
improves.
is
thin,
For
there
unknowns conservatively,
advanced naval
is a
natural
vehicles,
tendency
where
to
the
estimate
resulting in higher estimated costs and
risks. (Ref. 46). As the concept nears an acquisition commitment,
the
high
cost
alternative
estimates may eliminate
from competition.
an
otherwise
In a recent SWATH
feasible
design,
hull
structure estimates reflected a tremendous difference in cost per
46
pound
compared
to
conventional
monohulls,
even
though
the
material and fabrication specifications were identical (e.g., see
Ref. 46).
2.1
ESTIMATING BASIC CONSTRUCTION COSTS
2.1.1
General
Basic
for ship construction,
and
material
profit.
is defined as
construction
Note
contracting,
the contract award
including all production labor, overhead,
costs plus an amount for cost of money
the use of the word price in this
price
price
(COM) and
definition.
In
indicates dollar amounts inclusive of fee or
profit. If the fee or profit is excluded, the dollars amounts are
referred to as costs.
An
the
estimate
of
CC is developed
two major parameters of ship
costs
and labor hours.
separate
manufacturing
aggregate
rate;
by utilizing
construction
estimates
costs,
material
Direct labor costs are calculated
or engineering rates or
overhead
is estimated
an
for
using
appropriate
as a percentage
of
the
direct labor costs.
Overhead
costs
costs represent that portion of fixed and variable
allocated to the production of each ship but not
related
to its construction.
directly
Examples of fixed overhead include
rent, insurance, depreciation of buildings and equipment, and the
cost of money.
Variable overhead costs change with the
level
shipyard and include
in
the
taxes,
employee
activity
benefits,
communication and travel costs, and production related expenses.
During
the
feasibility phase of the ship
design
basic construction material and direct labor costs are
47
process,
estimated
using two primary approaches:
(1) wherever ship definition permits, vendor quotes are
solicited for the equipment involved;
definition
ship
wherever
(2)
sufficient
not
is
to
support a vendor quote, material and labor costs are estimated by
a CER.
of
form
either in the
CERs reflect historical ship cost data,
similar
return costs or cost proposals/pricing exercises for
Labor hour CERs are based on a review of CPRs (contractor
ships.
,
reports)
performance
historical
proposals,
costs,
return
previous estimates and shipyard experience.
relate non-cost design parameters to costs. Typically,
CERs
in
costs,
new ship cost estimating,
($),
functionally
are
(see Table 1.2), so thatl
related to SWBS weights
C = f
C
(2.1)
(weight)
represents some functional
Fig.
relationship.
where
f( )
shows
weight based cost estimates in terms of dollars per
for
a
wide range of products.
will
products
depending
vary
for
The cost per pound
upon
pound
similar
differences
their
2.1
in
complexity.
Other
costs,
systems
variables
besides weight may correlate
depending upon the application.
often
use
shaft
better
For example,
horsepower (SHP)
as
an
with
propulsion
independent
variable for estimating costs.
Material
estimates
for hull structure
48
include
a
scaling
Illl
--
*
^
9
-~~~_
lOOM
.S
I
100.000_
10.000-
Z
1000-
0-
100-
1010.10.010.0010.0001-
I
0.0001
I
I
0.001 0.01
I
0.1
I
1
I
10
I
100
---I
1000
----I
SELLINGPRICEOF UNITS (S/lb)
Weight Based Unit Production Prices
(Ref. 19)
Figure 2.1
49
----I
`-I
10.000 100000 1M
-- I -- '
10M 100M
factor
which accounts for the impact of
offs,
mill
tolerance,
weld
rod
and
scrap,
cut-outs,
transportation.
cut-
Typical
factors add an additional 15% to weight based quotes.
Ships
can have the same weight in a functional area
hull structure) and yet have different costs.
(e.g.,
Cost variations of
this type are due to differences in materials, fabricating, level
of
technology,
design
plus a variety of other factors.
process,
explain
Early
in
these factors can be incorporated into CERs
general cost trends.
However,
as design and
definition increases throughout the design process,
the
to
technical
the top-down
approach inherent in the use of CERs becomes less tenable.
Design definition supports the solicitation of vendor quotes
for
most
areas.
of
the high value components
Whenever
within
the
functional
actual equipment costs and weights are obtained
from quotes they can be incorporated into the cost estimates
for
ship construction costs. This is done by reducing the appropriate
weight
the
groups by the equipment weights available,
weight
group costs using the
relationships
(CERs),
and
appropriate
recalculating
cost
then adding the equipment
estimating
costs
to
these updated values. This method of shopping list additions will
tend
to
increase the accuracy of any cost estimates
data
is
obtained
The
since
the
directly from the manufacturer.
use of actual equipment
data is especially
helpful
for
determining accurate electronics and armament hardware costs. The
principal
their
reason for this is that their costs depend mainly upon
performance and sophistication,
quantified as weight,
characteristic.
As
space,
a
which are not
as
easily
or any other physical or technical
result,
it
50C
is difficult
to
develop
a
functional
relationship
such as Eqn.
(2.1) to explain historical
cost trends in electronics and armament.
For
Navy
.
systems)
In 1980,
on software costs.
dollars
the
effects
the DoD spent
However,
software
nearly
billion
$3
projected amounts for 1990
are in excess of $32 billion dollars (e.g.,
There
of
on costs is becoming more important (especially for
development
combat
acquisition programs,
see Ref. 17).
are several models in use for estimating
the
design
and support costs associated with software development. The names
of
some
of
the
(COnstructive
these models
2.1.2
more well-known
SLIM,
COst MOdel),
are
and PRICE-S.
can be found in Refs. 17, 18 and
Jensen,
COCOMO
Information
on
19.
Should Cost versus Actual Cost
When dealing with GFM quotes,
the
models
distinction
costs
it is important to understand
between should costs and actual
costs.
are based on time-and-motion studies of various
working
under
mean
generally
easily
conditions.
controlled
accessible
tradesmen
conditions
Controlled
that all tools and materials are
Should
available
are
few
generally supply should cost information
for
throughout
the work and that there
and
interruptions.
Manufacturers
major items and ancillary
with
these
tendency
vendor
for
hardware costs. Care must be exercised
quotes,
over-optimism
since there
is
and bias on the
an
understandable
part
of
vendors,
resulting in low cost estimates for the particular item(s).
In
materials
contrast,
must
be
under
actual working conditions,
shuttled between the
workplace
tools
and
and
supply
depot(s) and there are continual disruptions due to a variety
of
reasons. The result of these differences is that actual costs can
be expected to exceed should costs by 15 to 20 percent (Ref.
The effects of modular construction and other
considerations
on
investigated.
However,
this
difference have yet
it
would
seem
to
2).
producibility
be
adequately
logical
that
any
producibility improvement would reduce the difference between the
two.
Should
described
should
cost
have
different
above for vendor quotes.
cost
identify
can
interpretations
The US Navy may
study of a sole-source contractor's
inefficiences
reasonable
cost
eliminated.
The
to
in
the
results
initiate
facilities
the operation and determine
government
would
be
of the study form the
than
if
basis
to
what
these
a
a
were
for
the
government's position during contract negotiations.
Finally,
independent
a
should
review
team of analysts.
cost
study
can
also
refer
of a proposed program budget estimate
to
by
an
a
Their purpose is to establish a range of costs
that can be expected
for the program.
2.1.3
Learnin.gRate
as well as many other industries, experiences
Shipbuilding,
a
learning or improvement process when multiple units are
constructed
in
orderly
phased sequence.
The
accumulated
by
industry verify that learning takes place
being
historical
data
(Ref.
These empirical data provide the basis for what is referred
43).
to as learning curve theory.
is that each time the total quantity
theory
The
ships
of
built doubles, the manhours, material, or basic construction cost
of
the ships is reduced
by a constant
percentage
of the previous
manhours, material, or basic construction cost.
a
For
learning rate,
given
the cost
N
CSI N
Nt h
ship,
(2.2)
R / log 2)
C 1 = lead ship cost
where
the
relationship;
CN (S), can be calculated from the following
CN = C1 N(log
of
(.2)
($)
R = the fractional learning rate (%/100)
Eqn.
(2.2)
is
relationship.
curve
referred to as the unit learning or
Since
each
point
on
a
Boeing
curve
learning
a theoretical individual unit cost as a percentage of
represents
the lead ship production cost,
the area under this curve up to a
given quantity equals the total production cost for that amount.
Estimators have a choice of working with unit or
average
explained
manhours
cumulative
curves.
The
difference in these two approaches can
using
the
following example.
If
the
eighth
were 90 percent of the fourth ship manhours,
other
hand,
ship
then
learning would be expressed as a 90% unit learning curve.
be
the
On the
if the average manhours of all eight ships were
90
percent of the average manhours of the first four ships, then the
learning would be expressed as
learning
a 90% cumulative average
curve.
Fig.
2.2
illustrates
that a unit learning rate of 94%
to a cumulative learning rate of 97% .
equivalent
are
used correctly,
for
follow
the
rates
total
ship
have
been
learning rates of about 90%
Unit
cost.
acquisition
Provided they
the choice of unit versus cumulative
calculations does not affect
ship
is
noted for Navy destroyers and frigates (e.g., see Refs.
25
and
30).
The
of
factors,
level
skill
on
occurs
that
a variety
on
depends
schedule
learning
of
amount
any
including
required),
(i.e.,
technology,
construction time and time between starts.
skill
level
jobs
exhibiting
rates close to unity,
learning
in
reduction
Conversely,
systems
have
learning
there is little or
that is,
with subsequent
labor
minimal
system
manuf acturi ng
complexity
Low
production
performance
of
no
task.
the
requiring highly skilled tradesmen can show
significant learning effects,
so that there is a sharp reduction
of labor between the lead and follow ship.
less
learning that takes place between the first and lst
(e.g.,
see
result
outfitting
shipyard
Ref.
efficiency
20).
Innovations
like
robotics
in lower production costs from
and consistency of
learning effects.
54
the
is,
more highly automated the fabrication process
The
and
ships
zone
increases
application,
not
in
from
97 of
Lead Ship
Cost
SHIP
CONSTRUCTION
COST
AVERAGE COST
94 of
UNIT COST
Lead Ship
Cost
--
LEAD
FOLLOW
SHIP
SHIP
Unit and Cumulative Learning Curves
(Ref.
1)
Figure 2.2
55
Learning
phased
their
of
theory
sequence.
is defined for production
in
an
BIW (Bath Iron Works) reached a high point
in
production learning during a time period in which a number
Navy destroyers were built in a relatively short
(Ref.
orderly
time
frame
23).
When production occurs in multiple lots, too slowly, or with
too
great
a time gap between construction starts,
costs do not follow Eqn.
lapse
in production,
follow
ship
(2.2). Fig. 2.3 illustrates that for a
the shipyard experiences a jump
upon resumption of building.
in
costs
Break
in
Production
I
I
Recurring
Unit
I
Production
Cost
I
I
I
I
I
I
I
I
I
.1
I
I
Cumulative Quantity
Disruption of the earning Curve
(Ref. 17)
Figure 2.3
57
Scheduling
2.1.4
construction
The
highly
dependent
construction.
costs
upon
the
for a ship or group
procurement
of
ships
for
chosen
strategy
is
This strategy indicates the number of vessels, the
length of each construction period, the time between construction
and the number of lots that the
starts
in.
From
this partial listing,
vessels are to
it can be seen that
be
built
production
learning and scheduling effects are interrelated.
Pre-outfitting, modular and zone construction, increased use
of automation and other fabrication improvements are resulting in
construction times.
shorter
reduction
thereby
manhours,
reducing
offset by increased engineering and support efforts
being
as
production
costs.
current indications are that the production savings are
However,
16).
of
the
The effect of these changes is
(Ref.
An overall drop in total construction costs can be expected
these advanced techniques become more fully
the
fabrication
process.
The impact of producibility gains
reduced because shipyard costs only affect a
further
into
integrated
is
relatively
small amount of the total program budget (Ref. 24).
Fig.
2.4
illustrates
that
an
optimum
results in the lowest production costs,
(i.e.,
schedule
overtime,
increases in costs.
results
in
project
schedule
while compression of the
more shifts) results in
significant
Although stretching out of the schedule also
increased costs due to inflation and reduced
learning
U
A
IL
O
z
0 ,
0
z
UNRELIABLE
SCHEDULE (PERCENT OF TYPICAL) I
Schedule Length versus Optimm Program Cost
(Ref. 61)
Figure 2.4
59
effects,
these are generally less severe in magnitude.
also
an
optimum phasing of design and
when
construction
construction
starts immediately after design is
There is
schedules,
complete.
Overlapping of these schedules will result in increased costs, as
shown in Fig. 3.15.
Variations
can
in expenditures and government funding
significantly
affect
program
costs.
Fig.
2.5
patterns
compares
multiple and single lot production costs using the RCA PRICE cost
model
(see section
In
3.1.5).
the feasibility design phase,
there exists a great deal
of uncertainty regarding production scheduling.
Lowest costs can
be expected for an "optimum" construction period combined with
single lot procurement strategy.
60:
a
II
LOT 2
-
LOT 6
5 UNITS
LOT 6
LOT 5
4 UNITS
LOT 5
80
PROGRAM
COST%
60
LOT 4
4 UNITS
LOT 2
21 UNITS
LOT 4
1 LOT
LOT 3
4 UNITS
LOT 3
LOT 2
4 UNITS
LOT 2
DEVELOPMENT PLUS
PRODUCTION OF
29 UNITS
40
20
LOT 1
LOT 1
LOT
DEVELOPMENT PLUS
8 PRODUCTION UNITS
_
PRICE H
ESTIMATE
_
CURRENT
ESTIMATES
_
_
Program Cost Cmparisons
(Ref.
26)
Figure 2.5
61
_
_.....
PRICEH
ESTIMATE
_
A_
.
_
A_
J~
.
IJDo
P
rnLIC "
ESTIMATE
2.1.5
Inflation and Escalation
The
rate
inflation,
technical
of
change
in
economic
conditions,
aspects of a cost model
decrease
will
(Ref. 23). Inflation
or
is defined
alternatively,
of the amount of product a fixed amount
of
money
buy relative to some base year.
It
wage
is
important
inflation,
determined
by
a
to realize
each
that when looking at
product price or industry
unique set of cost and market
regional labor rate differences),
consequently
price
or
rate
is
wage
factors
(e.g.,
and their inflation rates will
differ from case to case.
Using forecasted
prices
that are not truly representative of a specific product can
to serious miscalculations of future
if
an estimate of the average
year
as
can affect cost estimates more substantially than the
as the price increase of a product over time,
the
such
lead
cost estimates. For example,
annual inflation is low by 3%
for a system to be manufactured seven years from
per
now, the
result is a cost overrun of 23 percent.
The
indices
visible
measures
of inflation are the
that are often quoted.
vari ous
These indices range from
price
highly
aggregate indicators such as the wholesale price index (WPI) down
to
disaggregate
commodities.
there
the
At
producer
higher
indices
(PPI)
levels of aggregation
is less fluctuation
disaggregate
price
levels.
for
(e.g.,
specific
the
WPI),
in the indices than would be found
Fig.
2.6
shows the
percent
at
yearly
variation in the BLS (Bureau of Labor Statistics) composite index
used
by
the US Navy for
estimating
62
contract
escalation.
The
4
10
(U
12 -
(50/
zw
(C)
I.li
4
_
1
n
1w
a
I
1966
1970
-. rL.... 9
I
1975
FISCAL YEAR
Ccaposite Index
(Ref. 62)
BIS Shipbuilding
Figure 2.6
63
1980
I
1985
fluctuations
can
technology,
be caused by a number
inventory,
competition,
of
and
factors,
including
the overall effect of
inflation on the economy.
Escalation
accounts for the change in cost expected between
contract award and ship delivery due to inflation. Currently, the
USN
has extended the
the
delivery date.
and
procedures
escalation coverage out to 8
Outyear pricing defines
inflation
factors
for
differences
past
estimating
general
developing
estimates with base dates in the future.
the
months
ship
costs
The failure to identify
among relative escalation rates
for
important
labor, material and energy cost inputs can lead to serious errors
in forecasting program costs.
Escalation
building
53).
amounts
are
computed
based
on
a
specified
period and an assumed manhour profile (e.g.,
Figure
see
2.7 shows a cumulative manpower curve for
representative
6
Ref.
ships
of a 36 month award to delivery period assuming a
6 month interval between construction starts. Total labor amounts
for this example are in the vicinity of 91,000 manmonths.
The use of more aggregate material price and earning indexes
can
lead
to
estimates.
escalation,
energy
errors
in
material
and
labor
cost
escalation
To provide a best estimate of a project's future cost
a
inputs
composite of the project's
labor,
material
should be constructed to the greatest
detail practicable.
64
degree
and
of
aM
4W
4S@
30
2O
oeo
S
6
12
18
24
38
36
MONTHS
Cumulative ManpowerCurves
(Ref. 62)
Figure 2.7
65
42
48
54
60
66
Historically,
costs
inflate
in Section
material
composite
material
Table
and
equipment
for
domestic
materials
1.5.1. Because
index illustrated
in
of this, the
the 50/50 labor
calculates its own escalation rates:
government
example.
shipbuilding
at different rates than
as mentioned
applications,
and
naval
2.6
Fig.
2.2 indicates the PPIs and weightings
is
that
used to calculate a whole ship BLS material index.
PP I
Description
10-1
11-4
11-7
Iron and Steel
General Purpose Machinery
Electrical
Machinery
and Equipment
BLS Material Index Breakdown
Table 2.2
66
Weighting
.45
.40
.15
an
are
NAVSEA
also develops escalation indices for the
SWBS weight
groups. Table 2.3 shows the weightings that are used to develop a
single ship weighted average for a non-combatant surface ship.
SWBS Group
o10
Description
Weighting
Hul 1 Structure
.092
200
Propulsion Plant
.118
500
Electric Plant
.093
400
Command & Surveillance
.015
500
Auxiliary Systems
.281
600
Outfit & Furnishings
.354
700
Armament
.005
800
Integration/Engineering
.026
900
Assembly & Support Services
.016
SWBS Group Escalation Weighting Factors
Table 2.3
67
importance cf
The
effect
on
and
escalation during the construction process cannot
overlooked.
accounted
program
accurately predicting inflation
The
for)
cost growth from unexpected inflation
may
result in decreased
stretchouts.
reduction
More
drastic
production
measures
may
its
be
(if not
rates
and
entai1
the
in units procured or even cancellation of the program.
In any event, these actions lead to higher per unit costs, deployment
shortfalls and reduced capabilities.
2. 1.5. 1
Calculating ROM Escalation
The
for
methodology
estimating
outyear
(i.e.,
ROM
and information contained in this
(Rough Order of Magnitude)
non-budget year) pricing
Memorandum dated 3 April
017
that
are
Table
The
escalation
is taken from a
escalation
an
contract award in the last quarter
fiscal year and post delivery,
in
months
of
the
8 months after the delivery date.
The cost for each ship is the sum of 100%of direct
overhead and 100%of material
NAVSEA
given
These percentages are based on the number of
assumed
for
percentages
multiplied against the cost of the ship are
2.4.
between
1986.
section
dollars
and cost
labor, 95%of
of
money (see
Section 3.1.1).
Other
escalation
assumptions
used in the determination
factors include:
of
the
base date is 8 months prior to
ROM
the
award date; start of construction 12 months after the award date;
50/50labor/material composite (e.g.,
see Fig.
ship construction expenditure curves (e.g,
escalation
for
2.6);
specified
see Fig. 2.7), and;
energy and fringe benefits growth. For
awarded after FY 90,
the factors for FY 90 are to used.
68
ships
Post Delivery - Award Date
(months)
FY
Award
60
72
84
96
9.4
10.8
12.1
13.5
14.8
6.8
8.0
9.3
10.6
11.9
13.2
5.8
7.0
8.3
9.6
10.9
12.2
24
36
48
6.4
8.0
89/07
5.5
90/07
4.6
Date
88/07
(*)
(*) year/month
ROM Percentage Escalation Factors
Table 2.4
The following example is given for a FY 88 shipl
(1)
(2)
(3)
(4)
(5)
(6)
Base date
Award date
Start of construction
Post delivery date
Target cost (000's)
Delivery - Award date
(7)
Escalation
= (5) x escalation
87/11
88/07
89/07
92/06
$250,000
48
factor
= $250 M x 9.4%
= $24 M
The escalation
or the FY 88 $250 M ship is $24 M.
69
Changes
adjusted
the
in
base date of 8 months prior to
For changes
0.4% for each month.
by
award
in the
start
construction from 12 months after award are adjusted by
each month.
For the following
example
of
0.1% for
of a FY 89 ship;
(1)
(2)
Base date
Award date
88/06
88/11
(3)
Start of construction
90/01
(4)
(5)
(6)
(7)
Post delivery date
Target cost (000's)
Delivery - Award date
Adjustment for Base date
88/11 to 88/06, change
93/10
$271,000
60
of -3 months
-3
(8)
are
x .4%
= -1.2%
Adjustment for Start of construction
88/11 to 90/01,
change of +2 months
+2
(9)
(5) x
Escalation
x .1% = +0.2%
(escalation + adjustments)
= $271 M x (9.3 -1.2 + 0.2)
= $23 M
for the FY 89 $271 M ship is $23 M.
The escalation
2.2
CER DEVELOPMENT
This
section
describes a methodology for calculating
estimating relationships (CERs) from given cost
illustrates
collection
data,
cost
the
six
steps
is the first step.
involved
in
the
data.
Fig.
procedure.
Data is obtained from
returns and ship operational information.
sources that have the data do not want to give it up,
cost
2.8
Data
historical
Often the
so that it
must be paid for in some manner (e.g., a contract to assemble the
relevant data).
In addition,
the data gathered is not always in
the form that it is needed for CER development. For example, most
shipyards
their
do not group costs according to SWBS weight group
internal accounting
procedures.
7o0
for
Therefore, a significant
CERDevelopnent Methodology
(Ref. 63)
Figure 2.8
71
portion
of the development work and expenses can be involved
in
data collection (Ref. 2).
The
second step of the process is primarily concerned
normalizing
the
converting
data.
The most widely
used
the costs to a common base year
.
with
normalization
Previous
is
mention
has been made for removing producibility effects from the data as
(e.g., see Refs. 9 and 23).
well
Once the data is in a usable format, the factors in the ship
design,
construction
and operation which may be thought to have
an effect on costs must be determined.
based
It
Variables are also chosen
upon their easy availability within the design
community.
was found that SWBS weight groups and SHP for propulsion
are
the factors used most in ship costing.
Having
costs,
the
relationships
selected
the
analyst
must
independent variables
decide
on
one
or
to
relate
more
with which to statistically curvefit the
to
rational
data.
A
rational CER means not just an equation that "fits" the data, but
one
which
has some basis in engineering.
For the
cost
models
reviewed in Chapter 3, three types of CERs are used;
(1) linear (e.g., NAVSEA Code 017 and NCA)
Cost = A + Bx
(2) log -
(2.3)
log (e.g., USCG)
log Cost = A + B log x
72
(2.4)
(3) log - linear (e.g.,
Cost
and B are unique
A
where
(2.5)
to be
and
analysis
for
determined
is
x
or
operational cost variable
or
construction
= A x
constants
regression
from
equation
ASSET)
design,
the
chosen.
element
These equations do not include the RCA PRICE model,
which uses a
multiple-variable regression relationship to
proprietary
each
relate
cost to non-cost variables (i.e., parametric analysis).
Although
any data can be curvefit,
the
instance,
uncertainty)
cost
cost
the
element
can
Experience
For
least
(i.e.,
variable giving the least variance
the
tests.
of fit" and other significance
for "goodness
curves
it remains to test
when compared to the data may be selected
used in the CER
also
(e.g.,
see
prove to be invaluable
2.2.1).
Section
in
the
as
selection
process.
a
As
result
is
of
these
tests
selected as the
and
most
a
rationalizations,
appropriate
for
the
particular
CER
situation.
It is important to realize that the CER represents an
historical trend to costs and therefore it is strictly applicable
only within the range for which the data exists.
these
This means that
algorithms should be used to predict costs of newer
having systems similar to those found in past ships.
73
ships
2.2.1
Reliability of CERs
There
are two areas in which errors in CER development
can
be found: measurement and stochastic errors. Measurement error is
not
addressed in this study.
This section discusses some of the
methods available for determining the probabilistic error
goodness
of
highly
fit)
dependent
available
for CERs.
upon
Confidence
in any derived
the quality and quantity
of
CER
cost
and on whether the derived CER can accurately
observable
cost trends.
However,
(i.e.,
is
data
reflect
if large departures are
made
from the database designs, errors will increase in magnitude.
Generally,
available,
for
the
any
better
statistical curvefit,
the fit,
provided the
the
more
variable
data
chosen
exhibits an explainable trend. Smaller subsystems are found to be
more
specifically
design
volume,
power
or
groups
(Ref.
23).
variables than are more general
example,
can
related to weight,
relationships for whole ship cost to ship
For
displacement
be expected to have a higher variance than those
individual SWBS group weight costs.
other
Because of this,
found
for
whole ship
costs are usually estimated as a summation of SWBS group costs.
Although
it
would
be desirable to reduce
error for every SWBS group CER,
of
the
dollars.
Section
costs.
material
typically,
the
estimating
only a small portion
system elements account for the majority
of
investment
Table 2.5 shows the SWBS weight group and subgroup (see
3. 1.2)
These
cost percentages of
construction
the
shipyard
costs include all direct
construction
labor
and
costs except for armament (SWBS 700) and communications
and control (SWBS 400), which are for installation only.
74
Percentages
Group
SWBS
Subgroup
Subgroup
Group
1A
11.67
C
2.33
2.33
2.17
D
1
2A
14.50
B
2.67
1.67
C
D
18.33
3.33
2
3.50
B
7.67
4A
2.67
22.33
11.17
4.67
Ec
7.00
4
7.33
5A
B
12. 00
1.00
1.33
C
D
5
21.33
6A
1. 33
5.
B
C
D
E
00
7.33
2.17
2.33
6
3.67
7
7
100.00
Percentage of Basic Construction Cost
(Ref. 23)
Table 2.5
16.00
3.67
100.00
The sum of the top ten of the subgroups in Table 2.5 account
for
75%
of the total
Therefore,
CER
for
shipyard
maximum
development
effort
construction
costs
(Ref.
the
majority
reduction of errors,
should be directed
at
2).
of
minimizing
the
uncertainty of the high cost driver categories.
The
main
propulsion,
SWSS
group cost drivers found in Table
auxiliary equipment and hull.
2.5
are
The ANVCE study found
that the relative importance of each of these three groups was
function of the particular ship design.
a
For example, the complex
lift system in hydrofoils made the auxiliaries group the dominant
cost
element,
while for advanced monohulls the propulsion group
was the highest percentage cost (Ref. 54).
Cost
studies
are
driver
impacts are particularly useful for
and are often expressed as marginal cost
usually
interfaced
generated
with
cost
using
synthesis
model CERs.
The
expressed in terms of dollars per ton.
analyses
In
can be found
order
to
developed
for
percentage
of
values
were
trade-off
factors.
programs
factors
which
are
They
are
typically
Examples of marginal cost
in Refs. 5, 13, 35 and 42.
evaluate the "goodness of fit"
the ANVCE study (see Appendix
deviation
calculated
of the estimated
for each
76
SWBS
of
E),
versus
group.
The
the
the
CERs
average
actual
cost
percentage
l
i
1
i
i
i
Percentage of Deviation from Data Bass
Upper Limit Lower LimiEt~
vere
Cost Group
-14.3
1.
Hull structure
8.2
2.
Propulsion
6.4
·26.0
-15.0
3.
Electric plant
11.1
28.7
-19.4
4.
Comnunication
13.0
37.9
-24.5
9.5
39.0
-22.6
and control
30.0
5.
Aux i liary systems
6.
Outfit and
furnishings
14.1
43.0
-23.1
7.
Armament
21.2
83.5
-27.3
8.
Design and
engineering
services
10.9
36.4
-22.0
14.4
54.3
-30.5
12.1
42.1
-22.1
9. Construction
services
Average Total Dev.
.
.
.
.
Percentage Deviation for SBS Group CERs
(Ref.
54)
Table 2.6
77
Ship
Deviatioan from -Ship
Class
Data Base (%) Class
Deviat on from
Data Base (%
PF 109
SWATH DE
5.8
6.3
PGH 1
PGH 2*
6.2
-37.4
SWATH MCM
DLGN 25
DLGN 35
DLGN 36
3.0
-5.1
-5.8
-4.2
PCH 1
PHM 1*
AGEH 1
DBH
8.5
-26.8
22.3
5.5
DLGN
-0.1
ARCTIC
13.9
SEV
JEFF A*
-28.1
JEFF B*
2KSES (B)
2KSES (R)
-30.6
-15.3
- 9.3
38
CPIC
PG 92
PG 92
PGG 1
PGG 1
(P)
(T)
(P)
(T)
Snip sample
data base.
-15.5
8.7
6.5
2.9
32.2
with actual returned costs in
Percentage Deviation for Total Ship Construction Costs
(Ref. 54)
Table 2.7
78
deviation for one point can be written as;
I Cestimated Cactual I / Cactual x 10%
FD
where PFD= the percentage deviation
of a single
(2.6)
data point
Cestimated = the cost calculated from the appropriate
SWBS CER
= the corresponding cost data point
Cactal
The
absolute
estimates
are
percentage
study.
value
weighted
deviations
Armament
signs ensure that the over- and
evenly.
shows
the
for the SWBS group CERs from
(SWBS 700)
shows
Although there are wide deviations
average for
Table 2.6
the
within
each group is muchless.
greatest
underaverage
the
ANVCE
uncertainty.
any one cost group, the
This further
illustrates
that the CERsrepresent overall trends and are not intended to be
representative of any one ship design.
Deviations of estimated total ship construction costs (i.e.,
total
labor
database
due
to
plus
material)
from actual
are shown in Table 2.7.
the
addition
of change
costs
for
ANVCE
The large under-estimates
orders
and
other
production costs in the return cost data which are not
for in the SWBS group CERs.
the
are
unforeseen
accounted
It is important to remember that the
deviations shown in Tables 2.6 and 2.7 indicate how well the CERs
can
reproduce the database,
not how successfully the CERs
will
predict the costs of ships not in the database.
Variance
Fig.
2.9
is often
used as a measure
of uncertainty
for CERs.
illustrates that the calculated variance can define
79
a
/
/
.0/
Y
11
/
Z/000,
w
I
_I
!
I
I_
I
I
__ I_
_ I
_I .
____
x
Confidence Interval
about a Regression Line
(Ref. 55)
Figure 2.9
80
-
__
l_
-
probability
allows
of fit
distribution
about
the CER regression
line
certain statistical tests to be carried out for
(e.g.,
see Refs.
19,
28 and 56). Most analysts
which
goodness
assume a
Gaussian or normal distribution (see Section 2.35.2.1.2)with zero
mean
about
range.
Ref.
the line and constant variance throughout
28
discusses
the
variance on cost risk analysis.
81
implications
of
the
data
non-constant
RISK ANALYSIS
2.3
2.3.1
General
in
Uncertainty
the ship design process has typically
as low,
medium or
The sum of these uncertainties is usually
accounted
for example,
described in qualitative terms,
high
risk.
been
for by incorporating various construction margins (Ref. 28). Risk
analysis attempts to quantify the uncertainties that are inherent
in
the
select
any predictive method in such a way that the user can
level and type of risk to be accepted and provide a means of
to
elements
the individual contribution of various
identifying
the overall risk.
In
simple terms,
schedule
are
(uncertainty)
can be divided
into
Included in these
cost, schedule or technical.
three categories:
categories
risk
the
changes,
effects
of
inflation,
technologies,
new
shipyard productivity,
plus others.
In this
section, some of the techniques that have evolved to estimate the
cost risks in shipbuilding will be examined.
An
addressed
important
area
in
risk
that
the uncertainty associated with the cost
is
of the estimator
will go a long way
The
experience
the
uncertainty in the system cost.
important
cost
terms of
Experience is
to
is
not
analyst.
reducing
particularly
in areas where costs associated with technical changes
must be extrapolated from existing data.
The general
In
the first
probabilistic
cost risk methodology
step,
input estimates
distribution,
as
is indicated
in terms of
are e:pressed
opposed
to
in Fig. 2.10.
a
a
traditional
deterministic (i.e., point) value. These inputs include the cost
II
--
Weight
WBS Structure
With Cost
Factors Or
Radr Renge
.
_
W
CERs Each A
IA
Probability
Distribution
All Categories
Summed
P(C)
Analytically
or by
Monte Carlo
Cost
Dollr
OUTPUT
INPUTS
* The Distribution of
Possible Costs as a CDF
* Physical or Performance
Parameters for CFqs
* Direct Cost Estimates
The General Cost Risk Analysis Approach
(Ref.
32)
Figure 2.10
83
of GFM,
performance characteristics, weight, and other technical
required in most cost estimating methodologies.
definitions
a
example,
cost
about
analyst may know something
the
For
lowest
most likely value, and upper bound of an input or perhaps
bound,
the mean and variance of its distribution.
second step in a general risk analysis is to
The
input
non-cost
the
variables
functions
distribution
probabi ity
separate
to
(PDF)
transform
subsystem
cost
the
CERs
using
developed. The PDF shows the probability of occurrence of a given
costs
estimate.
It
is
important to note that since
based on regression analysis,
are
they introduce an
the
CERs
additional
uncertainty into the process.
step is the summation of the separate
The final
PDFs,
subsystem
using either analytical or simulation (i.e., Monte Carlo)
to obtain the total PDF for the ship. However, the most
methods,
useful presentation of cost uncertainty indicates the probability
of
exceeding
a given sum
(generally the most
This is the type of distribution
the risk analysis
of
output
likely
shown in Fig.
process;
2.11 as the final
it is referred
probability distribution (CDF) and is the
cumulative
estimate).
to as
result
the
of
integrating the area under the PDF.
2.15
Fig.
and Eqns.
(2.11-12) illustrate the relationship
between the PDF ( p(x) or frequency ) and the CDF ( P()
and pessimistic values are the best estimates of
optimistic
lower
and upper
Note
the
The
).
bounds, respectively,
characteristic
effect
S-shape
the
for the cost of the system.
of the
of defining systems more
84
The
CDF.
fully
and
therefore
MOST LIKELY
PROBABILITY
DENSITY
(FREQUENCY)
I
I
I
OPTIMISTIC
PESSIMISTIC
COST
,,,
1uu
90
PROBABILITY
THAT COST
WILL NOT BE
EXCEEDED
10
0
OPTIMISTIC
PESSIMISTIC
COST
Density
Uncertainty Measures Using Probability
and Cumulative Density Functions
(Ref. 19)
Figure 2.11
85
obtaining
a
more
reduction
in
the
interesting
to
accurate cost estimate would be seen
width
note
of costs under
that,
according
the
to
S
by
curve.
Eqn.
It
with
zero
uncertainty would reduce to
is
(2.12),
probability of any point estimate being correct is zero.
system
the
a
the
Only
single
a
point
estimate.
In this study, risk is defined as the uncertainty measure of
a
ship costing estimate.
uncertainty
to be found
the variance (see Eqn.
Although there are several measures of
in the literature
(e.g.,
see
Ref.
18),
2.14) is probably the most familiar.
The
variance
is often used with the mean to define a coefficient
variance
(COV), which is simply the square root of the variance
divided
by the mean value.
Since
COV
of
the COV increases in relation to uncertainty,
values should be expected during conceptual and
design,
and smaller values for contract design.
the uncertaintyof weight using
existence
of
this
concept,
constant COVsfor classes of
larger
preliminary
Ref.
29 treats
and suggests
items
(i.e.,
the
hull,
propulsion, etc.).
Both material and labor contribute to the uncertainty
estimate.
However, the uncertainties
are usuallymuch more significant.
be
positively skewed (i.e.,
there
is
a physical
given task.
produces
of any
associated with labor costs
Individual labor costs tend to
skewed right in Fig.
limit on the minimum
time to
2.14),
since
accomplish
a
At the total job level, the:summation of these tasks
a close approximation to the normal distribution
(Ref.
29).
Cost
interval
estimates
are
often
used
to
epress
uncertainty
where
in
a point estimate:
is the most likely
X
for example,
estimate
value.
$X + or
- 10%,
such
However,
a
statement can be misleading without a proper statistical frame of
reference.
For
a
randomly distributed variable,
the level
of
provide
confidence or variance associated with the estimate will
the appropriate frame.
The level of confidence,
or confidence interval,
gives the
probability of the variable actually falling within the specified
limits
expressed
(i.e.,
$X + or -
10%) at any given time and is typically
as a percentage.
For a
normal distribution,
shows the variation of confidence levels with number of
Table 2.8
standard
deviations (square root of the variance) about the mean (also the
most likely value for a Gaussian distribution due to symmetry).
Number of Standard
Confidence Interval
(%)
Deviations
about the Mean
38.30
). 5
68.26
1.0
86.64
1.5
95.44
2.0
99.74
3.0
Relationship between Confidence Interval and
Standard Deviations for a Gaussian Distribution
Table 2.8
87
that the greater the range for
It is obvious from Table 2.8
the
estimate,
the
higher
expressed as standard deviations about the
mean,
the confidence that the estimate will be within
the
specified range. The relationship further illustrates that a cost
interval
or
estimate must be associated with a confidence
measure of the standard deviation as related to a
interval
particular
probability distribution.
The
previous
discussion points out the
used in risk analysis.
major
assumptions
These are as followsl
or subsystems;
is made up of a set of elements
(1)
a system
(2)
total system cost is the sum of the element
(3)
there is a cost estimating
relationship
costs;
(CER) for
each
element;
(4)
variables,
the inputs to the CER are treated as random
and;
(5) the element costs are stochastically (i.e.,
statistically) independent
The last assumption has been used in several analyses (e.g.,
see ,Cefs.
the
input
19 and 28). However, the influence
variables
can have a significant
of dependency among
effect
on the
system CDF and subsequent cost risk analysis (e.g.,
total
see Refs. 31
and 32).
The
input
variation in the CDF for independent
variables
reduction
is shown
in Fig.
2.12.
versus
The obvious
dependent
effect
in the cost uncertainty for independent inputs
dependent ones.
Therefore,
is a
versus
the assumption of independent inputs
B8
I _
1.0
.8
.6
r
0
VC
3
X
2
0
.4
240
260
280
300
Cumulative Dersity Functions for Dependent and
Independent Input Variables
(Ref. 33)
Figure 2.12
89
320
360
results
in an overly optimistic prediction of
uncertainty
when
dependency does indeed exist.
The cost estimating methods discussed in this study use SWBS
weight
groups
program
as major
will
show,
ship weights,
input parameters.
As any ship
synthesis
there is a high degree of dependency
performance characteristics,
and other
among
technical
inputs.
2.3.2
Analytical Methods
2.3.2.1
Cost Element FPDFs
In the context
of this study,
a cost element
can refer to a
sub-system cost (i.e., SWBS weight group cost, GFM, etc.), a noncost input for a CER
constants
cost
or the regression
in a CER.
Analytical
of
(i.e., weight, SHP, etc.),
methods provide analysts with a certain
expectations
and dispersion in
certainty
by
measure
fitting
probability distribution functions with well-known and understood
properties
to the cost elements.
In reality, the actual shape of the PDF is unknown. However,
it
is
possible
to identify
some logical characteristics
(e.g.,
see Ref. 31). These include;
(1)
it should have fixed, positive upper and lower
bounds;
There
including
(2)
it should not necessarily
(3)
it should be unimodal, and;
(4)
it should be computationally
are
several PDFs that
be symmetric;
satisfy
simple.
these
(in decreasing order of complexity),
90
requirements,
Beta,
Gaussian,
triangular
and
distribution
distribution
uniform distributions.
can
be generated
The Gaussian
as a limiting
form
or
of
normal
the
beta
(Ref. 33).
Table 2.9
indicates the variation of skewness and
kurtosis
among these four distributions. Skewness is a measure of symmetry
about the mean and can be calculated using the expression;
skew = ( mean - mode ) / standard deviation
where
the standard deviation is the square root of the
(e.g.,
see
Eqn.
2.14).
Kurtosis
provides
peakedness of the distribution (i.e.,
of
(2.7)
observations under the curve),
variance
a measure
with high
Distribution
Skewness
Beta
any value
Triangular
indicating
values
Kurtosi s
> 1.8
O
5
-. 565 to .565
2.4
0
1.8
Uniform
Comparison of PDF Characteristics
Table 2.9
91
the
a comparison of the spread
less spread.
Gaussi an
of
use of any distribution requires that a minimum
The
amount
information for each cost element be provided to the analyst.
of
This
method
(i.e.,
opinion) and is generally in the form of the most
likely
information
expert
can be obtained by the
Delphi
value (mode), lowest bound and upper bound.
In
some
cases the user may not wish to set absolute
lower
may
be
a low estimate and an associated
probability
of
the
probability
of
and upper
limits.
supplied
with
underrun,
the
In such an instance,
mode,
a
the cost analyst
high estimate and
overrun.
The
of
the
preceding cost element data will completely specify any
four PDFs.
If it is only possible
to assign
and
upper
lower bounds to the element, the uniform distribution is used.
92
2.. 2.1.1
Beta PDFs
A range of normalized beta F'DFsare shown in Fig.
indicated
required
by
Eqn.
to
function.
(2.8),
uniquely
parameters alpha and
determine
the shape
of
the
than zero
(Ref.
33).
Eqns.
how these parameters are related to the mean and
The random
beta
As
are
normalized
For the beta distributions to be unimodal, both alpha
and beta must be greater
show
the
2.13.
variable
X, on the interval
(2.9-10)
variance.
0 to 1 can be rescaled
and
shifted to the cost element range of "a" (low value) to "b" (high
value), using the transformation
The
beta
because
it
distribution is well suited to
can
assume
(peakedness) and skewness.
model
any
variety
"a + (b-a)X".
many
shapes
with
costing
varying
itself to the procedure of
graph
which the analyst selects a graph from a family of
skewness.
selection,
an example of such a series
of plots,
variance and skewness.
93
showing
This
in
distributions
which best characterizes the cost element uncertainty.
is
kurtosis
These properties allow the analyst to
amount of variance (dispersion) and
lends
analysis
Fig. 2.14
differences
in
4
3.6
F
U
N
C
T
I
0
3
2.5
2
N
V
A
L
U
E
1.5
!
0
X VALUES
whe re
f(x) - r(a+ )
xai (
r(a)*r(s)
having
mean 14 and variance
11 -
0
2
B
a
(C) (++)
(ar+B) 2(Q+B+1 )
The Normalized Beta Probability
(Ref. 33)
Figure 2.13
94
(2.8)
2
a
a+
B-
Density Function
(2.9)
(2.10)
ri
SkwSud
Skw^lod left
LI0
mo
XuX
H
L
M
L
M .
L
M
z i
H
L
M
H
L
M
L
M
N
L
M
H
L
M
n
L
M
H
Medium
LW
V'erCl
H
Shape Variations using Beta PDFs
(Ref.
33)
Figure 2.14
95
Gaussian PDFs
2.3.2.1.2
The
form
of
the
Gaussian
distribution function is shown in
(or
normal)
Fig. 2.15 and
probability
by Eqn. (2.11).
Whereas the beta PDF requires four parameters to define its shape
and range (i.e., alpha, beta, high and low values), the Gaussian
PDF only needs a mean and variance (see Eqns. 2.13-14).
The
symmetric
applicable
elements
of the Gaussian
curve
is
to the distribution of a large number of
and
elements.
shape
is
not
particularly
applicable
to
generally
independent
single
cost
Also, the distribution has no finite endpoints.
These
character stics limit its use in cost risk analysis.
Trianqular
2.3.2. 1.3
tri.ngular
The
simpl i city
Its
shape
most
li kel
shape
PDFs
distribution
anc' wide range
of applicability
(e.g.,
due
to
see Ref.
can be determined with only knowledge of
the
its
33).
lower,
y, and high estimates. Fig. 2.16 indicates the possible
a
of
triangular distribution for three
(most like,ly values),
between
is very popular
the
lower
m.
different
modes
The location of the mode in the
range
es;timate,
a,
and
the
high
estimate,
determines the degree c)f skewness for the cost element.
The analytic form for this PDF is shown in Eqn.
(2.15), and
the resulting mean and variance for a triangular distribution can
be
calculated
integration
using
Eqns.
(2.16-17),
respectively.
of this PD)F gives CDFs in good agreement
complex analytical functions (Ref. 3).
with
The
more
*a
I.
P(x)
.I
0.0
p(x)
where
1--
p(x)-
P(x) =and a,
,)/(
p(t)d
(2.11)
(2.12)
the mean value of x and oa the variance of x.
N
fx=N
~
(2.13)
Xi
il
2
k
N
(Xi ',
x,)2
i-i
where range of index i from 1 to
N is over all members of class
The Gaussian Probability Density Function
Figure 2.15
97
(2.14)
3
2
F
R
E
a
U
E
N
C
y
.I
where
-)
a)(b
2(x
-
x- )
2 (b
(b - m) (b - a)
im
p .
x< m
(2.15)
f(x) -
having
; a
mean
and variance
; m x< b
2
(2.16)
a + m +bb
3
2
aO
a(a - nm)+ b (b - a) + mn
18
The Triangular Probability Density Function
Figure 2.16
98
b)
(2.17)
Uniform PDFs
2.3.2.1.4
For
cost element are known, the uniform
a
estimate
lower
the
"b",
of
this
function,
mean and variance
f (x) = 1 / (b - a)
(2. 18)
mean = (a + b)
(2.19)
/ 2
= (a - b) 2/
(2.20)
12
Method of Moments
analyzing
method of moments provides the capability of
The
the
risk (uncertainty) inherent in the summation of
and
dependent
using
For
can be used.
of PDFs are as follows;
variance
2.3.2.2
distribution
a
to "a" and an upper value equal to
density
probability
simplest
equal
of
and upper bounds
when only the lower
situations
cost elements
polynomial CERs.
that are input directly
In this study,
the
independent
or estimated
following
additive
polynomial relationship will be discussed;
Cost = C 1Cos
1 + C22 + C 3 + ... + CN
N
(2.21)
(2.21)
C s represent the independent or dependent subsystem
where the
costs. The values of C are estimated from CERs of the form:
C = X
+ X2X
P
where the X s are cost elements
a constant.
ignored
in
(2.22)
+ error term
or regression
constants
and F
is
The error term is generally unknown and is typically
subsequent calculations.
99
For costs that
are
input
C = X1.
directly,
Using
the
moment of methods,
the estimation of the
total
system cost PDF involves four basic steps:
(1) calculate individual sets of moments for each CN
in Eqn.
(2.21);
(2)
add the N sets of moments
(3)
fit a PDF with moments
in
calculated
(i.e.,
the
first step,
Eqn.
used
of
first origin
(moments
(2);
additive moments,
Eqn.
2.22) are calculated.
expressions
knowledge
equal to the sums
numerically integrate the PDF to obtain the CDF.
(4)
In
together;
to
calculate
the
A s, for each
CER
(2.23) shows that
the
moments
additive
the Taylor series representation
moment
about
Ref.
CER,
and three central
(the mean value),
the mean).
of the
require
32 indicates
that
the
moments
these
four
will provide sufficient information to determine the FDF
moments
asymmetry and shape and yet allow easy computations.
The Taylor series
,i'
of
mean
is
X.
straightforward
Expansion
and
would
to
about the mean,
higher
probably
and
values
is
order
be
The first additive moment does not include the
accurate.
series
input variable,
each
derivatives
(2.23) is expanded
in Eqn.
Taylor
the
merely the value of the CER calculated using
for all Xis.
Subsequent
moments
the CER partials, each evaluated at its i
require
more
the slim of
mean value.
For directly inputted costs, the value of the partials would
be
mean
equal
to one and the
cost,
variance
moments would be equal
additive
(i.e.,
A 1 and A 2 ) and a combination
1 00
to
the
of the
central moments.
If
second
the values of C N
step
complete
properties
then
into
A k moments)
linear
D
the
N
(Ref. 33).
dependence,
which
moments,
as the A moments
the
of the sum is equal to the
(i.e., the A k moment
sum of the individual
transformed
independent,
in the method of moments is the addition of
sets of A moments
For
can be assumed
(Ref.
have
33).
of the sum is equal to the sum of the
the A
must
moments
the
same
Therefore,
be
additive
the D k moment
individual D k moments.
The
relationship between the moments is given by Eqn. (2.24).
Once
then
a
the
total system cost moments have
been
PDF can be fit using the values calculated.
calculated,
The
actual
system cost moments will fall between the independent
total
dependent values obtained using
Several
(2.23-24),
respectively.
methods have been used for fitting a given distribution.
28 and 33 outline
Refs.
Eqns.
and
the methodology
involved
in fitting
the
moments to a Beta distribution.
If
an independent
Gaussian
distribution
is assumed
for
the
elements, only the first two additive moments (mean and variance)
need
to be calculated (Ref.
29).
For dependent variables,
the
mean and standard deviation are summed for the total system cost.
Since
than
the
a
square of a standard deviation summation
variance
associated
with
summation (for
dependent
the
variables
same
is
values),
greater
independent ones. This is illustrated in Fig. 2.12.
1 01
is
greater
the
risk
than
for
Al=
C (ul, u2,
. .
Uj)
A2 = ( acGX)
ax,
)2 U 1(2)
A3 =
( ac(x)
X,
)S U 1(3)
A, =Z
aax
ac(x)
(2.23)
)4 (U,(4)
- 3u,((2)2
)
where the partials are evaluated at u,,
.
and u,(,) is the kth central moment of X,.
D
1
=
C (u,
2
)2 Ul(2)
a(x,
(2.24)
accx) )9 U((3)
3
£(x,
3
4
u2, )
aC(X)
D2
4
,
-
where
(
ac(x)
)4
-AX,
Dk
U/( 4)
dependent
Ak = independent
additive moment
additive moment
Additive Moments
for the Methodof Moment
Calculations
Figure 2.17
102
2.3.3
Monte Carlo Methods
Monte Carlo or computer simulation methods are techniques in
which
the
constant
regression
is determined by selecting random numbers subject to
probability law.
defined
within
value generated for a cost element or CER
If the generated random
number
the limits of a pre-determined probability range,
a
falls
then a
particular value for the variable is chosen.
Because
variables
the
method
involved,
that
the full
this
respect,
simulates a
random
sampling
of
the
several computer runs must be made to ensure
range of values for each variable
the Monte Carlo method is
is obtained.
iterative,
In
since
the
exact number of calculations is not known beforehand.
Several
simulations
techniques
numerical
more
that
Slice (see Ref.
Monte
Carlo
techniques have been developed
ef f i cient
and
less
time
to
make
consuming.
Two
are discussed include Stratified
32). They differ from so-called
random
sampling
methods by the
Sampling
and
straightforward
number
and
size
distribution of intervals along the probability range.
Figure
Sampling
as
estimated
the
The
additive
2.18 shows the decrease in error for the
compared
error
is
from the differences between the simulation values
of
greater
simple
moments and those
convergent
reduction
to
nature
Monte
Carlo.
obtained from exact calculations.
of the simulations is obvious
in errors associated with number of
efficiency
and
The
Stratified
accuracy of
simple Monte Carlo is easily seen.
103
runs.
Stratified
from
the
Also,
the
Sampling
over
_11_
9% f
la
C
to
(A
tZ
On
a.
LU
1000
10,000
100,000
Number of Runs
Decrease in Variance Error with Increasing Run Size
for MonteCarlo Methods
(Ref.
32)
Ficure 2.18
104
1,000,000
CHAPTER
3
DESCRIPTION OF SPECIFIC SHIP COST MODELS
There is certainly no shortage of cost models in use at this
time.
use
There
by
models
are probably a dozen or so official cost models
the US Navy alone,
kept
"official"
within
not to mention a myriad
the agencies
estimates.
In
to
addition,
involved in the ship or shipbuilding
of
independently
the majority of
in
private
check
the
companies
industry possess their
own
cost models.
It
is important to understand that these cost models
serve
different functions for different organizations. For example, the
cost
models at NAVSEA Code 017 are intended to provide
estimates
for ship construction.
budetary
The models in use at the Naval
Center for Cost Analysis are used by OPNAV to provide independent
checks
on
new ship construction bids that
are
received.
models are used by NAVSEA design teams to perform cost
Cost
trade-off
analyses.
Contractor
cost models are used to submit contract bids for
new ship construction.
Also,
there are government agencies that
are interested in the development of cost models.
the development
For
instance,
of ASSET and RCA PRICE cost models at DTNSRDC
future Navy applications.
for
The
intent of this chapter is to provide a
of
cross-section
conceptual/preliminary cost models in
use
at
The first part of the chapter is a description of five
this time.
cost
representative
which are applicable for this type of
models
stage
early
These include NAVSEA Code 017, Naval Center for
cost estimating.
Cost Analysis (NCA), ASSET, USCG and RCA PRICE.
All
order
models require an estimation of
the
that costs can be determined.
for
accounted
using
for each model
parameters
Due
to
the
databases,
the
qualitative
one.
variety
in
are
Technology differences
of
As
methods.
the models and their use,
of
description
a
weights
SWBS
the
as
well
a
input
necessary
are listed.
proprietary nature of cost
comparison
of
these
Perhaps further
models
models
is
their
and
pirimarily a
studies could focus on a more
quantitative study involving specific vessel costing comparisons.
Typically,
the
point estimates during conceptual design are
plus or minus 15-25% range.
range may be reduced to 5-15%.
of
assumptions
After preliminary
However,
design,
the
due to the large number
inherent in any cost estimate,
it would
that any "reasonable" cost estimate can be rationalized.
10)6
in
appear
3.1
ACQUISITION & LIFE-CYCLE COST ESTIMATING
3. 1.1
NAVSEA CODE 017
3. 1. 1.1
General
SEA
Code
is
This office
of the Naval Sea Systems Command (NAVSEA).
Analysis
charged
the responsibility for
with
official
ship
cost
purposes
and
for
for
estimates
the
Department
annual
shipbuilding budget (e.g.,
see Ref.
the
preparing
Navy's
programming
and
planning
and
Estimating
017 refers to the Office of Cost
(DoD)
Defense
of
16). These responsibilities
encompass ship cost estimating and analysis at the initial design
feasibility
serves
as
study phase through production award.
advisor
in
trends
emerging
on
to NAVSEA
all elements
the
historic,
of cost
SEA 017
current,
and
estimating
also
and
cost
analysis.
Generally,
the
Shipbuilding
and
Appropriation (SCN). An SCN procurement
item
procurement
shipbuilding
Navy
Conversion,
that
SEA 017 estimates are used in the preparation of
account,
has been authorized by Congress must be full funded or
work must cease.
the
This full funded policy ensures that monies are
for all reasonable and expected costs through the ship
available
construction
and
post-del ivery
period.
Ship
estimates
cost
prepared under this policy are said to be "end costed".
every
Since
treated
have
as
official NAVSEA ship cost estimate
a potential budget candidate,
been established
certain
to ensure the estimate
is
proper context. These are as follows:
1. a written OPNAV (Operations, NAVY) cost and
1(-)7
be
requirements
is treated
feasibility request must be in hand
to
in
its
2. formal technical design inputs (e.g., SWBS weight
groups) must be available
3. an approved acquisition strategy and shipbuilding
schedule must be available
4. a cognizant Program Manager must be involved
It
obvious that ship cost estimating within the
becomes
NAVSEA
community operates within a controlled environment.
There are four principal divisions in the acquisition design
process: feasibility studies, preliminary design, contract design
and
design.
detail
their relationship
and
The first three new ship design phases
milestones
to the key Navy acquisition
and
to
the quality of SEA 017 cost estimates are shown in Figure 5.1.
detail
The
phase
or design agent.
shipbuilder
various
design
phases,
the
is
the
of
responsibility
As the design proceeds through the
quality of the
cost
estimates
increases
inputs.
with the range and depthof the technical
proportionately
In
study,
this
a
calculating
Class
the cost estimating methodology
F estimate
is considered
estimates
The
quality.
calculated
level
of
breakdown.
is
is
This
outlined.
to be of feasibility
design
method is essentially the same used
Class D and C estimates.
However,
Class
used
class
"ball
to
D and C estimates
calculate
use
one-digit
two-/three-digit
based on technical definition available at the completion
1 ()8
of
park"
The Class C estimate is considered bdget-quality
the Preliminary Design.
for
whereas a Class F estimate is
from the light ship weight estimate at the
SWBS,
the
and
of
C
D
Acauisition
Review Milestones, Technical Definition
and Cost Estimate Quality
(Ref.
16)
Figure 3.1
109
3. 1. 1.2
Cost Estimating Relationships
The
based
cost estimating methodology outlined in this section is
upon
the
input/output
model.
information
printout
available
in
Ref.
16
from the Unit Price Analysis
and
(UPA) cost
UPA model was designed to address the cost
The
an
elements
discussed in this section.
Cost
estimating
relationships
used
within
and preliminary design phases,
feasibility
SEA
017
for
are calculated
from
manhour and material costs divided by the weight of the
selected
seven major SWBS groups. These factors are updated annually based
on
historical data as well as return costs on previously awarded
shipbuilding
awards.
contracts
In addition,
and
the past year's
bid
data
on
new
all GFM lists are reviewed and adjusted as
required.
As
SEA
previously mentioned,
ship cost estimates generated
017 are said to be "end costed".
constitute
major
a total end cost estimate
category
codes
The cost
categories
by
that
are shown in Figure 3.2. The
(MCCs)
conform
to
the
cost
collection/accounting and budgetary systems of NAVSEA.
The categories of an end cost estimate can be separated into
three
groupings;
shipbuilder
portion,
material (GFM), and miscellaneous GA
The sum of the basic construction,
escalation
portion
The
and
change
order
cost
government
furnished
(general & administrative).
construction plans,
categories
contract
constitute
the
of monies that are paid to the shipbuilder/design agent.
GFM portion is made up of government
ordnance/air,
propulsion.
H,M&E
The
(hull,
remainder
mechanical
supplied
&
electronics,
electrical),
of the program costs are
110
and
associated
-C------------------C-·l
Gro
100
200
300
400
600
704
800
900
Other
Support
MCC800
Test and
Inst rutentat on
MCC541
Stock ShoreBased Spares
NC 533
Program Manager
Reserve
lCC951
total Dollats
Categories of a Total End Cost Estimate
(Ref.
16)
Figure 3.2
111
--------
with
a variety
of G&A costs.
The basic construction cost (MCC 211) is the central element
of the NAVSEA ship cost estimating process. Basic construction is
defined
as
construction.
the
It
original
contract
includes
award
all allowable
price
labor,
for
ship
overhead
and
material costs plus an amount for cost of money (COM) and profit.
Labor manhours,
MH (hrs), and material costs, MC (M),
cost
factors are developed for each of the SWBS weight groups 100-700,
such that;
MH. = KL. W.
1
where
1
(3.1)
1
Wi = appropriate SWBS weight group (LT)
KL i = appropriate manhour cost factor (hrs/LT)
MCi =i KM
where
KM
(3.2)
= appropriate material cost factor ($M/LT)
Estimates
groups
W
of
labor
manhours and material costs
800 and 900 are typically
sum of manhours and material
estimated
dollars
for
SWBS
as a percentage of the
for groups 100 through
700,
such that;
MH = FL 8 (MH1 + MH2 + ..
+ MH7)
MH 9 = FL9 (MH1 + MH 2 + ...
+ MH 7)
(53.3)
MC8 = FMB (MC1 + MC2
MC9 = FM9 (MC1
+
MC2
+
+ MC7 )
+ ..
+ MC7)
112
where
FLi = appropriate labor fraction (%/100)
FM. = appropriate material cost fraction (%/100)
1
Representative
0.25-0.35,
values
for these variables are;
FL 8 and
=
FL
and FM 9 = 0.01-0.10. Labor manhours can also
and; FM
be estimated using manloading profiles based on previous programs.
The design and builder's (D&B) margin is costed and included
as
part of basic construction on the assumption that the
will
be
"used up" during the development of the design
construction
margin
and in
of the ship as awarded. The costing of the margin is
done by applying the DB margin percentage to the total
and material
dollars
manhours
for groups 100 to 900;
MHD&
= FD&
(MH1 + ...
MH&B
E
F.D&BB
)
+ MH9
9
(3.4)
MC
where
D&B(
= F
(MC
D&
1
+
+ MC
)
9
F
= D&B fraction (%/100)
D&B
The D&B margin is generally on the order of 12%.
For calculating
employed
for
Manufacturing
labor
manufacturing
rates
is
and
separate labor
engineering
are applied to labor associated
weight groups 100-700,
rate
costs,
900 and the D&B margin.
rates
are
operations.
with
SWBS
The engineering
applied to the labor required for SWBS
group 800 work.
For any given
labor
rate,
the labor cost,
113
LC ($M),
is given by
the simple expression;
LCi - MH i */HR
where
(3.5)
$/HR = appropriate labor rate (dollars per hour)
Overhead
costs,
OV ($M) are calculated
as a percentage
the labor costs associated with each of the SWBS groups
of
100-900,
such that;
OV. - F
1
ovhd LC i
where
F
=
The
cost
ovhd
margin,
direct
C
labor overhead fraction (%/100)
of construction for each SWBS weight
($M),
is
and overhead
Ci
The
simply
the addition
labor costs,
group
of material
plus
cost
and
so that;
= MCi + LCi + OVi
(3.7)
cost for each SWBS group can be summed to arrive at
intermediate
CC
The
(3.6)
ship construction
= C1 + C2 + ...
- cost
contractors
for
of
the
money
cost
cost, CCC
($M), where;
C9 + CD&B
(COM)
is
an
(3.8)
intended
of providing
to
capital
compensate
for
facility
fraction
of
facility costs that contractors can treat as capital invested
in
investments.
the
costs
Government
marketplace.
is referred
standards
specify
the
The rate of return allowed on these investment
to as the imputed interest
rate.
The amount used for the COM is calculated by multiplying the
sum of the estimated direct labor costs by an appropriate factor.
114
This factor is computed by multiplying the shipbuilder's net book
value of assets times the imputed interest rate and divided by
The base is set equal to the direct
cost allocation base.
labor
a
labor dollars expended in the shipyard for a particular year. The
for the COM is as follows;
equation
COM = Fcm
where
F
cam
=
(LC1 + ...
the COM
(3.9)
LC9 + LCDB)
factor (%/100)
= (net book valua) (im_ ted
interest rate)
(allocation base)
Cost
of
money is also referred to as facilities cost
of
money
of the
basic
which can be abbreviated to FCM.
Profit
construction
is the final element
cost.
Profit,
percentage of the sum of all
in the estimation
C
($M) ,
is calculated
SWBSgroup plus margin costs.
as
a
It can
be expressed as;
Cprofit
where
and
(3. 10)
Fp CCC
Fp = profit fraction (%/100)
CCC is the construction
cost estimated
After- the profit dollars are calculated,
costs,
cost
of
money
and
(3.8)
in Eqn.
construction
the
profit are summed to
arrive
at
a
complete basic construction price, PBC ($M), where;
PBC
It
is
construction
=
CCC
+
Cprofit
important
price
be
that
+
(3. 11)
COM
all elements
making
up
adjusted to a common dollar
the
basic
base
date.
Shipbuilding contracts are generally costed to a given near-
115
term
base date.
reimburse
The contracts include an escalation
the
shipbuilding
shipbuilder
for
inflation
to
in
the
occurring
industry over the life of the contract (as measured
from the base date in the contract).
is estimated , Ces c ($M),
and
clause
The dollar requirement that
reflects a specified building
assumed labor outlay profile (e.g., see Fig. **).
an
Several of the end cost categories are typically
calculated
as a percentage of the basic construction price (i.e.,
Construction
,
plans
change
reserve categories are all
Construction
to
period
the
orders and PM
Eqn. 92).
(program
manager)
examples of this type of methodology.
plans refer to the nonrecurring costs
related
development of detailed construction drawings and
other
associated engineering tasks by the shipbuilder or design
Associated
computer
tasks
include
programs,
damage control books,
lead
ship
related
contractor
engineering
responsible
calculIations,
technical
ships selected records,
agent.
manuals,
and mock-ups. The
normally carries the majority of the costs
for
this
category.
Historical data is reviewed to determine what percentage the
construction
similar
plan costs were of the basic construction price
new designs.
Percentages developed in this
then adjusted by judgement factors,
obtain
the cost of construction
C
CCP
where
F
Fconstr
constr
such as ship complexity,
plans,
=acto
a complexity factor (typically > 1)
construction plan fraction (%/100)
116
are
to
CCp ($M), where;
=FCfactor
F
F constr PBC
r
manner
on
(3
12)
(3.12)
A
typical
value
for Fconstr
expressed in terms of material,
estimate
is
and overhead costs,
COM
The
is 0.12.
labor,
final
and profit.
The change order category is simply an allowance of money to
fund
necessary
changes
after
the
shipbuilding
is
contract
awarded. Some of the reasons for these changes are:
- To include state-of-the-art improvements that come
about during construction
- To correct "fit-up" problems that surface due to
incorrect drawings
- To correct differences between contract drawings and
ship specifications
-
To incorporate safety items that become evident
during construction
- To incorporate improvements that are generated by the
operational forces afloat
- To have the shipbuilder repair or modify GFM
(government furnished material)
- To change the contract ship delivery point or the
contract date of delivery
For
the lead ship,
change order costs ,
Ccord
($M),
are
estimated by the following expression;
Ccord =
10 PBC
(3.13)
For follow ships, change order costs are given by;
Ccord
0.05
(3.14)
PBC
117
The
10
and 5% figures reflect new NAVSEA guidelines
programmatic costs.
the
lead
and
of
respectively,
ships
has
been
taken
as
basic construction price
the
reduce
the cost of change orders for
Historically,
follow
to
12
and
8%,
(e.g.,
see
the
ASSET cost model description).
Although the change order dollar amount is estimated in base
year
dollars,
it
Therefore,
it is not inflated over the life of the contract.
throughout
represents a total amount to be spent
the construction period.
Government furnished material (GFM), including both hardware
and
software,
installation
Navy
is
by the
provided
aboard
Navy
to
shipbuilders
for
period.
The
ships during the construction
chooses to supply these selected equipments for any
cost
The GFM categories can be a significant
savings and convenience.
portion
security,
safety,
including standardization,
of reasons,
number
these
The estimator must ensure that
of the ship cost.
costs are not duplicated in the basic construction category.
electronics,
are categorized (see Fig.
CGF M ($M),
GFM costs,
Electronic items include electronics
Electrical) and propulsion.
production
components,
training
and
services,
engineering
and
(Hull11, Mechanical
H,M&E
ordnance/air,
3.2) into
support
repair
equipment,
parts
test
associated
and
with
installation.
The
ordnance/air
systems,
search radars,
training
support
items
include fire and
missile
control
missile launching systems, gun systems,
equipment,
test
and
integration
services,
landing aids, and selected catapult components.
H,M&E
items
consist
of
H,M&E
118
equipments,
small
boats,
installation
with
Some
equipment.
H,M&E
of
training
and repair parts
H,M&E engineering services,
support equipment,
associated
equipment,
environmental protection
special vehicles,
"unofficial" estimates have calculated the cost of this
category
as 2% of the basic construction price.
GFM usually involves nuclear propulsion systems.
Propulsion
Components
responsibility
generally
conventionally powered ships are
for
the
of the shipbuilder and are therefore included
in
the estimate for the basic construction price.
estimates for each of the GFM categories are
Cost
These lists are reviewed
using appropriate GFM "shopping lists".
ship design compatibility and base year pricing.
for
prepared
Historical
data on similar purchases are used for comparison purposes.
The shopping list approach for GFM cost estimation is broken
into the following cost elements:
- major
possible
it
make
for
as the primary units
defined
hardware,
the total system
meet
to
the
that
mission
requirement;
- ancillary equipment, required to logistically support
the
such
hardware
major
special tools,
gauges
as
data/documentation,
the
data
installation,
equipment,
and jigs;
- technical
documenting
test
special-purpose
complete
integration,
package
operation,
and
developing
for
and
associated
with
maintenance of
the
hardware or system;
- spares,
of
an
required to ensure the operational readiness
equipment or system until the normal Navy
119
supply
system
assumes support;
- system engineering,
ensure
to
engineering support required
the integration of the various components that make up
a
major hardware item;
- technical engineering services, contractor/vendor and
services required to support efforts
engineering
government
to
attain system operational readiness;
- other costs, including nonrecurring production startup, training, software, test and evaluation, and changes.
there is
the diverse nature of GFM equipments,
to
Due
no
standard procedure for estimating GFM costs. Delphi (e.g., vendor
quotes),
analogy,
wherever
they
CERs,
are
or engineering methodologies are
used
the
best
deemed
appropriate
produce
and
results. Whatever method is used, material, labor, overhead, COM,
and profit estimates are obtained for each GFM category.
remaining
The
miscellaneous
the
include
cost categories comprise
end
number
a
of
G&A cost elements (programmatic in nature).
These
test
and
of
categories
instrumentation,
other
shore-based
stock
support,
spares and program
manager
reserve.
other
The
functions,
costs,
support
in
support
engineering
services,
engineering
services.
previous
a
variety
of
G&A
including but not limited to equipment transportation
travel
Other
involves
category
support
ship
of
ship
commissioning
costs,
Cothe
r
acquisition programs,
acquisition managers,
acquisition,
ceremonies,
($M),
are
and
contract
in-house
estimated
are supplied by
the
from
ship
or can be estimated as a percentage of the
120
basic construction price, so that;
Cother
where
Fsup PBC
(3.15)
F sup = other support factor (%/100)
value
A typical
for
is 0.0C8.
test and instrumentation category includes the cost
The
and instrumentation (T&I) related to routine or
testing
leading
trials
F
to qualifying a ship
for
special
service.
active
of
The
majority of these costs will be borne by the lead ship.
Estimates for test and instrumentation costs, C
(M),
ship
The
obtained by analogy/comparison with previous programs.
are
acquisition program managers can provide necessary information on
the
number of trials and special tests that are
costs
can be adjusted to reflect the current
must
costs
The
be
inflated
to
the
requirements.
proper
time
The
period,
is conducted
before ship
stock shore-based spares category includes
procurement
that most of the T&I effort
remembering
delivery
also
Fast
required.
and acceptance
by the Navy.
of back-up spares for stock ashore or aboard tender/repair
ships.
The stock spares funded in this category are limited to first-ofits-kind
installations
propellers,
on
anchor chains,
the
lead
anchors,
Examples
ship.
inc Lde
turbine generators, diesel
and gas turbine engines, and selected shafting.
Items included in the stock spares cost,
supplied
will
by the program manager.
already
assembled
have
priced
C
In most cases,
the equipment
from
($M),
the
estimator
shopping
lists
or
H,M&E
for estimating the basic construction costs
121
are
GFM.
If
this
is
not the case,
marine vendor
quotes
can
be
obtained for the item.
The
Program Manager (PM) reserve category provides a source
of funds to the project manager for unforeseen future problems or
actions. This cost reserve is made up of the following elementsi
- general cost reserve, for general risks in the
estimating and acquisition process, equal to 1% of end cost (less
contract escalation);
-
lead ship cost reserve, for cost uncertainties
associated with introduction of the first ship of a new class,
equal to 3% of end cost
(less contract
escalation);
- follow ship cost reserve, for uncertainties
associated
with follow-on
construction,
equal to 2 and 1% of end
cost (less contract escalation) for complex and non-complex
ships, respectively;
- cost reserve for less than budget quality (i.e.,
Class D and F), equal
to 1% of end cost
(less contract
escalation)
Complex
craft,
The
and
ships are defined as combatants,
unique ships
other ships with significantly complex GFM
appropriate percentages to be applied to
122
systems.
estimate
dollar amount for PM growth can be found in Table 3.1.
and
the
Budget
Less Than
Budget
Quality
Qual i ty
Program Type
Lead Ship - New/Modified Repeat
4%
5%
- Complex
3%
4%
2%
3%
Follow Ship
- Not Complex
Percentages to Calculate
Program Manager Cost Reserve
(Ref. 16)
Table 3.1
The PM cost reserve,
CpM ($M),
is then estimated using the
following expression;
CpM = FPM (PBC + CC
CGFM +
+ Ccord
(3.16)
CT&I
where
Cspares
Cother
Fp
PM = program manager reserve cost fraction (%/100)
(from Table 3.1)
The total end costed price for a lead ship,
Ptotal ($M), is
estimated by summing the individual cost categories in Fig. 3.2;
total
CCp
+
+
PBC
Cesc
+ Ccord CGFM
+
(3.17)
Cother + CT& I + Cspares +CpM
When
multiple ships are awarded for construction,
123.
learning
and
benefits
reduced
costs
are
The
anticipated.
learning is highly dependent upon the acquisition
degree
of
Another
plan.
important aspect of pricing follow ships is the exclusion of non-
of non-recurring costs occur in the
majority
only. The
ship
borne by the lead
are
which
costs,
recurring
stock
shore-based
test and instrumentation and construction plans end cost
spares,
categories.
If learning will take place, appropriate learning rates must
be
(Eqns.
applied to both manhour and material dollar estimates
.1-3.2).
These
reductions
reflected in a
are
basic
reduced
construction price, PBC (Eqn. 3.11).
bid
Historical
data
cumulative learning curves,
Values
for
return
cost
that
indicates
data.
A
applied.
and so these are generally
from
learning rates are estimated
the
learning
labor
typical
prefer
shipbuilders
historical
from
rate
an
labor
and
Additional cost reductions for follow ship construction
are
"unofficial"
estimate,
applicable
to
direct
both
overhead, was 92%.
realized on change order costs,
manager reserves,
change
as
Ccord (Eqn.
3.14),
CpM (Table 3.1). Escalation costs
and program
Ces c
a result of reduced manhour requirements and
will
varying
construction periods.
Provided that learning rates have been properly applied, all
non-recurring costs have been eliminated,
been
Ptotal
total
adjusted,
(M),
the
total
can
be
end costed
estimated
124
by
and related costs have
price of
summing
a
follow
the
ship,
following
categories;
Ptotal - PBC+ Cesc Ccord+ CGFM
+
(3.18)
Cother
To
assist
estimating,
provide
Navy
CpM
cost
estimators
involved
include labor rates,
shipbuilding
pricing,
ship
POMs (program objective memorandum) are
guidance on a range of cost issues.
addressed
in
indices,
ROM
escalation.
cost of money
(rough order
For
example,
of
the
issued
Some of the
shipyard profit,
to
topics
material/labor
(COM), overhead,
magnitude)
cost
outyear
estimates,
and
following percentage values
are
suggested for the following variables;
Fp
10.% (see Eqn. 3.10)
F com = 5.588%
Life
purposes.
making
cycle
formulation)
For
example,
period
cost estimates are not required
However,
process,
(see Eqn. 3.9).
life
cycle
especially
phase
design
costing enhances
budgetary
the
decision
during the early planning
(concept
of the acquisition
trade-off
can be evaluated
for
cycle
(see
studies conducted
an a total cost basis
Fig.
3.1).
during
as well as
this
on
a
performance/technical basis.
NAVSEA
(Visibility
program
cost
and
estimators
Management
use
of
ship
data
Operating
and
from
the
VAMOSC
Support
Costs)
to assist in the preparation of the O&S portion of life-
cycle costs.
The VAMOSC program presents O&S data for Navy ships
which
in
were
active
commissioned
125
status
throughout
the
particular
reporting year.
the
(Chjef
CNO
of
The cost data are of a quality
Naval
Operations)
has
that
approved
the
acceptability of VAMOSC data for decision-making usage.
Operating
sections
for
and support costs are presented in
121
cost elements.
The
first
two
separate
section
displays
average costs for each of 21 combatant ship types (e.g., C,
DDG,
six
FF,
FF6, etc). The second section gives total 08S data for
classifications
of ships,
warfare,
patrol combatants,
support.
The
explanation
The
CV,
format
by
including
warships,
amphibious
mine warfare, mobile logistics, and
which these costs are
of all the cost elements
listed
and
an
is found in Ref. 21.
121 data elements are grouped into the five major
cost
categories defined below:
Direct
Unit Costs - identifies the direct cost
the
associated
operation and support of an
with
individual
ship
Direct
Intermediate
Maintenance
expended
by
equivalent
- cost of material
a
tender,
ashore
or
and
repair
afloat
labor
ship,
or
Intermediate
Maintenance Activity (IMA) in the repair and
alteration of other vessels
Direct
Depot
Maintenance
- cost associated
with
depot
level
maintenance performed for the ship by public
or
Direct
Recurring
private
facilities
Investment - cost of
Account
Appropriations
Purchase
(APA) and Navy Stock Account
repairable
repair
126
parts
consumed
by
(NSA)
or
procured for the ship
Indirect Operating and Support - cost of those services and items
(non-investment)
that
are required
by
the
ship after commissioning and launching which
are necessary to continue operations but
not
result
Operations
in
and
an
expense
do
against
Fleet
Navy
(O&MN)
Maintenance,
appropriations
Table
categories
3.2 indicates the five major categories and the
category
that
are
summed
amounts. Successive
in order to
sub-categories
indicated by further indentation.
127
calculate
the
in Table 3.2
submajor
are
Direct Unit Costs
Personel
Manpower
TAD
Material
Ship POL
Repair Parts
Supp
1 i es
Training Expendable Stores
Purchased Services
Printing and Reproduction
ADP rental - contract services
Rent and Utilities
Communications
Other
Direct Intermediate Maintenance
Afloat Maintenance Labor
Labor Manhours
Ashore Maintenance Labor
Labor Manhours
Material
Afloat Repair Parts
Ashore Repair Parts
Direct Depot Maintenance
Scheduled Ship Over ha ul1
Regular Overhaul
Public Shipyard
Private Shipyard
Ship Repair Facility
Selected Restricted Avail.
VAMOSC-Ships
Cost Categories
(Ref.
21)
Table 3.2
128
Direct Depot Maintenance (continued)
Non-Scheduled Ship Repair
Restricted Availability
Public Shipyard
Private Shipyard
Ship Repair Facility
Technical Availability
Fleet Moderization
Overhead
Public Shipyard
Ship Repair Facility
Labor
Funded Material
Special Program Material
Other
Outfitting and Spares
Other Depot
Naval Air Rework Facility
Field Change Installation
Rework
Design Services Allocation
Direct Recurring Investment
Exchanges
Organizational Exchanges
Depot Exchanges
Organizational Issues
Indirect Operating and Support
Training
Publications
Engineering and Technical
Services
Ammunition Handling
VAMOSC-Ships Cost Categories
(Ref. 21)
Table 3.2 (continued)
129
3.1.1.3
Input Information
The following input information is required by NAVSEA Code
017's Unit Price Analysis (UPA) computer cost model for
estimation
in
Table
minimum
ship.
of
3.3
Navy ship acquisition costs.
of design definition necessary to
parameters
below
1i sted
inputs
below are considered to be representative
amount
The
The
the
are grouped into the
of
the
describe
the
common
driver classifications outlined in Table 1.5.
TECHNOLOGY
Ship type (e.g., destroyer, SWATH, etc.)
Hull & Superstructure material
Propulsion plant type
Electrical plant type
SHIP DESIGN
Hull structure weight (SWBS group 100)
Propulsion plant weight (SWBS group 20)
Electric plant weight (SWBS group 300)
Command & Surveillance systems weight (SWBS group 400)
Auxiliary systems weight (SWBS group 500)
Outfit & Furnishing weight (SWBS group 600)
Armament weight (SWBS group 700)
Design & Builders margin
PAYLOAD DESIGN
Cost of payload GFE
(includes armament and weapons portion of com. & surv.)
NAVSEA Code 017 Ship Acquisition Cost Parameters
Table 3.3
cost
MANUFACTUR
I NG
Learning rate (labor & material)
Labor rate
(manufacturing & engineering)
PROGRAMMATIC
Cost of Money fraction
Profit fraction
Overhead fraction
ECONOMICS
Base year $
Base year for BLS data
BLSmaterial indices
NAVSEA Code 017
Ship
Acquisition Cost Parameters
Table 3.3 (continued)
Output from the NAVSEAmodel gives cost
lead
and follow
Table 3.4
ships.
information
for
The following additional information in
is used to escalate/discount
the costs throughout
construction period to a commonyear dollar value.
the
Inputted Costs
Lead ship end cost
Follow ship end costs
PROGRAMMAT IC
Contract award date
Start of Construction
Manpower profile (length of construction)
Time between ships
Number of ships required
ECONOMICS
Base year S
Inflation rates
NAVSEA Code 017 Fleet Escalation/Discounted Cost Parameters
Table 3.4
35.11.4
Output Information
The
NAVSEA new ship costing estimating program gives a cost
breakdown of lead and follow ship material,
total
acquisition (i.e., end) costs.
cost information:
one is a one digit SWBS group
ummary information available.
132
overhead and
There are two classes
next is the so-called P-8 estimating format.
the SWBS group
labor,
summary,
of
and the
Table 3.5 indicates
Cost Category
Cost Variable (Eqn #)
Lead
Follow
Ship
Ship
MC. (3.2)
same (*)
MC
same
Constructi on
Material Cost
SWBSGroup 100-700
800-900
D&B Margin
1
(3.3)
MCD&
B (3.4)
D&E
same
MH
same
Labor Manhours
SWBS Group 100-700
800-900
D&B Margin
1
(3. I:)
MH. (3.3)
MH
D&b
(3.4)
(*) must include effects of learning, if applicable
NAVSEA Code 017 SWBS Output Summary
Table 3.5
133
same
same
Cost Variable (Eqn #)
Cost Ctegory
Lead
Fol 1ow
Ship
Ship
Labor Costs
SWBS Group 100-900
LCi
1
(3. 5)
same
Overhead
(3.6)
OVi
1
same
C.1 (3.7)
same
Total Const:;uction
CCC
same
Cost of Money
COM (3.9)
same
Profi t
Cprofit (3. 10)
same
Construction Price
PBC
same
SWBS Group
100-900
Material + Labor+ Overhead
SWES
Group
100-900
(3.8)
(3. 11)
NAVSEA Code 017 SWBS Output
Table 3.5 (continued)
134
Summary
Cost Category
Cost Variable (Eqn #)
Lead
Ship
Follow
CCP (3. 12)
NA
NA (**)
Ship
Add-on
Construction Plans
Change orders
cord
cord
GFM
CGFM
Other support
Cother
(3.13)
Ccord
(3.14)
same
(3. 15)
Spares
C
Program Manager reserve
CpM (3.16)
Test & Instrumentation
CT&I
Escalation
C
End Costed (sai laway)
Ptotal
same
spares
same
esc
(3.17)
(**) Not Applicable
NAVSEA Code 017 SWBS Output Summary
Table 3.5 (continued)
135
same
same (3.18)
For budget purposes,
the
P-8 format.
so-called
cost
breakdown
3.2.
A
acquisition costs are documented using
The P-8 format gives an
using the and cost categories
partial
acquisition
outlined
listing of the categories used to
in
output
costs is given below.
MCC
Category
100
Plan costs
200
Basic construction costs
300
Change orders
400
Electronics (GFM)
500
H,M&E (GFM)
800
Other costs
900
Ordnance (GFM)
951
Program manager growth
953
Escalation
NAVSEA Code 017 P-8 Output
Table 3.6
136
Summary
Fig.
ship
3.1.2
NCA
3.1.2.1
General
for Cost Analysis) was
Center
(Naval
NCA
within the Department
direct and strengthen cost aalysis
guide,
to
established
of
the Navy and to ensure the preparation of credible
of
resources required to develop,
estimates
military
procure and operate
systems.
Ship cost estimates are the responsibility of the
in
Their
NCA-2.
primary
costing
activity is
to
personnel
independent cost analysis of each major ship acquisition
program
or considered by the Secretary of the Navy or the
to,
submitted
an
provide
In other words, NCA-2 develops an independent cost check of
CNO.
NAVSEA Code 017 estimates.
the requirement
to
Due
personnel
limited
firm
architecture
the
estimating
NCA-2
resources,
Gibbs
of
develop
of
cost
a
and
check
naval
the
contracted
& Cox to
construction
basic
cost
for an independent
method
for
future
near-term
frigates, destroyers and cruisers of the USN. Table 3.7 indicates
the
six
ship classes used in the database.
built at Bath
Iron Works
All the ships
(BIW).
To account for variations in design definition,
were developed:
were
two
models
a one-digit cost model that utilizes weight data
available at the SWBS one-digit level and a two-digit cost
model
that uses SWBS three-digit level weight data. These two levels of
detail
allow
available
at
costs to be estimated from
technical
the NAVSEA planning phase up to the
(i.e., including
Class
F and D estimates
information
butdget phase
as shown in Fig. 3.1).
Gibbs & Cox states that the goal of the models is to provide
137
feasibility
vessels.
level
estimates
for
near-future
(i.e.,
1980's)
The scope of the estimates is limited to shipyard costs
only (material and direct labor portion of the basic construction
cost) and does not include
the cost of GFM for the command
and
surveillance and armament systems (i.e., SWBS groups 400 and 700,
respectively).
However,
the shipyard's material and labor costs
for GFM installation are included in the models.
Ship
Class
DD 931
DDG 2
CG 16
CG 26
FFG 4
14
23
9
9
6
55-59
60-64
62-64
64-67
66-67
77-80
11/55
8/60
7/62
11/64
4/67
11/77
3960
4500
7800
7900
3426
3605
407
420
510
524
414
408
FFG 7
Number
Built
Year
Comm.
8(*)
BIW
Deli very
Date
Ful
1
Disp. (LT)
LP (ft)
* as of 1980
NCA Cost Model Ship Database
(Ref.
23)
Table 3.7
138
3.1.2.2
Cost Estimatinq Relationships
The
cost estimating methodology outlined in this section is
based upon the information available in Ref.
developed
for
shipyard
Additional costs,
material
such as GFM,
costs
overhead,
Navy program support are not included.
23.
and
CERs were only
labor
manhours.
training, spares, and
However, these additional
costs would be available to NCA through the same channels as
NAVSEA,
and so NCA would be equally
able to estimate
for
each of the
end cost categories shown in Fig. 3.2.
Various weight groupings (e.g., SWBS groups) were identified
by Gibbs & Cox as the primary cost drivers for the development of
the cost estimating relationships. Other significant cost drivers
include shaft horsepower (SHF),
kilowatts (KW),
installed generating capacity in
type of propulsion plant, and tubic number (CN).
These variables are related to material costs and labor
manhours
for each cost group through simple cost algorithms.
These algorithms
regression
small
were developed using a linear
technique.
number
of
The linear
squares
least
fit was used because of
sample points and the
Uncertainty
of
the
future
technological direction.
The
the
costs of all the relevant groups are added
basic shipyard costs associated with a lead
costs
and
tabular
ship's
labor manhours are presented in
lists
of
delivery;
inflation)
productivity)
and
three
to
ship.
provide
Material
formats:
costs and labor manhours at the time
(2) graphs of material
labor
plotted
manhours
costs
(adjusted for
against a relevant
of
the
(adjusted for
BIW
parameter,
shipyard
and;
mathematical equations calculated from the graphs in (2).
139
(1)
(3)
-
1950
1960
1970
FFG-7
1.32
FFG-4
CG-26
CG-16
2.75
2.90
2.92
DDG-2
2.95
D0931
2.97
-
Shipbuilding Inflation Factors for 1980
(Ref. 23)
Figure 3.3
140
1960
BIW
return
cost
data and material
costs,
MC
($),
were
adjusted for inflation from the delivery date (see Table 3.7)
to
a 1980 standard using the following relation:
MC19
1980
MCDelivery
Delivery
ate
Date
(3.19)
Inflation Factor
where the Inflation Factor, shown in Fig. 3.3,
was based upon the
steel vessel index from the Statistical Quarterly (e.g., see Ref.
24).
BIW
by
use
total
of
labor manhours were adjusted to
the productivity curve shown in
1960
Fig.
standards
3.4
and
the
following simple expression.
MH1980 = MH
Delivery Date Productivity
Variations
workload
Factor
(3. 20)
in the Productivity Factor were due to the effects of
and
management practises over the building
period
of
interest.
3.1.2.2.1
One-D
i git Model
The one-digit level cost model relates one-digit SWBS weight
group estimates,
cubic number, KW, and SHP of a ship to material
cost
manhours graphically or through
and
labor
simple
linear
equat i ons.
Fig. 3.5 shows a representative
SHP
(shaft horsepower) for SWBS 200
points
were
obtained
from
the
indicated as filled-in circles.
by
plot of material
- Propulsion.
ships in
Table
cost versus
Actual
3.7
Starred (*) points were
and
data
are
derived
BIW estimators based on return costs of similar systems.
The
solid line in Fig. 3.5 indicates the calculated algorithm for the
actual data points. The dashed lines are projected algorithms for
141
PRODUCTIVITY
MH/TON
600-
500-
-0.83
I
-I
-
FFG-7_ 0.90
400-
AND PMHTO
FACTOR
PRODUCTIVITY
AND
FFG-4.
1. 18
CG-26
1.22
CG-I16-_ 1.17
DDG-2
DD931
300-
I 17
l 1.01
200100Y
,
-
-
-.
--
. 9
DELIVERY
YEAR
1955
28
I
I 960
C5
r
--
.
-1.25
v
I
I
I -
.
a',
-I
B.I.W.
I
1970
1965
DESTROYER TYPE VESSELS
*.
.
-
A
.
-
-
-
-
EXPERIENCE
23)
Figure 3.4
142
I
W
N.. ,
1980
1975
NO DESTROYER TYPE VESSELS
Total Ship Labor Factors for 1980
BIW4
(Ref.
.
.
m
-I .33
50
40
aCo
30
nv
la
O
C
20
.C
oli
10
40
60
50
70
SHAFT HORSEPOWER
(in thousands of SHP)
One-Digit Propulsion Systems Material Cost
(Ref. 23)
Figure 3.5
143
80
90
actual or derived data points,
Cost
cost,
algorithms were developed from the plots for
material
and labor manhours, MH, for each of the nine SWBS
MC ($),
groups (e.g., see Fig. *.*) according to the following equation:
MH 1 or MC i1 = Slope
i CD 1 + Intercept
'
where
(3.21)
CD. = particular SWBS group cost driver
Eqn.
shows
3.21
and
relationships
intercepts
for
the
functional form
is not meant to imply
cost and
material
of
the
identical
labor
estimating
slopes
and
Table
3.8
manhours.
indicates the cost drivers for material and labor for each of the
SWSS
groups.
There are several SWBS groups where more than
one
is satisfactory for the generation of
the
variable
independent
cost algorithms.
The
cost and labor figures
used in the model for design and
engineering services (SWBS group 800) are applicable to USN ships
built after 1970. They include costs from the fbllowing areas:
Program Manager's Office
Integrated Logistics Support
Reliability, Maintainability, and Availability
Data Management
Producibility Management
Test and Evaluation
Integration Engineering
Configuration Management
The
figures used for the lead ship should be reduced to 50%
for
the
first follow ship and to 20-25% for succeeding ships of
the
same class.
Construction
length
of
delivery).
time
If
the
services were found to be proportional to
vessel
is
the
in the shipyard (from keel laying to
the building time is unknown, 30 months can be
144
Cost Driver
SWBS
Group
Material
Labor
Application
- aluminumsuperstructure,
cubic number
100
-
100 wt.
300
400
600
- steam, geared GT, electric
GT - single & twin shaft
SHP
200
KW
300 wt.
- steam & diesel generators
400 wt.
- early & current technology
500 wt.
- steam
constant values
900
months construction
heat
- pre & post 1965 habitability
- early & current
700 wt.
800
electric
- pre & post 1965 habitability
length x beam
600 wt.
700
all hull steels
aluminum superstructure,
all hull steels
One-Digit SWBS Group Cost Drivers
(Ref. 23)
Table 3.8
145
technology
assumed for an FFG-7 type vessel.
In orderto estimate the total material and labor
for the
lead ship, the individual SWBS group costs and labor manhours are
summed.
rate.
this
Labor
costs are calculated using the appropriate
labor
The total direct labor plus material cost estimated
using
method
not
does
Adjustments
for
curves shown
in Figs.
The
include any design
margin.
builders
and productivity are
inflation
one-digit
&
based
on
the
3.3 and 3.4.
estimate
presumes a
subsystems within each SWBS group.
given
of
combination
The disadvantage of this type
particular
of estimate is its inability to differentiate between
design features of a new ship and those of the baseline ship from
which
the
algorithms
The two-digit
were derived.
cost
model
attempts to overcome this limitation.
3.1.2.2.2
Two-DigSi t Model
The two-digit
of
manhour CER.
groups,
of 22 cost
by at least one material
which is represented
labor
for
level model consists
cost
each
CER and
a
Table 3.9 indicates the cost group structure
This structure was developed after
the two-digit model.
was recognized that the BIW shipyard data was not accumulated
it
in
the identical manner that the USN used.
The definition of each cost group in Table 3.9 is determined
by
its
equipment content,
breakdown.
Appendix
D
as defined by the
shows
the
SWBS
contribute to each of the 22 cost groups.
the
two-digit
SWBS
weight
groups
The main advantage
level model is that the increased
groups are more specifically related to weight,
146
three-digit
number
of
that
of
cost
volume, or power
Cost
Cost
Group
-
Group Title
- - - - - -
1A
- - - - -
- - -
Structural
- - - -
Group
_ _ -
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _- -
Envelope/
Group Title
_ _ _ _ _ _ _ _ _ _ _ -
_-
_ _ _ _ _ _ _ _ _._
_ _
5A
Environment System
58
Fluid System
5C
Maneuvering System
5D
Equipment Handling System
6A
Hull Fittings
6B
Non-Structural
Subdivisi ons
18
Superstructure
iC
Foundations
1D
Structural
Attachmen ts
I
2A
Propulsion
Energy Sy stem
2B
Propulsion Train System
2C
Propulsion Gases System
6C
Preservation
2D
Propulsion Service System
6D
Facilities
3A
Electrical
Electrical
Power Generaticn
6E
Habitability
Power
7
Ordnance
8
Design & Engineering
Services
9
Construction Services
3B
Subdivisions
Di stri buti on
4A
Vehicle Command
48
Weapon Command
Two-Digit Cost Model Structure
(Ref.
23)
Table 3.9
147
Cost Driver
Cost
Group
_
_
_
_ _
Material
_
_
_- -
_ . _
B1
_
_
_
_
_
Labor
_ _
_
_
_
_--
_
_ _
Application
_
_
_
_
_ _-
_
_
_
_
_
_ _
_
_
_
_-- -
_ _ _ _ _ _ _ _ _ _ _ _ _ _
cubic number
- all hull steels
1A wt.
- all hull steels
1B wt.
- aluminum, steel
superstructure
1C
1C wt.
1D
ID wt.
- steam, gas turbine
SHP
- steam, geared GT, electric
GT - single & twin shaft
2B
SHF'
- fixed, controllable pitch
2C
SHP
- steam, gas turbine
2D
SHP
2D wt.
- steam, gas turbine
KW
3A wt.
- steam & diesel generators
38B
3B wt.
4A
4A wt.
- early & current technology
4B
4B wt.
- early & current technology
Two-Digit Cost Model Cost Drivers
(Ref. 23)
Table 3.10
148
Cost Driver
Cost
Group
Material
Labor
Application
5A
5A wt.
- steam & electric heat
5B
5B wt.
- steam & gas turbine
5C
length x draft
5D
5D wt.
6A
length
- early & current technology
6A wt.
x
beam
6B
1 ength
- pre & post 1965 habitability
x
depth
- pre & post 1965 habitability
6B wt.
6C
6D
bE
700
- early & current technology
length x beam
complement
- pre & post 1965 habitability
6D wt.
- pre & post 1965 habitability
6E wt.
- early & current technology
700 wt.
800
constant values
900
months construction
Two-Digit Cost Model Cost Drivers
(Ref. 23)
Table 3. 10 (continued)
149
than
are
the more general seven cost groups
of
the
one-digit
model.
Material and labor estimates are linearly related to various
cost drivers in a manner identical to that described for the onedigit
the
model
22
according
to Eqn. 3.21).
The cost drivers
cost groups are listed in Table 3.10.
3.8
Tables
(i.e.,
two-digit
A
comparison
for
of
and 3.10 shows that the cost drivers for the one- and
models are similar.
This is not surprising since
the
two models are closely related through the SWBS groupings.
As noted previously,
GFM,
but
does
installation.
the model does not include the cost of
account for the costs associated
Installation
includes
material
with
and
shipyard
labor
for
foundations, mounts, magazines and hoists, supporting hydraulics,
cables,
and electrical systems and their testing. Cost groups 4B
and 7 are most affected
The
more
by GFM considerations.
detailed two-digit cost
wherever
there
separate
groups
breakdown
is
warranted
is sufficient technical data to distinguish
listed in Table 3.9.
When there is a
lack
the
of
weight data for a particular cost group, the weight amount can be
estimated by use of a weight algorithm (see Ref. 235 Appendix B),
or by comparing the known new ship weight percentage distribution
with
a
baseline
distributions
ship.
Table
5.11
lists
the
percentage
of lightship weight for the six ships included
in
the Gibbs & Cox study (see Table 3.7) plus the average SWBS group
weight distribution for 12 Navy combatant ships.
150
Cost
Group
DD 931
DDG 2
CG 26
CG 16
FFG 4
FFG 7
Avg
_--______________________________________________________________
28
28
37
38
37
B
3
3
3
3
3
4
C
4
4
3
3
4
5
D
2
2
2
3
3
2
36
37
46
47
46
46
2A
21
18
12
11
9
5
B
4
4
5
3
3
C
I
1
1
1
1
1
D
4
3
2
2
2
1
30
26
18
17
15
10
2
2
2
2
2
2
2
2
2
4
4
4
4
4
4
7
4A
1
1
1
1
I
1
B
2
4
6
6
5
3
3
5
7
7
6
4
1A
B
Model Groups
(Ref. 23)
Table 3.11
151
16.9
4
Percentage Distribution of Lightship Weight
for Two-Digit
45.9
4.8
5.7
Cost
Group
DD 931
DDG 2
CG 16
CG 26
FFG 4
FFG 7
5A
3
3
3
3
3
4
B
6
7
6
6
6
9
C
1
1
1
1
2
2
D
1
1
1
1
3
2
Avg
_____---_______--____-----___-----____________________________
11
12
11
11
14
17
6A
1
1
1
1
1
1
B
1
1
1
1
3
C
2
2
3
4
4
D
1
1
1
1
2
3
E
1
1
1
2
2
7
8
7
8
10
12
8.3
9
a
7
6
5
4
5.5
100
100
100
100
100
100
13.0
7
Total
Percentage Distribution of Lightship Weight.
for Two-Digit Model Groups
(Ref.
23)
Table 3.11 (continued)
152
100.0
3.1.2.3
Input Information
The
Gibbs
&
material
ship.
between
digit
following
input information is required by the
Cox ship cost models in order to estimate the
inputs indicated in Table 3.12 do
the
model
not
the SWBS weight group data for
the
the
Ship type (frigate, destroyer, or cruiser)
Hull & Superstructure material
Propulsion plant type
Electrical plant type
level of technology
SHIP DESIGN
Hull structure weight (SWBS group 100)
Propulsion plant weight (SWBS group 200)
Electric plant weight (SWBS group 300)
Command & Surveillance systems weight (SWBS group 400)
Auxiliary systems weight (SWBS group 500)
Outfit & Furnishing weight (SWBS group 600)
(SWBS group 700)
Armament weight
Number of Propeller
Shafts
Length, Beam, Draft
Steam or Electric
Heating System
Missile magazineflooding requirement
Habitability Standards
Cubic Number
SHP
Generator rating
(KW)
_______--_-------__----_________________________----------_----
NCA Lead Ship Shipyard
Cost Parameters
Table 3.12
153
two-
one-digit
TECHNOLOGY
Early or Current
lead
differentiate
more detailed weight inputs required for
and
-
shipyard
and direct labor costs (excluding margins) of the
The
NCA
PAYLOAD DESIGN
Cost of payload GFE
(includes armament and weapons portion of com. & surv.)
MANUFACTURING
Labor rate
(manufacturing & engineering)
PROGRAMMATIC
Length of Construction
ECONOMICS
Base year $
Base year for
LS data
BLS material indices
NCA Lead Ship Shipyard Cost Parameters
Table 3.12 (continued)
model.
The
required
parameters
for
base
listed under the economic
year
calculations
outside
are
category
of
the
range
indicated in Figs. 3.3 & 3.4.
Calculation
Fig.
3.2
for
of the remaining end cost categories
the
shown
lead and follow ships are carried out
manner similar to that outlined for the NAVSEA Code 017 UFA
model.
154
in
in
a
cost
1.2.4
53.
Output
Output
Information
from the NCA - Gibbs & Cox cost model is limited
shipyard material costs,
for
the
major
MC ($),
SWBS groups
and direct labor manhours, MH,
(or a specified
values are shown below in Table 3.13.
labor
costs estimated using this methodology can
to within 20% of actual costs
Material and direct
be
considered
(Ref. 23).
Cost Variable (Eqn #)
Cost Category
Lead
Ship
Construction
Material
subset).
two-digit
These
accurate
Cost
MCi
(3.21)
1
SWBS Group 100-900
Labor Manhours
SWBS Group- 100-900
MHi
NCA Lead Ship Output Summary
Table 3.13
155
to
(3.21)
3.1.3
ASSET
3.1.3.1
General
ASSET,
which
Evaluation
Tool,
is
an
addresses
acronym
most
for
of
Advanced
the
Surface
major
Ship
technological
of naval architecture that are relevant to the design of
domains
Navy warships (e.g., see Ref. 15). ASSET was developed by Boeing
Company
for
the
David
Computer
Services
Research
and Development Center (DTNSRDC).
3.6,
and
Naval
Taylor
As indicated in Fig.
is intended to be used primarily in the
ASSET
Ship
exploratory
The
feasibility phases of the ship design process.
program
has proven useful in assessing a variety of whole ship technology
impacts
in
a consistent
manner.
Because it
is
essentially
a
synthesis model, ASSET also serves as a repository of current USN
design practises.
Technical information,
formulae
were
principally
ship data,
algorithms and empirical
supplied by
the
US
Navy.
versions of ASSET are available for a variety of ship
the present time,
ASSET can handle hydrofoils,
Updated
types.
At
monohull surface
combatants and SWATH (Small Waterplane Area Twin Hull) ships.
The
ASSET
program
has been divided into
initialization, synthesis and analysis. During
the
data
entered
completeness
to
define the current ship
Next,
and fatal errors.
the data
three
sections:
initialization,
is
checked
is
for
synthesized
until an integrated ship design is acheived, in that each element
of
data
element
analyses,
that
defines the ship is consistent with
of ship data.
such
as
Once the design
cost,
are
156
has
carried
every
converged,
out.
The
other
various
different
'I
MISSION
REQUIREMENTS
.............
~~~~~~~
I
. -.
I
I
I
II
I
I
I
.
.
I
-
I
EXPLORATORY
STUDIES
I
I
I
I
I
,.
I
I
i
I
,
I
-
I
I
FEASIBILITY
STUDIES
ii
I
I
i
i
ii
i
I
I
-~~~~~~~
PRELIMINARY
I
I
L
DESIGN
I
.
,,
_
I
i
I
I
I
I
-I
·
CONTRACT
DESIGN
I
DETAIL
I
.
ASSETApplicability
Within the Ship Design Process
(Ref. 60)
Figure 3.6
157
DESIGN
I
INITIALIZATION
INITIALIZATION
START
HULLGEOMETRY
HULLSTRUCTURE
RESISTANCE
PROPELLER
SYNTHESIS
MACHINERY
WEIGHT
DESIGNSUMMARY
CONVERGENCE
YES
END
PERFORMANCE
HYDROjTATICS
SEAKEEPING
COST
SPACE
MANNING
ASSETCaomputational Addules
(Ref.
1)
Fiaure 3.7
158
ANALYSIS
computational
Fig.
modules within each section are indicated in
3.7.
The
intent
of the costing module in ASSET is
to
provide
costs
data which can be used to evaluate the relative life cycle
of
competing ship systems.
major
three
investment
research
and
considered
cost
in
(R&D),
development
operations and support (O&S).
theoretical basis of the ASSET cost analysis module
The
the
and
categories:
Life cycle costs are
evaluation
program developed for the
Advanced
is
Naval
Vehicle Concept Evaluation (ANVCE) study. This study, done in the
early
used the latest available lead shipbuilding
to mid 1970s,
cost data on a variety
vehicles
to develop
of surface,
air and special
CERs of future vessel
types.
purpose naval
The ship sample
for the ANVCE study and the resulting SWBS group CERs
used
have
been placed in Appendix E.
The
ASSET
cost
development
of
module is shown in Table 3.14.
The list of
vessels
indicates the wide range of vessel types (i.e.,
SWATH,
the
sample database used in the
ship
monohull,
etc.)
and
accommodated.
159
displacements
hydrofoil,
that
can
ACV,
be
Base
Year
I
1960
1961
1964
1973
1957
1956
1958
1961
1973
1976
1959
1962
1968
1970
Contractor
Both Iron Works
a'
a
a
.
'
a
a
"
.
'
·
·
'
'
*
'
·
'
'*
O
"
.
Bethlehes
Steel
0
a
Newport News
sh p
Class
DC 1037
DE 1040
DE 1053
PF 109
DD 2
DLO 12
DLG 16
DLG
0 26
SWATH DO
SWATH CM
DLGN 25
DLGN 35
DLGN 36
DLON 36
Cost
Basis
Total iMaximum
Continuous shp
BID
'
.
PROPOSAL
BID
'
STUDY
20,000
35,000
35,000
40,000
70,000
8·
5,000
2,654
3,331
4,066
3,561
4,562
5,786
85,000
85,000
60,000
7,645
7,839
3,710
3,441
8,514
8,947
10,056
10,497
40,000
PROPOSAL
BID
.
PROPOSAL
60,000
70,000
70,000
70,000
6,140
15,600
25,850
1973
1967
1976
. Tacoma
Peterson/Tao,
U
.
CPIC
PG 92
PGG 1
BID
'
1967
Grumann
PGH I
PROPOSAL
.3,470
1967
Boeing
PGH 2
ACTUAL
3,260
1961
1974
1964
1975
1974
1974
1974
1975
1974
*
*
'
.
.
Aerojet
Bell
a
Rohr
PCH 1
PHM 1
AGEH 1
'
DBH
STUDY
ARTIC SEV
JEFF A
JEFF B
2KSES ()
2K1SE (R)
'
BID
a
491 ACTUAL
75i ACTUAL
PROPOSAL
ASSETCost bdule Ship Database
(Ref. 1)
Table 3.14
160
FLD (LT)
6,800
17,300
29,800
76
254
390
66
58
111
232
295
54,020
1,042
48,000
11,200
16,800
96,000
96,000
540
152
152
1,950
1,800
3.1.3.2
Cost Estimating Relationships
calculative sequence employed by the ASSET cost
The
is
a seven step process,
step
module
as illlstrated in Figure 3.8. The first
involves a complete review of all data input to the
via the current model.
invalid
The input data are scanned
Where appropriate,
entries.
module
for missing
or
replaced
missing data are
with default values.
The next step involves the calculation of miscellaneous ship
data
that
are required as input to
various
CERs.
These
data
include the following:
NCi
th
=
the i
element of the number
of
crew
aboard
2=petty officers, 3=enlisted
ship (officers,
men),
PSum = the total power available from all of the ship's
propulsion
engines,
NHH = the number of helicopters aboard ship,
WL = the lightship weight,
SWBS weight groups
equal to the sum
plus margins,
the
of
and
WMp = the weight of the ship's costed military payload,
listed in Appendix
Following
lead
and
Construction
acquisition
the
follow
costs
A.
calculation of the miscellaneous ship
ship
acquisition
costs
are
data,
determined.
are the first costs estimated for the
total
cost.
For the lead ship, the cost of the ship elements or services
within each of nine major SWBS groups,
161
plus the margin cost,
is
PROCESS
INPUT
DATA
X
CALCULATEMISCELLANEOUS
SHIP DATA
.I
~~
DETERMINELEAD SHIP
ACQUISITIONCOSTS
DETERMINEFOLLOWSHIP
ACQUISITIONCOSTS
DETERMINEFLEET LIFECYCLECOSTS
I____
i-
CALCULATETOTALSYSTEMSCOSTS
(UNDISCOUNTED
AND DISCOUNTED)
-
--
-
-
--
.
.-
GENERATEPRINTED
OUTPUT
ASSETCost Module Calculative
(Ref. 1)
Figure 3.8
162
Sequence
to
computed according
C1 =
2 =
C =
3
C
=
C
=
4
5
C =
6
(33.95
(1.86
(75.05
HP
)0.808
)(W
)
KN2) (PuM)
N3
300
0.910
617
KN4 )(W40)
(108.57
N4
400
(94.87
KN5 )(W
(98.59
KN )(W
)C78
N6
600
(W
/
= (0.019
C9 = (0.385
*N
500
K )(W
N7 W700
(WLS -
equations;
772
100 )100
Ni
K
C (8.38
C8
KN1) (W
C (1.86K
=
7
CM =
the following
(3.22)
0782
0 987
W )) (CL1 + ... + CL7)
KN8) (CL1 +
KNg)
~N9 (CL
LI
+ ...
C
+~C
CL7
)1.099
LM
++C
+ C ) 0.839
CL7 + CLM
where C 1 = the appropriate lead ship SWBS group cost, $K
KNi - the technology factor for the i
th
SWBS group
Wi = the appropriate SWBS group weight, LTON
and subscript M refers to the D&B and growth weight margins.
163
The KN technology factors used for Eqn.
based
of
(3.22) are selected
upon the characteristics of the ship to be costed.
KN
values for various ship configurations
Appendix
B.
are
Tables
included
These tables were originally derived from the
in
cost
evaluation program of the ANVCE study.
Follow ship construction costs are principally a function of
lead
ship
learning.
costs
and
a factor
that
For the first follow ship,
accounts
for
production
this factor, F, is defined
by the following equation;
= 2.0 R LL -
F
(3.23)
1.0
where RL = learning rate
The
value
ASSET
equal
rate
(default
of the
average
model uses a cumulative learning
to 0.97) to account for the decrease
cost per ship resulting from increased production.
Ci
Follow ship construction costs,
($K),
are given by
the
following expressions;
C j2
1,2,
..
7
= F CL
=M
7,9,M
L1,L2, ..
L7,L9,LM
(3.24)
C 8 = (0.019 KN8) (CL1 + ...
where
= appropriate
C
+ CL7 + CLM)
lead ship SWBS group and
Li
from Eqn. (3.22).
164
margin
costs
The ship construction
CCC ($K),
cost,
is the sum of weight
group costs, such that;
CcC = C 1 + C 2 + ... + C 9 + C
Profit, C
($K),
and
of
is assumed to equal a fraction
P C ($K),
construction cost and the price,
cost plus
(3.25)
equals
the
construction
profit.
As a result;
Cprofit
FP CCC
(3.26)
+
(3.27)
PBC
=
CCC
Cprofit
where Fp = profit fraction (ASSET default value of 0.15)
A
series
of
miscellaneous
program (G&A)
costs
must
computedand added to the price to determine the acquisition
of
the
ship.
be
cost
These costs are described below:
Ccord = cost of change orders,
Cnsea = NAVSEA support
$K
costs, $K
Cpdl
= post delivery charges,
Coutf
= outfitting costs, $K
K
Chmeg = hull/mechanical/electric plus growth, $K
They are estimated as a percentage of the acquisition price, F'BC,
according
to the values listed in Table 3.15 for lead and follow
ship applications.
165
Cost
Lead Ship
El ement
Percentage
Follow Ship
Percentage
C
12
8
2.5
2.5
C
cord
nsea
5
Cpdel
Coutf
Chmeg
4
4
10
10
NAVSEAProgram Cost Percentages
Table 3.15
The total ship acquisition cost, Cacq ($K), is given as;
Cacq
PBC
'cord
Cpdel
The "sailaway"
C
where
C
P
S
acq
+ Cnsea
+
+
(3.28)
Coutf + Chmeg
cost of the ship, C S
(WK), is given
by;
(3.29)
p
ship payload cost,
166
The
value
of
C
P
by
may be input or estimated
the
following
equation;
C
where
payload WMp) + (18.71 N H )
Vpayload = 0.373
= 0.326
The
(3.30)
) 1000.
for the lead ship
for the follow
ship
in
factor 18.7 approximates the cost (in millions)
FY
(fiscal year) 81 dollars of acquiring a helicopter. Payload costs
are defined as the sum of ordnance and electronics costs.
The next sequence
in the ASSET cost module
the estimation of fleet life-cycle costs,
and
categories.
The
these categories
R&D
cost
of
and
investment,
development,
(see Fig. 3.8) is
consisting of research
operations
and
support
major life cycle cost elements addressed
under
can be seen in Fig. 3.9.
costs are calculated as the sum of two components:
the
test
and
design
and
development,
and the
cost
of
evaluation.
Design
and development
costs
are computed
167
as the sm
of ship
ASSETLife Cycle Cost Elements
(Ref. 1)
Figure 3.9
168
design
and development
CSD D
cost,
($M),
and other
design
and
development cost, CODD ($M);
= 0.0011
C
(0.571 C
SDD
) + C
/F + 0.072 C
Lp
Facq
TECH
(3.31)
CODD = 1.1 (0.06 CF
where
=
CFacq
I )
acquisition cost of the first follow ship, $K
CLp = lead ship payload cost, $K (see Eqn. 3.30)
=
CTECH
technical
advancement cost,
M
(user
supplied
input)
CFI = additional facilities cost, $M (user supplied input)
The
ship cost,
life-cycle
value
of C Facq/F
represents an adjusted lead or
first
and is probably the most important input to the ASSET
cost model
1). This cost is intended
(Ref.
to capture
the recurring production cost of the lead ship.
The technical
advancement cost is the cost of developing
any
technological advances that will be necessary before the ship can
be
bui It.
The cost of additional facilities refers to the construction
of facilities
such as shipyards
and piers that will be
necessary
for the support of the ship.
Test
ship
and evaluation
costs are computed as the sum of
test and evaluation cost,
the
CSTE($M), the payload test and
169
evaluation cost, CpTTE ($M), and other test and evaluation costs,
COTE
($M);
C
0.0(012 ( 0.499 C
STE
CpTTE
+ C
/F + 0.647 C
Facq
Lp
)
OAS
(35.2)
1.2 CPHTE
COTE = 1.2
(0.02
CF I )
where CPHTE = payload hardwere test and evaluation cost,
M
(see Eqn 3.33)
COS
=
total operations
(see
and support
cost,
$M
Eqn. 3.49)
The payload hardware test and evaluation
cost may be input
by the users or estimated by the following equation;
CpHTE = 0.0646
(3.33)
WMp
The total research and development phase cost,
CRD (*M),
is
given by the following equation;
CRD = CSDD + CODD + CSTE + CTTE
Investment
support
costs
are
equipment cost,
spares and repair parts,
underway
replenishment).
made up of
+ COTE
prime
cost of facilities,
(3.
4)
equipment cost,
cost of initial
and the cost of associated system (e.g.
Investment
acquisition costs of these equipments.
has two distinct components,
costs
are
primarily
The prime equipment
the
cost
the ship prime equipment cost, CSpE
1 70
($M), and the payload prime equipment cost,
CppE ($M), given by
the following expression;
5PE
( (C
SPE
q/F) /1000o) ) N (ln(2RL )/ln(2))
Facq
(3.35)
CppE = ( CLp + (NS -
where
1) CFp )/
1000
RL = average learning rate
NS = total
number of ships to be acquired
CLp = lead ship payload
cost, $K
CFp = follow ship payload cost, $K
ln = natural logarithm
An average
of
the
ship cost,
C avg
($K),
is computed
ship prime equipment investment for use by
as a function
other
life-
cycle cost algorithms;
Cavg
10C)C)0
(CSp
/NS
E+Cpp
E)
The
equipment
C
cost of support equipment is divided into ship
cost,
CSSE ($M),
and payload support equipment
(3.36
support
cost,
($M), such that;
CSSE = 0.05 CSPE
(3. 37)
CPSE
=
2c
CPp'E
171
The
costs of initial spares for the ship,
for the payload, C
P
CS
($M),
S
and
($M), are given by;
166SP
C S S = 0.03 CSE
(3.38)
C IS
P = 0. 13 C.FE
SP
'E
The final investment cost to be estimated is the associated
systems investment cost, CIASOC ($M);
C IASOC =
25 N S H (WS +
. 1 Wp) x
(3.:39)
(CURU/UCAP) (L/30)
where
H = ship operating hour per year, HR/YR
WS = ship fuel rate, LTON/YR
Wp = payload
CURU
=
cost
fuel rate, LTON/YR
of underway
replenishment
unit,
$M
(user
supplied input)
UCAP = underway replenishment unit fuel capacity, LTON/YR
LS = ship service life, YRS
The
cost of underway replenishment is the acquisition
of each underway replenishment ship.
unit
The underway
cost
replenishment
fuel capacity is the weight of fuel that a typical underway
replenishment
associated
ship can deliver per year to the ship
systems
cost
accounts for the cost
of
fleet.
The
buying
and
operating UNREP ships over and above the support forces currently
operated by the USN.
172
The
total investment phase cost is given by
the
following
sum:
CINV
=
(CspE + CppE
)
+ (CSSE
+
CFSE) + CFI
+
(3.40)
(CiS
S + CS P ) + CiASOC
Operations
of
seven
maintenance
spares,
and support (O&S) costs are computed as the
components:
costs,
personnel
energy (fuel) costs,
major support costs,
data base
costs,
operations
cost of
costs,
replenishment
and associated systems costs.
sed to estimate the operations,
sum
The
maintenance and major
support cost CERs can be seen in Table 3.16.
Personnel
CpAY
costs are divided into pay and allowances
costs,
($M), and temporary additional duty pay costs, CTAD ($M);
CpAY = ( (26,184 NC1 ) + (11,510C(NC2+ NC )) ) x
(NS L S / 106)
C
.65(40) ( (Nc1 + NC2 + N)
Operations cost, COOS
COpS
(3.41)
+
(Nc1 + NC2 + NCS) -
(0.4459 H/12) + (1000 CT)
where
)/106
($M), is given by the following;
NS L S (188 + 0.0013 C
2.232
(N L)
(3.42)
) /1000
CAT
C = annual cost of training ordnance, $M
rn-cC
73
Maintenance
(OOOs)
Operations
(OOOs)
(SFY81)
Ship Class
fY81)
CG26
C616
D0637
D00631
D0062
D00931
D00963
LST1179
LPD1
LSD28
LPD4
ULA113
LHA
LSD36
LCC19
LPH2
AE26
4,689
1,212
4,770
3,888
5,661
5,592
1,212
4,139
858
934
715
AE22
AOEI
AFS1
AOR1
ASR21
ATS1
FF67
FFG1
FF1052
FF1040
FF1037
CVS9
3,268
2,926
3,884
5,389
3,847
2,264
8,963
4,214
3,899
5,047
3,705
3,811
5,685
4,670
3,676
3,141
2,908
1,176
1,014
1,095
1,055
857
1,055
989
2,724
1,055
1,712
1,779
701
745
1,440
998
1,181
448
454
796
796
3,999
4,721
3,646
5,253
3,896
22,331
·
729
729
729
7,659
Major
Support
(000s)
{SFY81)
3,184
3,089
3,105
2,819
3,271
2,733
2,492
1,707
2,839
2,743
2,894
1,703
6,425
2,547
4,652
4,192
2,460.
2,295
3,996
3,058
2,893
1,489
1,103
1,422
2,555
2,377
2,724
2,138
19,546
OOOs)
$FY81)
1)
Table 3.16
174
Ship
Personnel
266,104
434
271,302
407
213,189
391
120,646
163,863
122,148
347
315
188,649
66,023
174,566
99,466
251
293
221
403
333
Hours
Per
Month
296
373
308
291
319
393
463
198
428
142,732
419
89,968
364,797
92,898
209
338
293
289
287
229
398
216,083
205,666
787
.381
641
115,437.
345
321
565
362
360
330
74,632
286,129
884
427
97,896
120,209
173,863
33,255
124,155
119,543
252
80,840100,008
68,977
1,071,799
2,781
Data for Selected USNShips
Operations and Support
(Ref.
Steaml~
Ship
Cost
420
181
101
201
270
261
204
457
356
371
188
325
404
316
316
236
364
425
The
annual
cost of training ordnance may be input
by
the
user or estimated using the following expression:
CATC = 0.0051
The
as
CFp /1000
(5.43)
decrease in operations cost with ship operating
shown
in Eqn (.42),
reflects
the increased
hours,
operations
cost
associated with being in port.
Maintenance
costs,
CMTC
($M),
are
estimated
by
the
expression:
MTC
(NS L S ) (2967 + 0.0064 C
+
4.814 (NC1 + NC2 + NC 3 ) -
(3.44)
(3.939 H/12) + (10 HDEF /168) ) /1000
where HDEF = deferred maintenance manhours, HRS
The deferred maintenance manhours is the average number of
additional
maintenance
manhours per ship that will
have
supplied each week by a shore based facility to perform
to
be
required
preventative and corrective procedures. A value of zero indicates
that
ship's
crew
can
perform
these
tasks
between
overhaul
periods.
Eqn (3.44) indicates that maintenance costs decrease as ship
operating
expected
hours
to
increase,
whereas maintenance costs
increase under such
anomaly
reflects
minimal
maintenance
the fact that
circumstances.
historically,
and therefore minimal
175
would
This
ships
be
apparent
requ iring
maintenance
costs,
were
able to operate for longer
The cost of energy
(fuel),
periods at sea.
CEGY
that used by
($M), includes
the ship and helicopters, if any. The expression isi
C
(N
S
EGY
L
S
) (C
FUEL
/10
)
(2240
H/6.8)
x
(WS + O. 1 Wp)
(3.45)
where CFUEL = fuel cost per gallon, $/GAL
The cost of replenishment
CREP = ( (LS
Values of CS
expression
the
-
($M), is8
4) (CISS + CSP ) )/4.0
and CS
S
spares, CREP
P
(3.46)
are calculated in Eqn.
(3.38). This
reflects the assumption that the initial spares cover
first four years of ships operations and that
replenishment
spares are required at four year intervals thereafter.
Major support costs, CMSP ($M), are calculated as follows:
(NS L )
C
MSP
S
S
(698 + 0.00(:))22
Cavg +
avg
(3.47)
5.988 (NC 1 + NC + NC) ( 1.158 H/12) )/1000
The associated systems cost, COASOC
COASO C = 0.25 NS L S H (WS +
where
CAS
=
($M), is;
. 1 Wp)(CAS/UA
annual cost of operations and
underway
replenishment unit,
input)
176
support
(3.48)
P
of
each
$M (user supplied
The total operations
and support cost,
COA S
($M),
is given
by the following expression:
COA
S = CA Y + CTAD + COF
S + CTC + CMTC
CREP
+ CMSP + COASOC + CEGY
(5.49)
Total systems costs are computedin the next step. The total
systems costs equals the research and development,
investment,
and operations and support costs minus the residual value of
platforms
costs
at
are
the end of their useful life.
computed
both
in terms
of
the
The
total
systems
constant
year
dol 1 ars
(undiscounted) and discounted dollars.
The
residual value of the platform,
R ($M),
is
estimated
using the double declining balance method (e.g., see Ref. 1):
= CSpE
This
decline
(1 -
depreciation
/L S
(5.50)
)L S
method
most
closely
approximates
the
in value of capital assets which are not worn out at the
end of a nominal service life, but are capable of being restored,
modernized and/or converted to other uses (Ref.
The undiscounted life
ships,
CLIFE
cycle total systems cost
for
the
($M), is given by the equation:
CLIFE
The
1).
discounted
CRD + CINV
+
(3.51)
COAS -
cost of a given phase is a function
non-discounted cost and a discounted cost factor.
177
of
the
The discounted
cost
factor
the
is calculated
cost phase,
B,
as a function
of a beginning
and an ending year,
E.
year
The beginning
of
and
ending years are assumed to be integers and are normalized to the
base
year.
The
discounted
cost factor,
Fd '
is given
by
the
following equation:
F
-
+ 1) (-1)
(E -
(1/1.1)
+ (1/1.1)
+
... + (1/1.1) E-1+ ( 1 /1. 1 )E )
(3.52)
This expression assumes that the costs associated with
particular phase (i.e.,
end
of
payment
In
each
each
R&D, investment, O&S) becomes due at the
year that the program runs and
that
each
yearly
is equal to the average cost/yr over the program length.
addition,
Eqn (3.52) assumes a constant discount rate of 10%
for all discounting calculations.
The
ending
and
development phase,
beginning
ERD and BRD,
years
for
the
research
and
respectively, are calculated as
foll ows:
ED
= YI0C + 2 - YB
RD
(3.53)
BRD
where
=ERD - LRD +
YB = the base year (ASSET default value of 1981)
LRD = length of research and development phase, YRS
YIOC = year of initial operational
The
end
capability
of the R&D program is assumed to occur
after the IOC date.
178
two
years
The
C*RD
discounted
($M),
is given
C
where
FdRD
cost of the research and development phase,
as:
dRD
=
RD
discounted
development
(3.54)
cost
factor
for
the
research
and
phase using the beginning and ending
years calculated in Eqn. (3.53)
The
beginning
and ending years for the
investment
phase,
BINV and EINV, respectively, are calculated belows
ERD
R +
BINV
(3.55)
EIN
V = BIN
INY
NYV + (N /Rd ) + 0.9999 where
1
Rd = ship production rate
If
the
value
of
EIN V is
a
non-integer,
the
value
is
truncated to the next lower integer. The length of the investment
phase
is
ships,
the number of years required to construct all
including
the lead ship,
at the given production
of
the
rate.
Also, no overlap between the R&D and investment phases is assumed.
The
($M),
discounted
is calculated
C
where
FdINV
cost of
the
investment
phase,
from:
NV = F
C
INV
dINV
INV
=
C* INV
(3.56)
discounted cost factor for the
using
in Eqn.
investment
the beginning and ending years
(3.55)
179
phase
calculated
The
beginning
and
ending years
support phase, BOAS and ES,
OAS
E INV
B
for
the
operations
and
respectively, are given as:
1
(3.57)
O&S
EOAS
BOAS + LS
costs
are not charged until all the
constructed,
and
the operating
period
ships
have
been
is set equal to the
ship
service life.
The
discounted
C*OAS
S ($M)
support
phase,
is;
S = Fdo S C 0
C
FdOAS
where
cost of the operations and
=
(3.58)
discounted
support
cost factor for
phase
the
operations
using the beginning
and
and
ending
years calculated in Eqn. (3.57)
The discounted residual value of the platform,
given
R* ($M),
is
by:
R*
f
R
(1/1.1) (EAS
+
1)
(3.59)
where R = the non-discounted residual value from Eqn. (3.50)
The
total discounted life-cycle systems cost,
C*LIFE
LIFE ()
is given by the following equation:
C*
Cost
dollars
of
LIFE
C*
RD + C* INV + C OAS - R
outputs from
any year.
(3.60)
the module may be expressed in terms
Because all
1 80
cost equations
used by
of
the
module
reflect
dollars of the year
1981,
cost figures
must
be
adjusted to account for inflation occurring between the base year
and 1981.
Table
actual
3.17 indicates the calculated escalation indexes
and
convert
projected
to
base
inflation
year
rates from
dollars,
the
1977
1981
to
dollar
for
1996.
To
figure
is
multiplied by the appropriate base year index.
The Costing and System Design Office at DTNSRDC has recently
modified
the life cycle portion of the ASSET cost module.
improvements
relaxation
include
a
greater
emphasis
of some of the simplifying
on
assumptions
These
discounting,
in the original
discounting analysis, an ability to compute and display cash flow
diagrams,
and
escalation
of
the ability to study the impact of relative
particular
cost categories
such
as
energy
cost
or
personnel.
The
in
improvements are accomplished primarily through changes
the structure of the calculations.
For the
most
part,
the
original module's cost estimating relationships (as documented in
this study) are retained.
181
-L
-
I
INFLATIOI RATE S
YEAR
.6722
.7287
.8001
.8995
1981
11.30
1.0000
1982
1983
1984
1985
1986
8.50
8.90
7.50
7.60
7.60
1.0850
1.0816
1.2702
1.3667
1.4706
1987
7.60
1.5823
1988
1989
1990
1991
1992
1993
1994
1995
1996
7.60
7.60
7.60
7.60
7.60
7.60
7.60
7.60
7.60
1.7026
1.8320
1.9712
2.1210
2.2822
2.4557
2.6423
2.8431
3.0592
1)
Table 3.17
182
--
I
FACTOR
5.87
8.40
9.80
12.30
(Ref.
~ ~ ~ ~
II
1977
1978
1979
1980
ASSETUSNShip Construction Escalation Indices
-~~~--
I··.
·
IR
3. 1 . 3
Ot5
Information
Input information required by the ASSET cost module has been
divided
into ship acquisition
plus payload cost parameters
and
life cycle cost parameters. These parameters are grouped into the
common
cost driver classifications outlined in Table
acquisition
plus
payload
costs
are the
costs for the lead and follow ships
so-called
1.5. Ship
"sai1away"
(e.g., see Eqn. 3.29).
TECHNOLOGY
Ship type
(i.e., SWATH,
Hydrofoil,
etc.)
Hull & Superstructure material
Propulsion plant type
Electrical plant type
Control systems complexity (*)
Electronics complexity (*)
Weapons complexity (*)
SHIP DESIGN
Weight limited considerations (i.e., SES vs Monohull)
Amount of trials
testing
(**)
Hull structure
weight (SWBS group 100)
Propulsion plant horsepower
Propulsion plant weight (SWSS group 200)
Electric plant weight (SWBS group 300)
Command & Surveillance systems weight (SWBS group 400)
Auxiliary systems weight (SWBS group 500)
Outfit & Furnishing weight (SWBS group 600)
Armament weight (SWBS group 700)
Margin weight (SWBS group MOO)
(*) simple, modest, complex
(**) limited, extensive
ASSET Sailaway Ship Cost Parameters
Table 3.18
185
-
PAYLOAD DESIGN
Number of helicopters
Costed military payload weight
MANUFACTUR ING
Tooling complexity (*)
Learning rate
PROGRAMMAT
IC
Profit fraction
ECONOMICS
Base year $
ASSET Sailaway Ship Cost Parameters
Table 3.18 (continued)
184
Inut ted Costs
Lead ship acquisition cost
Lead ship payload cost
1st follow ship acquisition cost
Follow ship payload
cost
SHIP DESIGN
Ship fuel rate
Crew accommodations
PAYLOAD DESIGN
Costed military payload weight
Payload fuel rate
MANUFACTUR
I NG
Learning rate
PROGRAMMATIC
R & D program length
Technological advancement costs
Payload testing and evaluating costs
IOC (initial operational capability) date
Number of ships required
Ship production rate
ASSET Life Cycle Cost Parameters
Table 3.19
185
ECONOMICS
Base year $
Inflation rates
OPERATIONS AND SUPPORT
Additional facility costs
Deferred manhours required
Annual cost of training
ordnance
Underway replenishment ship operating and support costs per year
Underway replenishment ship acquisition cost
Underway replenishment ship fuel delivery capacity per year
Annual operating hours
Service life
Fuel cost/gal
ASSET Life Cycle Cost Parameters
Table 3. 19 (continued)
The inputted cost information required to estimate the fleet
life
cycle
costs
calculations
for
can
be
obtained
from
lead and first follow ship
the
previous
construction
ASSET
plus
payload costs.
The ASSET cost module makes use of several default values or
expressions
to
representative
facilitate
values
to
ease
of
operation
the program.
and
to
Table 3.20 lists
cost parameters and their default values or expressions.
186
supply
these
ASSET Value
ASSET Parameter
or Expression
Year $
Inflation
1981
Rate
Table
Production Rate
4
5 ships/year
Learning Rate
0.97
Fuel Cost
$1.20 /gallon (*)
Payload Testing
and Evaluation Cost
$0.0
Lead Payload Cost
Eqn.
(12)
Follow Payload Cost
Eqn.
(21)
Annual Training
Ordnance Cost
$0.0
Payload Fuel Rate
0.334 LTON/hr/helo
(*) in 1981 dollars
ASSETParameter Default Values
Table 3.20
187
ASSET Value
or Express ion
ASSET Parameter
R&D Program Length
0 years
Number of ship
Acquired
25
Profit Fraction
0.15
Service Life
30 years
Annual Operating Hours
2500 hours
Technology
Advancement Cost
$0.0
Additional Facilities
$0.0
Cost
Deferred Manhours
Required
0.0
UNREP Unit Capacity
258,585 LTON/year
UNREP Unit Cost
$120.209 Million
(*)
UNREP O&S Cost
$14.656 Million
(*)
(*) in 1981 dollars
ASSET Parameter
Table
Default Values
3.20 (continued)
188
5.1.3.4
Output Information
Output
breakdown
information
of
lead
from the ASSET cost
and follow ship
model
acquisition
includes
costs,
and
a
a
breakdown of life cycle costs.
The
ship
acquisition
costs
consists of the follow:
ing
output
cost variables and their applicable equations.
Cost Variable (Eqn #)
Cost Category
Lead
Follow
Ship
Ship
Construction
SWBS Group
100-900
Ci
Ci
(3.22)
C.152)C
(3.24)
Margins
C
(3.22)
CM
Total Construction
CCC
CC
Profit
Cprofit
Construction Price
Miscel 1aneous
Change Orders
NAVSEA SuoDort
-
r-
-
PBC
C
C
same
(3.25)
(3. 26)
(3.27)
cord
nsea
same
same
same
same
ASSETShip Sailaway Output Summary
Table 3.21
189
(. 24)
Cost Category
Cost Variable (Eqn #)
Lead
Ship
Follow
Ship
Post Delivery Charges
Cpdel
same
Outfitting
CoLtf
same
H/M/E + Growth
Chmeg
same
Total Acquisition
C
Payload
C
_--____________________________________________________________
Miscellaneous (continued)
(3.28)
same
(3.30)
same
(3.29)
same
P
Sailaway
C
ASSET Ship Sailaway Output Summary
Table 3.21 (continued)
190(:
For
the
life
cycle
costs
the
output,
ollowing cost
variables and their corresponding equations are listed below.
Cost Category
NonRecurring
Cost Variable
Ship
Recurring
(Eqn #)
Payload
Other
R &D
Design
& Development
CSDD
CODD
(3. 31)
COTE
(3.32)
Test
& Evaluation
CSTE
CPTTE
CSF'E
SF'E
CPPE
Investment
Pri me
Equi pment
(3. 35)
Support
Equipment
(3.37)
CSSE
Initial Spares
ISS
(3. 38)
CIS
ISF'
Associated Systems
CISOC
IASiOC
Facilities
CFi
** user input value
ASSET Life Cycle Cost Output
Table 3.22
191
(**)
(3. 39)
NonRecurring
Cost Category
Cost Variable
(Eqn #)
Payload
Ship
Recurrng
Other
Operations& Spport
Personnel
Operations
Maintenance
Energy
Replenishment Spares
Major Support
Associated Systems
CFAY
(3. 41)
CTAD
(3.41)
COPS
(3.42)
CMTC
(3. 44)
CEGY
(3. 45)
CREP
(3. 46)
CMSPF
(3. 47)
C
OASOC
Residual Value
R
(3.50)
ASSET Life Cycle Cost Output
Table
. 22 (continued)
192
(3.48)
Cost Category
Cost Variable
(Eqn #)
Total Systems
UnDi scounted
CLIFE
LIE
Discounted
C*LIFE
(3.51)
(3.60)
ASSET Life Cycle Cost Output
Table 3.22 (continued)
193
USCt
3.1.4
3.1.4.1
General
ship cost estimating in the US Coast
Preliminary
a
from
derived
is
to
as
referred
which is generally
procedure
Guard
"Flanagan's Method". This method is based upon an unpublished
The
in 1969 (see Ref.
by Flanagan
report written
in this
described
methodology
information found in Flanagan's report,
study
computer ship synthesis models.
uipon
personnel.
has also been used in
Method
based
is
previous MIT theses, and
from discussions with USCG cost estimating
Flanagan's
12).
Goodwin (Ref.
conjunction
with
7) first used the
method to estimate conceptual and preliminary lead ship costs for
and
rescue and patrol type
between
150
and
400 feet.
program
predicted full load displacements within 4% of
Coast Guard cutter designs.
Goodwin stated that
lengths
with
cutters
search
USCG
synthesis
the
existing
There was no mention of the accuracy
of the cost estimating procedure.
A
synthesis
application
to
150
for
Coast Guard search
and
feet
rescue
for
in
patrol
Since Flanagan's Method was developed for cutters, Tuttle
boats.
(CERs)
from
The CERs were
also
modified
some of the cost estimating relationships
existing
cost
updated
to
data for patrol-sized boats.
reflect more recent Navy information on
propulsion machinery,
see Ref.
changes
in
hull materials and overhead charges (e.g.,
10).
Flanagan's
cost
(Ref. 8)
going boats between 50 and
ocean
especially
length,
was developed by Tuttle
model
Method
is the basis for the current
estimating methodology.
USCG
Changes have been incorporated
194
ship
to
account
for
accounting
the addition of return
methods
cost
data,
and administrative changes
and
that
different
have
been
introduced since the late 1960's. The output from the present day
cost
model
provides estimates ranging from concept
decisions to budget submissions.
195
development
3. 1.4.2
Cost Estimating_Relationships
Flanagan's
Method
is based upon the weights of
If these weights are
standard US Navy weight groups.
seven
the
incorrect,
the estimate will be correspondingly incorrect. Flanagan reasoned
the average dispersion of bids for new
that
since
vary
by 20% about the mean bid,
construction
any technique that can come
to
within 10-20% of actual costs should be considered adequate.
Estimates of the SWBS (Ship Work Breakdown Structure) weight
groups
are used to determine material costs
weight
groups include all government furnished equipment
all
including
In Flanagan's report,
weight
manhours,
agencies
weight.
the following
MHi
1
were
group
weight
a representative plot of material
Fig. 3.10 presents
versus
developed
(GFE),
log-log plots of material cost versus
manhours versus weight for each
and
presented.
cost
government
supplied by other
The
manhours.
US Navy).
the
(e.g.,
material
and
these
From
plots,
CERs for material
Goodwin
cost,
MC
i
(Ref.
($M),
7)
and
($M);
MH
10 l1og(W 1 C) FDG)
MH 2 =
10
.752
+ 2.973]
1og(W2 0 0 FDG ) 1.027 + 2.344]
(3.61)
MH
=
MH4 =
1Clog(Wc)00FDG )
10
1.078 + 2.630)
log(W4 0)0 FDG) 1.078
196
+ 2.6303
0
0
Y=
LOG
0
A
COST
X
=
LOG WEIGHT
Y = A+ BX
Basis of USCG
Material Cost to Non-CostRelationshins
(Ref.
64)
Figure 3.10
197
0
MH5 =
10
FDG) .763 + 3.2373
Clog(W
.974 + 2.7423
[
6
MH 6 = 10 log(W )C)FDG
MH 7 =
1 0 log(W700
MC 1 =
1lo1g(Wl
0
FDG)
.752 + 2.9733
FDG)
.954 + 2.790]
(3.61)
and
MC2 = 10lC
0
(W2
0
1.019 + 3.634)
0( FDG)
1
MC 3 = 10 Elog(W 3 0 0 FDG) 1.073 + 3.6683
MC 4 = input
1
g ( W5 0 0
MC 5
=10
MC6
= log (W
MC 7 = input
where
value
(3. 62)
FDG) .871
FDG)
+ 3.974]
1.068 + 3.60()4]
value
FDG = design and builders margin -factor
(setequal to 1.02)
F 1 = $0 for diesel
= $1,500,000
W.
for
propulsion
CODAG(combined diesel and gas)
= the appropriate SWBS group weight, L.TON
The design and builders
margin of 2%
is distributed
among the seven weight groups that make up the lightship
198
even l y
weight.
As indicated in Fig.
3.10,
the manhour and material cost versus
weight curves are assumed linear when plotted on log-log paper.
Since
inflation
years.
all
material costs curves are based on 1959
prices,
indices must be used to correct prices to more
These
Statistics
indices
(BLS) data
can
be obtained
(e.g.,
from
see Ref.
Bureau
recent
of
Labor
11) for the shipbuilding
industry.
In
cites
order to account for different hull materials,
data from the US Navy and Newport News
indicates
into
that
structure
HY-SO
Shipbuilding
HY-80 requires 40% more man-hours
than mild steel.
In
Flanagan
to
fabricate
1969, the material
was 3.4 times that for mild steel.
that
cost
of
Ship estimates must be
adjusted to reflect the percentage of structure which is made
of
high-strength steels (HSSs).
For aluminum structures, US Navy and Maritime Administration
(MARAD) data
from
aluminum structure
times
the
However,
cost
has a total
(material
of the same weight
Flanagan
contracts
June 1969 indicate that a
notes
of
given
weight
+ labor)cost 1.8-2.2
mild
steel
stru(Cture.
that computations from WMEC
and
more recent Navy information on changes in hull
Tuttle
for
WHEC
ignored the presence of aluminum in their construction
estimates -without introducing any significant additional
From
of
found that the effects of all aluminum hull
patrol
boats
could be accounted
material costs only, by 2.192 (see Eqn.
Material
costs
for
by
error.
materials,
construction
multiplying
the
.64).
for command and control equipment
and
for
armament (SWBS weight groups 400 and 700, respectively) are input
directly into the model.
Better estimates are generally obtained
199
when
there
particularly
exist
good
electronic,
shopping
lists
for
material
armament and machinery
using a shopping list in conjunction with Eqns
weight used for the labor estimate
weight group,
particular
item
the
equal to the total for
the
that used in
must be reduced by the equipment
calculation
the
remains
(3.61)
Although
not to "double-bill".
it is important
(3.62)
the
year
price,
and weight.
of purchase,
When
The
hardware.
minimum cost data required for any equipment is the
and
costs,
cost
material
weight.
The cost of
is then added to the revised material cost
calculated
from Eqn (3.62).
(3.61) and (3.62) were developed by
Eqns
from
report.
Flanagan's
directly
Tuttle modified some of these CERs
order for the results to be more consistent
in
Goodwin
the 50-150 foot length range.
Eqns.
with CG patrol
in
boats
(3.63) and (3.64) show
these adjusted relationships.
For calculating manhours, only the equation for armament was
so that:
modified,
MH 7 =
10
og(W70
material costs,
For
equipment
propulsion
FDG ) .954
+ 2.7903
(3.63)
structural
the relationships for
costs were updated using more recent
and
Navy
data from Ref. 10:
Mc
=
10
log(W 1(
FDG) 1.019 + 2.7903 F
1
2
(3.64)
MC
2
= 10 Clog(W 2 0 .- FDG ) 1.019 + 3.634
F3 + F
3
4
where
F 2 = 1.0
for steel construction
= 2. 192
F3 = 1.0
for aluminum construction
for diesel prime mover or CODAG
= 1.57
F 4 = $0
for gas turbine prime mover
for diesel and gas turbine
= $1,500,00
PM
for CODAG
For calculating tne labor cost, the labor rate is assumedto
be the same for all weight groups. Therefore, the labor cost, LC.
($M), is given
by the simple expression:
LCi - MH. $/HR
1
where
(3.65)
1
MH 1 = estimated manhours from Eqn. (3.61) or (3.63)
$/HR = labor rate (dollars per hour) , adjusted
to the
appropriate year of shipbuilding
For
the lead ship,
the labor and material
costs
of
and engineering services (i.e., SWBS weight group 800),
design
LC 8 and
MC 8 ($M), respectively, are given by the following expressions:
LC
= (.114 LOA - 1.6)(LC
+ ... + LC 7 ) / 100
(3.66)
MCe = (-.01 LOA + 7.5)(MC1 + '
where
LOA = overall
ship length
+ MC7) / 100
(ft)
The percentage calculated in Eqn. (3.66) for
the
costs
associated with design and engineering services assumes that
contract
design
is
available to the shipyard at
2( )1
the
time
the
of
construction.
The
services
labor
and
material
(i.e., SWBS
costs for lead
weight
group 900),
ship
LC9
construction
and
MC9
(sM),
respectively, are as follows:
LC9 = (.014 LOA + 7.0)(LC1 + ... + LC 7 ) / 100
(3.67)
MC, = (-.015 LOA + 6.5)(MC1 + ...
For
vessels
under
+ MC7) / 100
approximately 160
ft
in
length,
the
material cost for SWES weight group 900 is estimated at 4% of the
sum of the construction material costs.
Overhead
cost,
OVi
($M),
is calculated
from the labor cost
of each SWBS weight group by the simple relationship:
OV.
1
where
F
ovhd
LC
(3.68)
1
F ovhd = labor overhead fraction
The
cost
of construction
($M), is the sum of material,
for each SWBS
MC,
weight
group,
C1
and labor, LC i, plus overhead,
OV i , costs and can be written as:
C. = MC
+ LC
(3.69)
+ OVi
The ship construction
cost, CCC
($M), is given as the sum of
the weight group costs, such that:
CC
Profit,
C1 + C
+
C+
C9
profit ($M), is assumed to equal a fraction
Cprofit
(3.70)
of the
construction
cost
and the price or
base
estimate,
PBC
($M),
equals the construction cost plus profit. As a result:
Cprofit
and
PBC
=
CCC
Fp = profit
where
A series
price
(3.71)
= Fp CCC
+
(3.72)
Cprofit
fraction
(set equal to 0.15)
nd added to the
of add-on costs must be calculated
the "sailaway"
of the ship to determine
cost of the
ship.
These costs are determined by the following expressions:
= 0tf
.02 PBC
C
Cretro = 0.04 PBC
(.3. 7)
C
C
where
= input value
misc
spare
Cutf
= .09 MC 2 + .08 MC
+ .25 MC
+ .10 MC 5
= initial outfit costs ($M)
Cretro = retrofit costs (SM)
Cmisc = miscellaneous and contingency costs (M)
C
spares = cost of spares
Instead
modifications
accumulates
of
decided
some
construction period.
this
issuing
(M)
change
upon during
orders
ship
for
design
all
construction,
the
CG
throughout
the
Retrofit costs are the costs incurred
when
of
these
modifications
work is done by CG shipyards immediately after the ship
built.
2 C)3
is
Miscellaneous
design
work
automation,
technical
initial
and
studies,
contingency costs include such items
training
specialized
services.
ship
use
training requirements,
Other
load
mock-ups,
of
costs under this
all fluids and
the
as
of
large-scale
and
additional
category
inclusion
are
of
the
cost
margins.
Administrative costs,
Cad
admin ($M),
are estimated from the
following equation:
C
+
Cadmin =F
Fadmin (P
PBC
Cretro +
Coutf
C
Cmisc + C spares
where
Fadmi n = administrative
POGA
=
cost
cost fraction
(including profit)
P OGA )
(3.74)
(set equal to 0.035)
of material
supplied
by
other government agencies
Administrative
personnel
attached
costs
to Resident Inspector Offices (RIOs) and
at
other related
expenses.
claims,
the
office equipment
purchases,
on a particular
project
supplied
are known,
this portion
of
command
and
the
into the estimate.
by other
government
agencies
(OGAs)
the US Navy contributions to SWBS weight groups 700
weapons
and
When the size of the RIO and HO personnel
cost can be input directly
include
salaries
CG
travel
Equipments
made up of the
of
headquarters,
working
are
control portion
of
the
400
Included with these items are the spares supplied by OGAs.
204
and
group.
The "sailaway"
cost of the ship, C
($M), is given by:
Cs - PBC + Coutf + Cretro+
Cmisc + Cspares + Cadmin
The
(3 75)
lead ship cost for the CG would be equal to Eqn.
actual
(3.75) minus POGA'
the cost of material from
other
government
from
multi-ship
agencies.
For
contracts
follow
ship
for
that
costs,
Flanagan
found
succeeding ships there was a
reduction
total labor manhours according to the schedule in Table 3.23.
PERCENT DECREASE
IN LABOR FROM
SHIP #
1
PREVIOUS SHIP
Eqns.
(3.61)-(3.67)
2
10
3
5
4
2
5
2
6
2
7
1
8
1
> 8
same as 8th
Follow Ship Reduction in Total Labor Manhours
Table 3.23
20(5
in
This
delivered
little
reduction
in labor accounts for learning.
more than six (6) months apart,
learning
For
ships
Flanagan states
occurs other than for design
and
tnat
construction
services.
For
design and engineering services on
follow
ships,
the
percentage of costs in Eqn. (3.66) becomesa
LC
= (.025 LOA) (LC1 +
' + LC7) / 100
(3.76)
MC 8 = (-. 0(:02
LOA + 2) (MC+
Eqn.
(3.76) accounts
for
+ MC 7 ) / 1 :
...
a substantial
portion
of
the
reduction in labor noted in Table 3.23.
For construction services on follow ships, the percentage of
costs
in Eqn.
(3.67) is modified
to:
LC9 = (.022 LOA + 1.5)(LC1 + ...
+ LC7 ) / 100
(3. 77)
For
material
MC9 = (-.0125 LOA + 5.5)(MC + ...
) /1oC
+ MC7
vessels
ft
cost
under
approximately 200
of construction
services
in
length,
is estimated
at 3.C)%
the
of
the sum of the construction material costs.
Regardless
of
the
magnitude of
the
reduction
in
labor
calculated using Eqns.
(3.76) and (3.77), the total reduction in
1abor
ship must be consi stent with
for
each follow
the
val Les
found in Table 3.23.
Apart from any differences associated
contingency costs,
with miscellaneous and
Flanagan noted the following
22 -)6
adjustment to the
cost of spares for follow ships:
Cspares = .045 MC 2 + .07 MC
+ .23 MC 4 + .1 MC 5
(3.78)
Administrative costs are estimated in an identical manner as
for the lead ship, using Eqn. (3.74) with appropriate follow ship
costs.
cost of the follow ship, C
The "sailaway"
PBC is the price of the follow ship adjusted
(3.75), where
Eqn.
the required reduction in manhours.
for
($M), is given by
ship cost for the CG would be equal to Eqn.
follow
POA,
the
As before,
actual
minus
(3.75)
the cost of material from other government agencies for the
The
cost
of OGA material includes
follow
ship.
portion
of the cost of spares calculated
in Eqn.
relevant
any
(3.78).
previous discussion was based entirely upon the
The
by
written
Flanagan in 1969.
estimating personnel,
USCG
After discussion with
the following changes and/or
report
cost
explanations
were noted as being applicable to the CG model as it now exists.
The procedure for estimating the ship price or base estimate
the
variables
used
for -calculating material and labor costs may be
different
than
those listed in Eqns (3.61)-(3.67). This is due mainly
is
essentially correct.
The numerical values of
to
additional data becoming available in recent years.
effects
The
related
into
learning on the
to the rate of production.
the
between
of
learning curve,
construction
base
This introduces
so that learning
starts
estimate
increases.
For
cost
a time factor
decreases
as
time
periods
of
time
greater than 2 years, learning is taken to be non-existent.
207
are
Return
data
from recent multi-ship acquisitions
indicate,
that for CG vessels, it is more appropriate to apply the learning
curve to the labor plus material costs.
The
categories
under
which costs are
grouped
have
been
changed from those reported by Flanagan to reflect differences in
current
cost accounting procedures.
These categories are listed
in Table 3.24.
Categories
Contract Award
Contract Spares
Contract PTD
Escal ati on
Related Costs
Outfitting
RIO
Retrofit
Ship & ICP Spares
Self Insurance
Training
USCG Cost Model Categories
Table 3.24
208
The
costs
categories of contract award and PTD and related
require some explanation. The dollar amount of the contract award
is
PBC in Eqn
to
equal
accounts
and
documentation
technical,
provisioning,
to
refers
PTD
(3.72).
for
of
most
the
miscellaneous costs (e.g., Cmisc )
The
related costs listed in Table 3.24 make up the rest
the add-on costs
(inventory
similar
listed
control
(3.74). Ship and ICP
spares are calculated
point)
to that outlined
(3.73) and
in Eqns.
in Eqns
(3.73) and
in
a
manner
(3.78). RIO costs are
Inspection
the administrative costs associated with the Resident
Offices at private shipyards.
of
The related costs are estimated to
be some percentage of the contract award amount.
A
simple
analysis
risk
has been
incorporated
into
the
present model by presenting a range of cost estimates. This range
of
values
is
economic (i.e.,
for inflation
obtained
by varying
design
forecast inflation) inputs.
rates
of
%,
forecast,
(i.e.,
and
Costs are estimated
and 6% and SWBS weights of
10% below estimated, estimated, and 25%/.above.
209
weight)
3.1.4.3 Input Information
following input information is required
The
Method
for
by
the estimation of CG ship acquisition
costs.
These
equal to the "sailaway" cost of the lead
and
follow
costs
are
ships
minus the cost of equipments supplied by other
agencies
into
the
Flanagan's
(e.g.,
government
The parameters below are
the US Navy).
common cost driver classifications outlined
grouped
in
Table
1.5.
TECHNOLOGY
Ship type (e.g., cutter or patrol boat)
Hull & Superstructure material
Propulsion plant type
SHIP DESIGN
Hull structure weight (SWBS group 100)
Propulsion plant weight (SWBS group 200)
Electric plant weight (SWBS group 30)
Command & Surveillance systems weight (SWB.Sgroup 400(:)(
Auxiliary systems weight (SWBS group 500)
Outfit & Furnishing weight (SWBS group 600)
Armament weight (SWBS group 7)(:))
Design & Builders margin
Overall ship length
PAYLOAD DESIGN
Cost of payload material
(includes armament and weapons portion of com. & surv.)
USCGShip Acquisition Cost Parameters
Table 3.25
21
MANUFACTURING
Learning rate
Labor rate
PROGRAMMATIC
Profit fraction
Overhead fraction
ECONOMICS
Base year $
Base year for BLS data
BLS material indices
USCG Ship Acquisition
Cost Parameters
Table 3.25 (continued)
211
OuLtpt
lead
used
from
Flanagan's Method gives cost
and follow ships.
to
information
The following additional information
escal ate/di scount
the
CG
costs
throughoLut
for
is
the
construction period to a common year dollar value.
Iputted Costs
Lead ship acquisition cost
Lead ship payload cost (*)
Follow ship acquisition costs
Follow ship payload costs
FROGRAMMATIC
Contract award date
Length of constrcttion
Time between ships
Number of ships required
ECONOM ICS
Base year $
Inflation rates
(*) payload refers to material cost (plus profit)
weapons portion
700 & 400) -
of command and surveillence
USCGFleet Escalation/Discounted
Table 3.26
212
of armamentand
(SWBS weight
Cost Parameters
groups
3.1.4.4 Outpt
Output
breakdown
Information
information
from Fl anagan's
of lead and follow ship material,
Method
includes
labor,
overhead
a
and
total acquisition costs.
Cost
Category
Cost Variable (Eqn #)
Follow
Lead
Ship
Ship
Construction
Material Cost
SWBS Group
100-200C
(*)
(3.62 or 3.64)
MCi
1
same
300-700
MC. (3.62)
1
800
MC
8
900
MC 9 (3. 67)
MC 9 (3.77)
MH
(5.61)
same
MH 7
(. 61 or 3.63)
same
(3.66)
same
MC8 (3.76)
Labor Manhours
SWBS Group 100-600
.700
(*) all
construction costs include the design and builders margin
USCG Ship Output
Table 3.27
213
Summary
Cost Category
Cost Variable (Eqn #)
Lead
Fol 1 ow
Ship
Ship
Labor Costs
SWBS Group
100-700
LC.
(3.65)
same
800
LC
(3. 66)
LC
(3.76)
900
LC9
(3.67)
LC 9
(3.77)
1
Overhead
SWBS Group 100-900
OVi (3.68)
same
SWBS Group 100-900
Ci1 (3.69)
same
Total Construction
CCC (3. 70)
same
Profit
C
same
PBC (3.72)
same
Material + Labor + Overhead
Total
profit (3.71)
Ship Price
USCG Ship Output Summary
Table 3.27 (continued)
214
Cost Category
Cost Variable (Eqn #)
Lead
Ship
Follow
Ship
Outf it
Coutf (3.75)
same
Retrofit
Cretro
Mi scel 1 aneous
Cmisc
.
(.73)
same
Spares
Cspares (3.73)
C
Administrative
Cdi
admi
same
Add-on
n
Sail away
C
s;
(3. 7.3)
(3.74)
(3. 75)
USCGShip Output Summary
Table
3.27 (continued)
215
spares
same
(3. 78)
3.1.5
RCA PRICE
3.1.5.1
General
RCA
cost
PRICE has developed a family of automated,
estimating models.
for Costing
parametric,
PRICE (Programmed Review of Information
and Evaluation)
was originally
developed
for internal
RCA use in the early 1960's. Commercial operations began in 1975,
with
applications
to
hardware
development
software design and implementation,
maintenance and support costs.
Fig.
3.11 gives an indication of
modeling systems.
For ship acquisition cost estimating,
used extensively.
the PRICE H model has
The PRICE H model is applicable
aspects of hardware acquisition,
government furnished,
be it development,
to
all
production,
or modification of existing equipment. The
model estimates costs associated with design,
management,
production,
microcircuits and associated
the diversity of areas covered by PRICE
been
and
drafting,
project
documentation, engineering, special tooling and test
equipment, material, labor and overhead. Also, costs to integrate
subassemblies
into a system
and to test the system for
required
operation can be estimated. A detailed explanation of the PRICE H
model
can be found
PRICE
H
methods (.g.,
a
in Ref. 20.
is characteristic of traditional
of
estimating
NAVSEA, NCA) in that it performs cost estimates on
cost per pound basis.
breakdown
cost
However,
its output does not contain a
material and labor costs.
These figures
must
"backed" out using a post-processor (i.e., PRICE LABOR) or
be
some
other method, as outlined in Section 3.1.5.4.
There
is
strong
resistance
to the
use
of
PRICE
among
traditional "material list and labor" ship cost estimators. There
216
HARDWARE COST ESTIMATING
PRICE H
PRICE HL
PLANNING AND MANAGEMENT
PRICE PM
MICROCIRCUIT COST ESTIMATING
PRICE M
SOFTWARE COST ESTIMATING
PRICE S
PRICE SL
SUPPORTING MODELS
PRICE A
PRICE D
PRICE Parametric m3delina Systems
(Ref. 20)
Figure 3.11
217
are
several reasons for this distrust,
points:
black
(1)
the model
box;
aerospace
(2)
is proprietary
including the
and must be operated
was originally developed
it
applications;
following
for
as
avionic
a
and
(3) the user must pay to use it, and; (4)
it doesn't give material/labor figures.
Despite these reservations,
years
PRICE has been used for several
for early stage estimating of Navy weapons systems.
Also,
the Costing and Design Systems Office, Code 1204, at David Taylor
Ship Research
Naval
the
PRICE
H
and Development
model useful for:
(DTNSRDC)
Center
early-stage
ship
has
found
design
cost
assessment; advanced technology cost impacts; alternative systems
cost analysis, and; RD
PRICE
Journal
modeling
of
systems
Parametrics
International
Refs.
resource planning.
17-19).
are extensively discussed
and
in
the
proceedings
relating
the
of
Society of Parametric Analysts (ISPA)
Articles
in
the
(e.g., see
PRICE to ship costing can
be
found in Refs. 25-27.
Cost estimates obtained using PRICE are
generally
intended
for acquisition planning purposes and not for budget submissions.
Even
a
so,
PRICE can estimate costs at any level of detail,
from
The
DCAA
states
the
whole ship viewpoint down to individual equipments.
(Defense
Contract
"parametric
Audit
estimating
Agency)
provides
completeness of proposal coverage."
21.8
Audit
an
Manual
excel lent
cross-check
on
3.1.5.2
Cost Estimating Relationships
This section is intended to provide a brief overview of
PRICE
H
model.
discussion
available
information
personnel
The
in
that
follows
in the users manual
is
(Ref.
the
based
upon
20)) and
from
the Costing and Design Systems Office at
DTNSRDC.
For additional clarification, PRICE offers users an intensive two
week
training program to familiarize users with the concepts and
use of the model.
and
PRICE H model estimates costs for both development
The
the
program.
Table
production
elements
of
categories
included
under the development and
3.28
the
lists
production
cost
headings and a brief description of their make-up.
Cost
Category
Development
Production
Description
Engineering
- drafting, design, systems
engineering, project management and
data
Manufacturing
- labor and material associated
with prototype production
- tooling & test equipment costs
Engineering
- non-recurring production costs
such as drafting, design, project
management and data
Manufacturing
- production costs
- tooling & test equipment costs
PRICE H Cost Output Categories
Table 3.28
219
The
is
basis for the development of the PRICE proprietary CERs
multiple
result
regression curve fitting of
historical
of this analysis is literally thousands
of
data.
The
mathematical
equations relating the various input variables to cost.
Input
data
physical,
consists of 67 variables used to
qualitative,
programmatic,
describe
economic,
engineering
dependent and system dependent characteristics of the
system.
However,
with
minimal amount of hardware
a
input
variable
makes
the
the
user
model useful
specification
statistical
Of
all
the
information,
since
missing
This
feature
generated.
for cost estimating
It is important
in
to realize
of all the input variables
uncertainty
particular
model has been designed to estimate costs
values are internally
stage of development.
the
the
conceptual
that the proper
will
reduce
the
of the model.
inputs,
the more fundamental
parameters
listed in Table 3.29.
Description
Number of production units built
Learning curve
Integration difficulty
Schedules for development and production
Weights
Amount of new design required
Operational environment
Manufacturing complexity
Technology improvement
Fundamental Cost Drivers in the PRICE H Model
Table 3.29
220
are
Weight
and
cost drivers,
manufacturing complexity are the most
so that in its simplest
powerful
form;
Cost = f (Weight x Manufacturing Complexity)
where
f(
)
applications,
is
a
nonlinear
function.
For
(3.79)
ship
costing
weights are based on SWBS. PRICE can be run at any
level of detail (i.e., whole ship, one-, two- or three-digit SWBS
levels) provided the necessary information exists.
Separate
manufacturing
mechanical /str ctural
applications
are
and
almost
complexities
electronics
are
computed
items.
exclusively
based
Ship
for
costing
estimated
upon
mechanical/structural complexities (MCFLXS).
MCFLXS
can
manufacturing
be
thought
processes
engineering
work.
Typical
Fig.
and
in
terms
of
a
cost/lb
It exhibits a wide range of values
the
for
and a cost/drawing or required effort for
upon the technology required
environment
of
for its fabrication,
employment history
values for a variety of non-ship
of
the operating
the
systems
depending
manufacturer.
can be found in
3.12.
An important trend to note from Fig. 3.12 is the increase in
manufacturing
complexities
requirements.
Fabricating
standards
for
with
severe
more
and
specifications
are more stringent for space applications
ground assemblies.
operating
specification
environment
level
The number associated with a
is
(FLTFM).
referred
to
The variable
as
reliability
than
those
particular
platform
WSCF is equal to the
density of the structure in pounds per cubic foot.
2~ 1
the
operating
Equipments
TypicalExamrnples
WSCF
1.0
1.4
Ground Mobil
1.8
Airborne
2.0
Space
2.5
Manned
Space
Antennas
Drive
Assembliles
Microwave
Transmission
Small,Spiral,Horn,Flush.Perabolic 4
4.75
5.39
Scanning
Raedr10-40'Wide
8
5.3
5.4
Phased
Arrays(Ll Redators)
6.8
5.9
6.2
Autornobile.00 to 400 H.P.
25-35
4.30
TurboJet(PrimePropulsion)
25-35
RocketMotors
14-15
ElectricMotors
75-100
4.47
5.0
MachinedParts,Gears.
etc.
7-10
5.115.24 5.5
Mechanisnsw/Stamnpings
(HiProd)
12
3.33-3.73
Wevu;de, IsolItors Couplers,
11-20 5.4-5.6 5.4-5.6
StriplneCrcuitry
9
5.7
5.
Optics
Good (Commercial)
Engines
& Motors
5.64
5.5
6.4
6.6-7.9
6.16.5
5.3
5.
.
5.5-5.7
59
6.55-7.04 6.92-7.44
7.0
7.2
-
6.47.3
5.46.3
7.28.
5.46.3
-
-
5.5-5.9
6.0
-
5.5-59
61
70-90
5.1
5.4
6.3
6.7
7 3.
Ordnance
Fuze
Servo
Tools
Printed
Excellent(Militery)
Highest(Add 0.1per 10%YIeld)
AutomatedProduction
SmallProduction-Min.
Tooling
Maechrive &Coupliq Networks
MachineTools
PaperPhenolic
CKT Cards
GlassExpoxy, Double Sided
110
5.3
5.3
5.3
5.3
5.3
Cabling
Add 0.1 for Platea.Thru Holes
Multiconductor wiMS Connectors
Same rv/ Hfwmtically Sealed
Connectors
40
40
4.9
5.1
5.0
5.2
5.0
5.2
5.1
5.3
52
53
Nickel- Hydrogen
NickelCadmium
InertialPlatformTypo
80
75
79
5.39
6.01
5.83
6.56
6.
6.73
6.8
7.81
7.63
6.9-9.1
7.6
6.5
9.4
9.5
(BoardsOrdy)
Battery
Gyro
Lr
(Add0 2for3Layers&0.05forAddn'l)
Module
70-90
5.4
5.8
7.3
7.8
8.0
70-90
5.9
6.8
8.0
8.3
85
14-20
4.3-4.65 4.3.465
14-20
5.11-5.33 5.11.5.33 6575
5.63
5.63-5.7 5.76 26 5.7-6.86 .7.6 86
25-30 4.45-4.52
83
4.14.3
4.14.3
4.14.3 4.14.3
4.14.3
Typical MCPLIXS
Factors for Mechanical Assemblies
(Ref. 20)
Figure 3.12
222
8.55
8 38
7.0-9.4
9.5
Ships also experience a range of manufacturing complexities.
Fig.
3.13
several
illustrates
USN
the variation
vessels.
Note
that
of "whole ship" MCF'LXS for
the
cost
per
pound
figure
increases proportionately with MCPLXS. The platform specification
level used in deriving these complexities was estimated at 1.4
for conventional ships and at PLTFM=1.6 for weight critical ships
(i.e., hydrofoils,
The
affect
complexity
of
technology on the
value
boats,
technology
manufacturing
designs done at DTNSRDC (see Ref. 27).
equal to 5.531 and 5.082 were estimated
planing
of
is clearly shown in the comparative cost analysis
alternative WPB
MCPLXS
SESs and ACVs).
respectively.
for the hydrofoil
of
Values of
for hydrofoil
and
The result of the higher level of
was a cost that was 70%
more
than
the planing boat.
The
effects
complexity
on
indicated
costs
vary with time
in Figure 3.14.
expected
and
according
with time.
indicates
technology.
manufacturing
to
and
year
As expected,
of maturity
for
trends
reductions
manufacturing
The year at which the curve
the
the
The curves show the cost
with improvements in design methods
efficiency
abscissa
of the level of technology
crosses
the
the
particular
lower technology matures earlier
than
high tech., but typically has less effect on costs.
Scheduling
effects
production
phases
Penalties
are
on
the costs of
the
can also be explored with
assessed
for
accelerated
schedules as shown in Figure 3.15.
development
the
and
PRICE
and
model.
stretched-out
,P
0.5
_
a
x
I-
0DG993
6.0
w
-a
a.
5.5 _
LS041
·
C.
/A041
~AS33
t
5.0
-
Lid
T-AOT177
O T-AOT168
Ce
a.
T-AFT166
*T-AFT166~~~~ I
10
20
45;
0
00963
j
FFG7
SHIP
TYPE
S/LB.
NICPLXS
CG47
ODG993
FFG7
00963
CRUISER
DESTROYER
FRIGATE
DESTROYER
LS041
AD41
AS39
T-AOT177
T-AOT168
LANDING SHIP
TENDER
SUB-TENDER
OILER
OILER
54.0
42.9
32 3
28 6
138
13.7
111
9.6
86
T-AFT1E6
OCEAN TUG
6.29
6..12
5.82
5 73
5 36
5 30
5.19
5 08
4 96
4 53
20
40
COST PER POUND(S/LB)
MCPLXS
Factors for Selected USNVessels
(Ref.
60)
Figure 3.13
224
53
50
60
.HIGH TECHNOLOGY
w
J
Q
4
z-a
I
4)
a
1970
1980
1990
Effect of Technology Inprovement on Costs
(Ref. 64)
Figure 3.14
225
. 2000
2010
$
$
$
DEV
PROD TOTAL
190
6,887
7,077
190
7,133
7,323
190
7,841
8,031
DEVELOPMENT
PRODUCTION
DEVELOPMENT
PRODUCTION
-~~~~~~~~~~~~~~~~
DEVELOPMENT
PRODUCTION
Influence of Schedule on Costs
(Ref.
64)
Figure 3.15
226
The
PRICE
including:
model
incorporation
production units;
runs;
a
incorporates
of
a
several
learning
other
curve
features,
for
multiple
estimating the cost of multiple lot production
factor to account for component integration difficulty;
allowing
design
costs in proportion to existing
discounting
the
through-put
of vendor quotes
to
data,
and;
double
prevent
billing.
PRICE also has a design-to-cost mode,
to
input
a target cost,
information.
quantities
which allows the user
and level
technology
consists of design limits for weight. This
Output
analysis would be especially beneficial
type of
of
during
concept
development and the early stages of feasibility design.
One
of
widespread
through
Figure
the
use
an
main reasons that the
model
has
is its adaptability to any hardware
easy to use calibration process,
found
such
application
as illustrated
in
3. 16.
Basic
inputs
for calibration consist of cost and
data plus the physical
characteristics
of the system.
schedule
The
model
then
iterates on MCPLXS values until a complexity is
calculated
that
matches the input data.
calculated
when
the
The empirical
factors
are
now
representative of the system and the supporting organization
and
data
is
ECIRPed (PRICE spelled
can be used independently to estimate
Once
a
user
backwards)
costs and schedules.
has performed this process on a
variety
of
related cost data, a pattern usually emerges for the complexities
that
can
effectiveness
be
related
of
the
to
the
system
calibration
227
characteristics.
is highly
dependent
on
The
the
JO
OM
l
l
The PRICECalibration
(Ref. 64)
Figure 3.16
228
Process
availability and accuracy of the cost data used.
Fig.
3.17
(i.e., whole
refers
their
shows
ship) calibration.
value
data
of
PEND
and
respectively,
month/year
(e.g.,
252
is
Costs are given in British pounds.
was
obtained
from
an
(Institute
IDA
were
increase
in
3.17,
Defense
a trend showing
complexity with
manufacturing
production was developed and is shown in Fig.
the complexity has been
that
for
(e.g., see Ref. 44).
study
From the ship data contained in Fig.
shows
box
by Cost and System Design Office personnel at DTNSRDC.
performed
the
1
1 box calibrations for selected US Navy ships
Similar
Analyses)
production,
given in terms
is
a
The variables PSTART and
the start and end of
to
February, 1952).
The
the information required to do
3.18.
steadily
start
the
The
of
figure
with
increasing
time.
The
model
with
conjunction
manufacturing.
uncertainty
of
a
incorporates
the
However,
r i sk
program
in
development
and
analysis
cost
estimates
this
analysis
for
is developed
the PRICE algorithms and not of the input
Understandably,
the
accuracy
dependent
upon the input data.
indicates
that
of
the
model
Bearing this in
is
mind,
the PRICE model produces results within
actual costs.
229
the
using
data.
highly
Ref.
20
10.
of
CLASS
TYPE NAME
COSTM
STANDARDSCHEDULEDLENGTHHP
(TOTAL) (ATPEND)
OISP'T(WT) PSTART-PENO
SALIS8URY 2170
SALISBURY 61
2170
LINCOLN
61
2300
LYNX
LEOPARD 41
2380
BRIGHTON
12
WHITOY
2300
ASHANTI
81
ASHANTI
2450
LEANDER
LEANDER
2500
..
~ ARIADNE
5440
KENT
COUNTY
5440
ANTRIM
..
6100
BRISTOL
82
BRISTOL
2750
AMAZON
21
AMAZON
2750
ANTELOPE
21
".
2750
ARROW
21
..
SHEFFIELD 42() SHEFFIELD 3500
42(i) BIRMINGHAM 3500
42(I) GLOUCESTER 4000
42(II) EDINBURGH 4000
22(i) BROADSWORD3500
BROADSWORD
22(0) BATTLEAXE 3500
3800
22(n) BOXER
".
3800
22(0) BEAVER
".
450
MCMV WILTON
TON
615
MCMV BRECON
HUNT
32 MCMV (PROPOSAL) 120
HYMARINE
4000
SSN WARSPITE
VALIANT
SSN CONOUEROR 4000
..
SWIFTSURE SSN SWIFTSURE 4000
4000
SSN SUPERB
.
SSBN RESOLUTION 7500
RESOLUTION
7500
SSBN RENOWN
.
7500
SSBN REPULSE
"
7500
SSBN REVENGE
..
PRICE H Ship Calibration
(Ref.
25)
Figure 3.17
230
3398
252-257
3398
655-660
3398
853-257
370
857-961
158-1161 360
372
459-363
1169-273 372
520 5
360863
266-770
520 5
1167-373 507
1169-574 384
384
471-775
1072-776 384
412
170-275
372-1176 412
1179-1283 463
463
980-1284
430
275-579
430
276-380
1179-1183' 471
471
680-684
1170-773 150
197
975-380
107
980-683'
1263-467 285
1257-1171 285
272
469-473
372-1176 272
264-1067 360
664-1168 360
360
365-968
565-1269 360
Data
14400
14400
14400
30000
20000
30000
30000
60000
60000
60000
64500
64500
64500
64500
64500
64500
64500
64500
64500
64500
64500
3000
3540
1000
NK
NK
15000
15000
NK
NK
NK
NK
33
33
32
3.6
5 22
4.7
70
13.8
168
27
144
144
20 2
23 2
309
78 5(1980)
850(1980)
68 6
69 2
125(1981)
130(1981)
23
30
135'
24
30
37 1
41 3
40 24
3995
375
386
77 7
Inl
I
r
I
- 1
Iq I I 1-1
'
1 1 I
I]
FRIGATES&
DESTROYERSMCPLXSvs PSTART
r/I:
5.8
6I
5.,4
E
~~~~~BF~~~~~~~
T22(11.
.
I
I
T22()_
I
_
5;2
C.
r
5
.4-142 :, . _/ ' T21II , T21T821
.121 ---1 II
I
_
.
T
. _.^
I
-1
--l COUN'TY -- r ''- - IcIL-e I
I I·
LEANDER
I II
r,
1
D
t-
r
A a:
IE ]
-
AA
-·
7
_
4
~E
54
.-
l PJTV
J
I.1l I
I
1-1
J 1-
.I-
ANDEi
II
l----· -
l
II II
58
,
.
-
,
-
.
r
._ _ -
-
_T
I1
_
--
H-e
__.
& Destroyer MCP(XSvs PSTARTTrend
25)
Figure 3.18
231
-
.
t
l-
|-
__
lA
_ 7
IE3
: E_ lEI 74_
1960 62 64
66 68 1970 72
(Ref.
_I I
-
x
PSTART
Frigates
_
.
I I
fi -E
56
r
.
.j
T81,~
, _/
I
T61 T41 lk - - T12---I -W "T611 I' "
4
I
1 -
2
_
U)
x
t,/'
F
T42(i)-
-
-
-
_
76
I
=
7
_
J
78 1980 82
84
When
life
the
output
from PRICE H is coupled
cycle costs can be estimated.
as the cost of supplying
with
PRICE
Life cycle costs are defined
and maintaining
a particular
system. The
model can discount future costs or present funding by
The
costs
equipment;
HL,
year.
estimated depend on the characteristics of
the:
employment/deployment; organization, and; maintenance
concepts used.
The model uses over 250 preset variables, each of
which
altered by the user,
may
be
support environment.
to
describe
the
system's
Input I__nformat
i on
3.1.5.3
The
minimum
following
data
"reasonable"
input
required
estimate
information
by
of
the
is considered to
PRICE
H
model
development and
the
a
costs
A glossary
the input variables and the data sheet used to input
information
The
obtain
manufacturing
during the feasibility/preliminary design phase.
all
to
be
for
system
into the PRICE H program can be found in Appendix F.
parameters
below are grouped into the
common
cost
classifications outlined in Table 1.5.
TECHNOLOGY
Manufacturing complexities
Technology base year
SHIP DESIGN
Hull structure
weight (SWBS group 100)
Propulsion plant weight (SWBS group 200)
Electric
plant weight (SWSSgroup 300)
Command& Surveillance
systems weight
(SWBSgroup 400)
Auxiliary systems weight (SWBS group 500)
Outfit & Furnishing weight (SWBS group 600)
Armament weight (SWBS group 700)
Design & Builders margin
Operating environment
PRICE H Development & Production Cost Parameters
Table 3.30
driver
MANUFACTUR I NG
Manufacturing complexities
Learning rate
PROGRAMMAT I C
Scheduling
Development start & end dates
Production start & end dates
Number of ships
ECONOMICS
Base year $
______________________________________________________________---
PRICE H Development & Production Cost Parameters
Table 3.30 (continued)
Output
.1. 5. 4
The
Information
PRICE H output
production costs,
cost data is
in terms of development
not the usual material/labor format seen
"shipbuilding only" cost methods. Table 3.31
categories
brief
total.
previously
description
and
with
shows the major cost
indicated in Table 3.28,
of the various elements that
and contains
make
up
a
their
~~-----,~~--
Prropgram Cost~,;~+ Cost Element
----
,~---
-
--
-
-
--
Q~eRg_!ippt,i on
-
-
--
-
-
--
-
-
--
-
-
--
-
-
--
-
-
Engineering
Data
Documentation costs for manuals,
lists, reports, deliverable
drawings, and other related items
Design
Cost of design and development
engineering
Drafting
Cost of manufacturing drawings,
data lists, specifications, and
incorporation of engineering
changes in drawings
Project
Management
Cost of program management and
control to include travel expenses,
computer operation costs, in-house
reports and the "ilities"
Systems
Cost of conversion of performance
requirements into design
speci i cations
ManLtufacturi
ng
Producti on
Manufacturing costs to include
material, labor, st-up, overhead,
and quality control
Prototype
Production costs for prototypes
including material, labor,
overhead, and qualification testing
Tool -Test
Cost of all special
Equipment
equipment to include their design
and any refurbishment
Description of PRICE H Cost Elements
Table 3.31
tools and test
The Costing and Design Systems Office at DTNSRDC has
familiarity
output
with
cost
PRICE
to correlate USN SWBS based
accounting categories in order to obtain
recognizable ASSET and NAVSEA formats.
enough
and
PRICE
the
Table 3.32 indicates
more
the
relationship for lead and follow ships between PRICE cost outputs
and
NAVSEA
end
cost
respectively,
construction plans (CP) and basic construction
(BC)
categories
211,
(e.g.,
see
MCCs
111/113
and
in Fig. 3.2).
PRICE
Cost Category
Development
Production
Drafting
CF (lead)
0.0
Design
CP (lead)
0.0
SWBS800 (lead)
Not Applicable
SWBS800 (lead)
SWBS800 (lead & follow)
SWBS800 (lead)
SWBS800 (lead & follow)
Engineering
Systems
Project
Mgmt
Data
Manuf acturi
ng
Production
SWBS1-7, margins,
NA
GFM
(lead & follow)
Prototype
Tool-Test Eqmt.
0.0
0.0
NA
SWBS900 (lead & follow)
Total Cost = Development + Production Costs
Relating
PRICE and SWBS-based Ship Costs
Table 3.32
236
The
following
PRICE
sum is used to estimate the
construction plans, CCp (see Eqn.
Cp
CP
cost
.12):
(3.80)
(Dev. Drafting) + (Dev. Design)
The basic construction cost,
of
CCC (see Eqn.
3.8),
is
then
estimated from the remaining PRICE cost elements, so that:
C
= FG& A ( (Total cost) -
( (Dev. Drafting) +
(3..Desi1)
(Dev. Design) ) )
where
F
G&A = 1.165
F
a
is
factor
that
accounts
for
general
and
G&A
administrative
overhead
not
included
in
the
FRICE
cost algorithms. The nLlmerical value of this factor is the result
of
experience
gained by DTNSRDC in comparing
NAVSEA
cost estimates with PRICE outputs (e.g., see Ref.
1ists
are
separately
through-put
by
PFFRICE and
from the total cost figure
their
in Eqn.
Once the construction cost is obtained,
categories indicated in Fig.
in
Section
26). GFM cost
sum
wi11
237
appear
(3.81).
the other end
3.2 can be calculated as
3. 1. 1.
formatted
cost
described
3.1.6
Other Cost Models
There
are several cost models which have not been discussed
in this study. In this section these models are broken into three
groups:
the FAST-E parametric model;
models for merchant ships,
and; contractor/shipyard models.
3.1.6.1
FAST-E
FAST-E
is
- Equipment)
(Freiman Analysis of Systems Technique
a parametric
cost model for estimating
cycle costs of various equipments.
acquisition
and
life-
the model is very similar
to
the RCA PRICE model.
This is not surprising, since Frank Freiman
originally
the RCA PRICE model back in the 1960s.
FAST-E
study
developed
model is explained in greater detail than found
in Refs.
This
cost
change
express
etc.)
in
this
18 and 37.
model
adjustable,
The
has
which
a
feature
of
being
is said to emulate designers'
perceptions
change
unique
(Ref.
18).
from a known design
to a new one using
This allows
(or
descriptions
and
linearly
managers'
analysts
shedule,
to
reliability,
such as 25% less, 50% more,
etc.
Costs- are related to an overall complexity factor which
itself
related to a variety of general
integration
factors
equipment
technologies,
and reliability & maintainability
impacts.
Complexities are calculated by calibrating the model using
data.
FAST
complexity
mechanical
system
factors
unlike
represent a
those
in
PRICE
is
complete
which
input
electrorepresent
structural and electronic separately.
This
one
feature
variable
makes
of combining the two PRICE complexities
it
easier to
calibrate
FAST
because
into
no
of
to be made as to the weight and density
has
assumption
the
various components. Only the total weight is required as input.
An unpublished Navy report on the FAST-E model suggests that
approximate
total ship
costs.
it
is a useful tool for obtaining
To
date no Navy personnel are actively pursuing the use of
in Navy ship costing,
model
Costing
System
and
although
this
the
in
several personnel
taken
FAST
cost estimates are usually broken down
into
Design Office at DTNSRDC
have
training.
3.1.6.2
Merchant Ship Cost Models
ship
Merchant
manhours
cost
material costs for three
and
These
categories.
are;
categories
(1)
Steel
(2)
Outfit
(3)
Machinery
Cost estimates are obtained in each area through the use
cost
engineering
detailed
for
CERs
weight/cost
general
estimating
cost
Conventional
trends.
for
more
field
was
procedures are employed
design definitions.
Perhaps
developed
at
the
most well-known
cost model in this
MARAD (Maritime Administration)
(e.g., see Ref. 38). A description
by
39).
There
are
of the procedures
several other references
merchant ships in the open literature.
under Refs. 40 - 43.
239
Fetchko
J.A.
available
MARAD for estimating ship costs is given in a paper by
(Ref.
of
at
Landsburg
to costing
Some of these are
of
listed
3.1.6.3
Private Contractor Models
Private
costs
contractors
typically estimate
construction
ship
using the conventional engineering method.
Labor
manhour
to
material costs are estimated on a system basis and sliummed
and
obtain a total ship cost.
Generally,
the work breakdown
system
by private contractors is not the SWBS system found in
used
the
Navy.
Private
shipyard
contractors
experience
rely
heavily
ulpon their
(most have spent several years
extensive
working
in
shipyards). Their databases are considered highly proprietary and
are closely guarded.
the open literature,
Because they seldom publish cost studies in
methodologies is
access to even their
difficult.
240
3. 2
COST MODEL COMPARISONS
This
section is intended to provide a comparative study
the NAVSEA Code 017,
comprehensive
examination
studies,
costs,
cost
of
of
NCA, ASSET, USCG and PRICE cost models. Any
model
comparison will
input/output
involve
sensitivities,
a
detailed
comparative
cost
the influence of design standards, treatment of program
database
requirements,
plus
a
variety of
other
cost
issues. The ability to perform such an analysis requires complete
access
to
the
models
and
data
bases
proprietary nature of costing in general,
involved.
Given
this access is
the
seldom
granted.
The
research
above
result of
papers on costing in the open literature apply
issues
based
this reluctance to share information is
in a general way.
is
to
the
primarily
pon data available in published journals and unclassified
government
documents,
comparisons
is also forced
3.2.1
Since this study
that
General
the
following discussion on
cost
model
to be of a general nature.
Model Characteristics
The five cost models selected in this study present a fairly
complete spectrum of methodologies used for new ship construction
costing. These include the following categories;
(1) conventional engineering
(2) CER
(3) parametric
The
NAVSEA
Code
017
model
is
representative
of
conventional engineering or so-called bottoms-up approach,
the
ship is broken down into discrete blocks for which
and
labor
costs
are estimated.
241
The highest
level
the
where
material
of
detail
for
obtained
this
approach
occurs when the
estimator
job.
for the entire
possession
of a bill of material
individual
material and labor costs have been
is
Once
in
the
they
calculated,
are summed for the total ship construction cost.
The
NCA
USCG,
and
ASSET
use
models
cost
estimating
relationships (CERs) to estimate costs. CERs employ a single nonto cost relationship to estimate ship
cost
can be developed down to any level
CERs
Although
of
costs.
technical
are usually applied on the 1 digit SWBS level
they
detail,
construction
and
are referred to as a top-down approach.
is representative
The RCA PRICE model
of
end
cost
the
parametrically.
approaches,
spectrum
methodology
Parametric
intended
estimates
and
at the other
costs
estimate
also
are
top-down
for use under conditions in which there is
not a detailed breakdown of the ship.
approach dffers
of models
the parametric
Typically,
from the use of CERs,
in that it uses more than
one non-cost variable in its regression.
plus
the models use weight as the main cost driver,
All
accouint for
different
method
to
include
a multitude of influences,
variations in materials,
Technologies
technologies.
including but not limited to,
propulsion, electrical power generation
and distribution and electronics.
invariably
a
are "high tech",
Combat related systems,
which
to be
input
are generally
intended
using a shopping list.
Technology differences are accounted for in the NAVSEA model
through
linear
the
extensive
data base.
The NCA model
trend lines for the various technologies
.242
has
different
represented
in
The ASSET and RCA PRICE models use technology
its cost elements.
to describe the variations in technology (although there
factors
is no relationship between each model's factors).
The USCG model
adjusts CER outputs directly using appropriate multipliers.
Rangeof Appelicability
3.2.2
3.2.2.1 Vessel
Range
Type and Displacement
for which each model is capable
displacements
and
of
producing
Accurate model results are assumedfor ships
"accurate" results.
place them within the range of ship costing
whosedesign features
accurate
information available in the database. In other words,
are produced whencost information on similar
results
types
of applicability refers to the variety of ship
ship types
has been incorporated into the model CERs.
Table 3.33
indicates
the vessels for which each
model is
was obtained
intended and has in fact been used. This information
model.
from personnel using the specified
The wide range of applicability
for the NAVSEAcost model is
of
the existence and continual updating
primarily
due to
database.
The NCA checks the NAVSEAestimate
for
conventional
monohull combatants using the model discussed in this study.
vessels,
other
NCA
Due to the
Corporation.
uses
a
model
developed
its
the
by
For
RAND
classified nature of this model, it could
not be documented.
The
it
ASSET model indicates a wide range of vessels for which
can be used.
ASSET
CERs
It
is important to note that the data
is from the early to mid 1970's and so
243
is
for
the
becoming
Cost
Range of Applicability
Model
Vessel Type
Displacement
(LT)
NAVSEA Code 017
any new construction
NCA
frigates, destroyers,
cruLi sers
ASSET
all monohulls, SWATH,
hydrofoils, SES, CV
10 - 20,000
USCG
all CG vessels
(i.e,
patrol boats to polar
69 - 12,000
2 - 100,000
5,000-
12,000
i cebreakers)
RCA PRICE
any new construction
2 - 100,000
Cost Model Vessel Applicability
Table 3.33
dated.
the
In addition, several of the advanced vehicles included in
ASSET database
environment
and
are
were built
therefore
in an aircraft
not indicative
of
and/or
a
R&D
production
setting.
The
vehicles
Coast
Guard model has incorporated a wide range
in its database.
Due to limited construction
of USCG
in the CG,
the database for each vessel type is typically small. However, it
has been found to be representative of the indicated range.
The RCA PRICE model
ability
has an all encompassing
range due to its
to be calibrated and the universality of its
244
complexity
factors.
vessel
The
flexibility of the complexity factors
type ot be estimated,
allows
even if extrapolating
outside
any
the
range of the database.
Cost analysts,
a
ship
cost
for
by necessity, may be called upon to estimate
a vessel type not
included
in
the
current
database. For these situations, it is important to understand how
each
model
can be perturbated
(and by how much) to account
for
new design variations.
As mentioned for the RCA PRICE model, the complexity factors
allow
for
technology
extrapolation
factors
in
outside of
the
the
model
ASSET
database
do
not
range.
The
have
any
universality, in that a new technology factor cannot be estimated
from a knowledge of the current database values,
and the factors
scale differently for each SWBS group.
By the nature of their development, cost models are intended
to be used to estimate ships similar to those in the database. If
it
becomes necessary to extrapolate outside of
must
be
done
so logically
and
this
intelligently.
range,
Phrases
it
like,
"relate existing data tempered with experience" become applicable
(but
certainly
not
helpful)
and
show
the
importance
of
knowledgeable personnel in the cost analysis process.
3.2.2.2
All
provide
Ship Desian Phase
of
some
conceptual,
the
the models discussed in this study are intended
weights in the
to
of
ship
cost
estimates
during
the
and on into the preliminary design phase. Typically,
technical
Due
measure
to
definition
available during
these
periods
are
- 2 digit SWBS level.
its conventional engineering approach,
245
the
NAVSEA
Code
017
is applicable from
model
the
provided
appropriate
level
ROM
to
budget
is
definition
technical
of
estimates,
available. Although, any of the cost methodologies will produce a
more
accurate estimate (as related to its database) if there
is
uncertainty in the input data.
less
model was intended
The ASSET cost
for
comparative
cost
of naval vehicles rather than specific point estimates.
analyses
Therefore,
ASSET
estimates may be the
cost
questionable
most
(i.e., uncertain) for specific ship design applications.
Model _In_pt/Outtput
3.2.3
the
Al1
estimation
weight
models
Comparisons
discuLssed in
this
technology
construction
costs.
level
are
address
the
As previously
mentioned,
the main inputs for
estimating
of ship acquisition costs.
and
study
Scheduling and programmatic costs are
also
important for calculating total acquisition costs.
Only the ASSET cost model directly estimates the total lifecycle cost (LCC) associated with a ship. LCCs are most frequently
for select comparative studies and require
used
operating
equipment
profile,
analogy
the
personnel
salaries,
such as those
obtained
consumption,
-spares and maintenance policies.
Financial
using
fuel
infor-mation on
spreadsheet comparisons,
USCG's CASHWHARS program (e.g., see Ref. 34), or direct
comparisons
with
the
USN's VAMOSC
methods of doing these LCCstudies.
246
data,
are
popular
3.2.3.1
Minimum Ipt
Requirements
In this section, the cost parameters that are representative
of
the technical definition necessary to cost a ship during
conceptual/preliminary design phase are presented.
data required for each model,
the
The amount of
or its definition, may differ from
the inputs given below.
The
user
NAVSEA
model
would require
the most input data
because of its conventional engineering approach.
PFRICE model
would
require the least amount
by
the
The
RCA
of data to obtain
a
reasonable estimate, provided it was calibrated initially.
Table
estimate
listed
3.34 shows the representative input data required
the
construction cost of a lead ship.
The
in this table are the general cost driver
Table 1.5 and were used previously
to
categories
categories
in
to group the cost parameters
associated with each model.
Weight
cost drivers
in Table
.34
are in the form of the 1-
digit level SWBS groups and the margin weights.
Cost
influences
from the appropriate levels of technology apply to manufacturing,
materials, propulsion, electric power generation and electronics.
Expensive
shopping
combat
lists.
systems
Scheduling
related
effects
components
enter
are affected
through
through
the
interaction of length of construction and the rate of inflation.
Other
parameters
reflect the shipyard labor
and
material
costs or provide a base date for technology effects and
monetary
conversions.
247
TECHNOLOGY
Ship type (i.e., SWATH, Hydrofoil, etc.)
Hull & Superstructure material
Fropulsion plant type
Electrical plant type
Year of technology
SHIP DESIGN
Hull structure weight (SWBS group 100)
Propulsion plant horsepower
Propulsion plant weight (SWBS group 200)
Electric plant weight (SWBS group 300)
Command & Surveillance systems weight (SWBS group 400)
Auxiliary systems weight (SWBS group 500)
Outfit & Furnishing weight (SWBS group 600)
Armament weight (SWBS group 700)
Margin weights
F'AYLOAD DESIGN
Cost of payload material
(includes armament& weaponsportion of cornm.
& surv.)
MANUFACTURING
Labor rate
Production technology
PROGRAMMATIC
Profit fraction
Overhead fraction
Construction start date
Delivery date
ECONOM ICS
Inflation rates
Base year $
BLSmaterial indices
Base year for BLS data
Representative
Lead Ship Construction
Table
248
.34
Cost Parmeters
Table
3.35
shows the additional data required to
estimate
construction costs for a multi-ship procurement. Frimary inputted
data
include
a labor and material cost breakdown for
the
lead
ship, inflation rates, and profit and overhead fractions.
Major
from
the
inflation.
influences
on multi-ship construction
relationships
between
learning,
costs
result
scheduling
and
Drawn out scheduling can substantially increase costs
due to inflation.
Multiple production lots increase costs due to
a reduction in production learning effects.
Increased costs due to uneconomical scheduling or unforeseen
inflation
can easily result in a decrease in the number of ships
that can be produced
if there is only a fixed amount of money
be spent.
Inp!tte d Dat a
Lead ship construction costs
Inflation rates
Profit fraction
Overhead fraction
Base year $
MANUFACTUR ING
Learning rate
(material and labor)
PROGRAMMAT IC
Number of ship required
Construction start dates
Delivery dates
Representative Multi-Ship Construction Cost
Table 3.35
249
armeters
to
LCCs
discussed
in this study deal mainly with
and operations & support (O&S) costs.
to
the
acquisition
cquisition costs are equal
construction price plus various program and
G&A
add-on
costs. These add-on costs are generally estimated as a percentage
of the construction price.
Typically,
the total program cost is
30-40% above the construction price.
O&S costs include contributions from the areas of personnel,
operations,
maintenance,
related to salaries,
the
time
the
fuel and spares.
Personnel costs
are
benefits and training. Operations determine
ship will
be
underway.
Maintenance,
fuel
and
equipment spares costs are self-explanatory. Table 3.56 indicates
the
additional information required to estimate
LCCs.
Inputted
data consists of total ship acquisition (i.e., program) costs and
the necessary economic data.
Research
incurred
into
the
& development plus testing & evaluation costs
to enable the introduction of high
the ship design.
construction
or
technology
are
systems
The additional facilities costs relate to
refurbishment of
maintenance
or
docking
facilities.
The
vessel
ship's residual value refers to the scrap value of
at
the end
of its service
life.
generally in the range of 25-30 years.
25 0)
The
service
life
the
is
Inputtted Data
Total ship acquisition costs
Ship delivery dates
Inflation rates
Interest rates
Base year
SHIP DESIGN
Ship fuel rate
Crew accommodations
PAYLOAD DESIGN
Payload fuel rate
PROGRAMMAT I C
R & D,
OPERATIONS
T & E costs
AND SUPPORT
Annual cost of training ordnance.
Equipment spares
Additional facilities
Salaries,benefits
costs
and training
Annual operating hours
Ship residual value
Service life
Fuel cost/gal
Representative Life-Cycle Cost Parameters
Table 3.36
251
3.2..3.2
Cost Output Sensitivities
In this section,
to
variations
accomplished
subsystem
in
the sensitivity of the cost model
the
input
by examining
cost
data
marginal
are
discussed.
cost factors
drivers) and total
outputts
This
(i.e.,
acquisition
is
the major
(program) cost
sensitivities.
There
is
no
quantitative
comparison
between
cost
sensitivities of the different models reviewed in this study.
To
have done this would have required a detailed comparison of model
outputs
for
various
ship types.
Unfortunately
this
was
not
feasible given that these results would have been considered of a
proprietary nature.
3.2.3.2.1
Marginal Cost Factors
Marginal
determine
cost
the
factors
provide
For
convenient
method
effect of subsystem changes to ship costs
see Ref. 5, 13 and 35). The marginal
associated
a
cost is the change
(e.g.,
(in cost)
with a unit change of weight of a particular
this study,
to
system.
the subsystems are identified by their
1-digit
SWBS designation.
In
found
terms of this breakdown,
that
leverage
the
four
cost analysts
(4) SWBS groups with
the
have
typically
greatest
cost
are;
(1) combat systems & related electronics
(2) electric generation and distribution
(3) propulsion systems
(4) axiliary
This
subsystem
machinery
list is representative of the naval surface combatants
cost
percentage
breakdowns
noted
in
Ref.
36.
The
and
electrics
upon
the
systems
cost/weight
depending
propulsion systems may be interchanged
chosen
for
each.
Table
3.37
lists
the
sensitivities calculated for the DD3 51 navy surface
combatant. Note that the factors for combat systems and command &
control groups do not include the cost of equipment.
Marginal Cost*
Factor ($/LT)
SWBS Group
17,500
100
Structure
200
Pr opul s i on
110, 800
300
Electrics
180,200
400
Command
46,300
500
Auxi iary
80,800
600
Outfi t
78,500
700
Armament
16,000
*
costs in FY 83 $
** shipyard installation costs only, equipment procured
separately
Naval Surface Combatant Marginal Cost Factors
(Ref. 5)
Table 3.37
in
Personnel
DTNSRDC
costing and
the
systems
office
at
for
the
the cost/weight sensitivities
that
indicated
design
ASSET and RCA PRICE models were similar.
3. 2. 3.2.2
Program Cost Factors
cost sensitivities examine the
Program
cost
contributions
in
addition
from a one
resulting
loaded
labor
(i.e.,
additional
plus
or
reduction
dollar
direct
program
overhead)
and
material costs.
Table
shows
that effect of the
dollar
NAVSEA
format.
dollar change in labor
cost
in
results
This example provides
change of $2.73.
cost
program
A
in
increase
labor during construction on subsequent elements of
loaded
using
3.38
a
direct
a
illustration of the importance of add-on costs to ship costing.
Table
The
increase.
lower
3.39
dollar
shows
a
similar effect for
a
material
smaller program cost is primarily the
for
percentages
engineering
and
cost
result
of
construction
services material.
3.2. 3.
Cost Otput
Accuracy
This section is an attempt to provide a quantitative measure
of
the
accuracy
that a user can expect with
the
cost
models
presented,' and a qualitative measure of cost estimating accuracy
in general.
The
personnel
using
the five models in
this
study
contacted and asked to provide a representative percentage
figure
that
would given an indication of the uncertainty
point estimate.
For example,
the value of the estimate
- 10%. Some interesting points were brought Lip.
.54
were
error
of
a
is $X + or
Cost Factor
Element
Calculation
No.
Description
1
SWBS 100-700
2
Margins
12.5%
3
SWBS 800
4
-
$
Amount
1.00
(1)
.125
30%
(1 + 2)
.396
SWBS 900
27%
(1 + 2) LR*
.304
5
SWBS 100-900
1 + 2 + 3 + 4
6
Profit
10% (5)
7
FCM
5%
8
BCC
5 + 6 + 7
9
Change Orders
1)% (8)
.205
10
Escalation
20% (5)
.365
11
PM
12
Program Cost
*
**
Growth
5%
(5)
.183
OV
(8 + 9)
8 + 9 + 10 + 11
LR = engineering/manufacturing labor rate ratio
OV = engineering direct/burdened labor rate ratio
Labor/Program Cost Sensitivity
Table 3.38
1.825
.044
2.052
.113
2.735
Cost Factor
Element
Calculation
$ Amount
No.
Description
1
SWBS 100-700
2
Margins
12.5% (1)
.125
3
SWEBS800
1% (1 + 2)
.011
4
SWBS 900
4% (1 + 2)
.045
5
SWBS 100-90)(
1 + 2 + 3 + 4
6
Profit
10% (5)
7
FCM
8
BCC
5 + 6
9
Change Orders
10% (8)
. 129
10
Escalation
20%
.236
11
PM Growth
5% (8 + 9)
12
Program Cost
8 + 9 + 10 + 11
-
(5)
Material/Program Cost Sensitivity
Table 3.39
256
1.00
1. 181
.118
1.299
.072
1.737
First
of all,
each person contacted was more than a little
hesitant to give a representative figure.
If nothing else,
this
demonstrates the uncertainty that pervades cost analysis. Figures
that were given were typically in the range of plus or minus
15-
25%
the
for
estimates during the conceptual phase and 7-15% at
end of preliminary design.
Such
a
percentage figure is strictly
there is no uncertainty involved.
estimate
is
percentage
applicable
if
For instance, if a single cost
compared to the return costs for that
error
only
would completely define the
vessel,
accuracy
the
of
the
estimate. However, if the figure is given as a general measure of
the
accuracy,
then
in
order
for
it
to
be
statistically
meaningful, it requires an indication of the analyst's confidence
that
any
given estimate will fall within the
specified
range.
estimates
versus
This point is illustrated in Section 2.3.1.
Percentage
error
differences
for
return cost data vary significantly.
to
cost
Differences can be expected
decrease as input data becomes more accurate.
accuracy
is usually
is
a necessary requirement
not
similar
costs
accompanied by greaterdetail,
for
accuracy.
Although
more
greater detail
Ref.
26
shows
percentage errors for ACVcost estimates versus return
for input data ranging in detail from whole ship
down
to
the 3-digit SWBS level.
Current
indicated
to
point
be
estimate
double those of the
reasons were given for this;
uncertain
database
to
errors for the
other
ASSET
cost
model
were
models.
Two
(1) the KN technology factors
within 2(:%of their values due to the 15 year
from which they were derived,
2.57
and;
(2) the' model
were
old
was
initially intended for comparative studies.
The relative cost percentage errors for the ASSET model
be
expected
estimates.
since
to be significantly lower than those of
This
statement
is true for any
the
comparative
can
point
study,
only selected baseline characteristics are prturbated
create
the variant.
Because of this strong correlation
the baseline and variant,
to
between
there is less overall cost uncertainty
(i.e., a smaller variance) and therefore reduced errors.
It
not
a
is important to note that in cost estimating,
lot
of
confidence associated
with
any
there
one
is
estimate.
Although this can easily be shown to be statistically true (e.g.,
see
leve
Section
2.3.1),
it
can also be seen at a
bids
received,
characteristics
bids
practical
1.
For any request for bids,
the
more
there will be a wide variation in
each reflecting the
of the yard at the time.
particular
needs
Because each of
and
these
can be rationalized in relation to the yard submitting
it,
the bid is "accurate" for the assumptions and data used (provided
that there are no blatant errors).
The
preceding discussion indicates that a comparison of the
accuracy
or
requires
a
correctness of any cost
methodology
with
another
detailed examination of the assumptions used in
the
development of the model. These assumptions will impact the areas
related to standards,
management and scheduling,
few.
258
to name but
a
CHAPTER
4
DEVELOPING A SHIP COST ESTIMATING CAPABILITY
The
design
increased emphasis on the economic feasibility of
(e.g.,
involvement
and
However,
any
within
the
constraints dictated by the perceived needs and resources of
the
visibility
costing
see Ref.
from
the cost
capability
must
requires
estimating
be
a greater
community.
developed to
operate
throughout the conceptual/preliminary design
Even
organization.
52),
ship
phase, these needs can vary from cost comparisons for feasibility
studies
trade-off
ship
a costing capability.
require
Fig.
4.1 groups these
priorities
may
that
Therefore,
in
addition
associated
also
experience changes.
is
estimating capabilities should be
view
of
current
management
to fLifilling these needs,
quantifiable
accuracy
and
available within a specified timeframe.
require
their
sense.
the
important
Accuracies
to
requirements.
periodically
direction.
there must
estimate
Estimates for
must be accurate
in the range of 10-20% are
be
an
must
be
trade-off
relative accuracy between comparative
estimates on the other hand,
Budgetary
It
a priority shift can affect costing
cost
re-evaluated
absolute
that
functions
most organizations are continually evolving,
Because
studies
specific
to their applicability within the ship design process.
according
In
for
Table 4.1 lists a wide variety of functions
designs.
realize
through to budgetary estimates
designs.
in
an
usually
indicated for estimates in the early stages of design.
The obvious time for an estimate to be available is whenever
a
decision
point is reached in the design process.
259
During
the
conceptual/preliminary design phase,
translates
has
into
this requirement
immediate availability.
led to the development of cost
interface with the otput
This rapid
Design-To-Cost
Source Selection
Froposal Review
Trade-Off Studies
Budgeting
Cost Control
Technology Impact
Specification/Standards Impact
Economics Effects
Scheduling Effects
Bid - No Bid Decisions
Contract Negotiations
Organizational Cost Estimating Functions
(Ref. 58)
Table 4.1
260
turnaround
estimating models which will
from design synthesis models.
Contract
generally
* COST IMPACT
* COSTDRIVERS
* COST STANDARDS
* ADV. VEHICLE
* COST AS A
DESIGNPARAMETER
* IMPROVEDCOST
ESTIMATINGTOOLS
ABILITYTO
ESTIMATERELATIVE
COST CHANGEAS
DESIGNADVANCES
* IMPROVEDMETHODS
TO ESTIMATE
NON-HARDWARE
COSTS
* COST TRADEOFF
ANALYSIS
DATA BASE
* R&D PROGRAMS
FORCOST
REDUCTION
* ESTIMATE
RELATIVE
COST
* IMPROVEDCOST
DATA BANK
* HIGH QUALITY
COST ESTIMATES
* COST IMPACT
OF ACOUSTON
POLCES
* INCREASEDCOST
DETAILS
* EFFECTIVECOST
ESTIMATING
PARTICIPATION
o
INDEPENDENT
COSTESTIMATE
* ANALYZE
CONTRACTOR
COST ESTIMATE
Cost Estimating Functions for the Ship Design Process
(Ref. 59)
Figure 4.1
261
* COST IMPACT OF
CHANGES,DELAY.
DISRUPTION.ETC.
* ABILITY TO
ADJUDICATE
COST CHANGES
* AWARENESSOF
COSTDRIVERS
4.1
COST MODEL DEVELOPMENT
This section deals with some of the factors to be considered
for
developing a parametric cost estimating model (of which CERs
are
assumed to be a subset).
The primary motivations
for
using
parametrics are to reduce the costs of estimating, and to provide
consistent,
primarily
there
(e.g.,
repeatable
results.
Although
used as an independent check for
parametrics
contract
proposals,
is increasing interest in using them for contract
see
Ref.
57).
A
list
of criteria
for
objective is given in Table 4.2.
Cost-to-Noncost Estimating Relationships
Must Be Logical
Data Used For CERs Must Be Verifiable
A Significant Statistical Relationship
Must Exist Between The Variables
CERs Must Predict Well
Parametric Estimating Systems Have To Be
Easy to Monitor
Compliance To The Previous Criteria
Must Be Verifiable
Criteria for Parametric Contract Fricing
(Ref. 57)
Table 4.2
262
are
pricing
meeting
this
The
first
Chapter 2.
within
in
Table 4.2 has
been
discussed
Data verification will ensure that the estimates
the
database
criteria
database's
range
is kept current.
of
applicability
In order to predict
and
well,
that
the
in
are
the
model
must be sensitive to a multitude of factors, including technology
improvements,
design and manufacturing complexities, scheduling,
programmatic
considerations and the "ilities"
reliability,
etc.).
should
be
criteria
Ease
of monitoring indicates that the CERs
able to be easily updated and
requires
(maintainability,
calibrated.
that documentation sufficient to
The
last
verify
the
previous requirements accompanies the estimate.
Although
cost
relationships (e.g,
particular
and
most models are based upon similar non-cost versus
model chosen
experience
of
the
weight versus construction cost),
is
largely dependent upon the
organization
requirements of its customers.
requirement for budget
format
and
should
For example,
submissions
the
the
interests
reflect
the
labor/material
makes the
engineering
costing method the only practical approach for NAVSEA (e.g., see
Section
3.1.1).
particular
the
group,
of
the
but also the resources that were available to
group during the model development.
Before
form
Cost models not only reflect the needs
of
prospective
benefits
a
parametric estimating system is
cost/benefit
analysis
should
be
developed,
some
performed.
The
costs of a new system can then be weighed
of
potential
savings
in
estimating
against the
effort
and/or
increased estimating accuracy.
Most
data
parametric models require essentially the
(e.g.,
see
Section 3.2),
263
although they may
same
input
weigh
these
parameters differently within the algorithms of the model.
Also,
since these models are trend indicators, they tend to reflect the
projects in the database.
4.1.1
Data Gather ing
The most difficult
phases
part of cost estimation
in early
program
is that of finding the proper mix of cost model and model
input
data.
properties
The
of
model
the
shoul d
system
require
inputs
that are well
based
defined
on
and
can
the
be
established with a reasonable degree of certainty at the time the
estimate
is
required
(e.g.,
SWBS
1-digit
weights
during
feasibility studies).
Data
gathering
considered
is
just one of the factors
for the development of the cost
that
database.
must
Fig.
be
4.2
indicates six factors, including data acquisition, that determine
the
characteristics
considerations
are
of
the
cost
database.
important for information of
Data
a
access
proprietary
nature. Cost definitions and standards (e.g., sailaway costs, mil
specs,
etc.)
specify
the format
in which the cost data will
be
found. The economic factors and definitions allow for consistency
in
the
cost reported
(e.g.,
see Section
level of detail of the available
database
1.5. 1).
greatly
The size
and
influences
the
choice of costing technique.
If
then
there
it
is an abundance
becomes
much
of data for a particular
easier
to
develop
representative cost estimating relationship.
are
system,
accurate
Whether these CERs
developed within the organization or by outside agencies
264
aid
is
COST
DATA BASE
DATA
ACQUISITION
A
BA!
COST
MODELS
COST
PREDICTION
DESIGN
/
% za
.,,~
~
<
Sunni nrtV
I
l[.
MANUFACTURING
TECHNOLOGY
*'-U,)
CcT
DnlIrT Ul
I I\LLJUL. I Tnl
Factors Affecting the Cost Database and Cost Predictions
(Ref. 60)
Figure 4.2
265
i
on
dependent
the
monies
and
to
available
personnel
the
organization. The cost models at NAVSEA Code 017 and the USCG are
examples of internally (i.e. in-house)
the
When
undoubtably
cost
data resides with other
proprietary
of
BIW
NCA
data
source.
it
companies,
easily
is
obtainable.
contracted the naval architecture
& Cox to develop a cost model using
Gibbs
proprietary
not
and therefore
Due to limited resources,
firm
developed CERs.
Gibbs
& Cox
in
shipyard to supply cost information,
turn
an
outside
subcontracted
nwilling
which it was
to supply of its own volition.
such as
When there is little data available for the design,
for
it becomes difficult
to
of existing CERs can lead
to
technology applications,
advanced
The extrapolation
estimate costs.
erroneous results. Models that have been developed to account for
all ranges of technology include the RCA FRICE and FAST-E models.
Fig.
4.2 also illustrates
that a cost database
can create a
level of cost awareness sufficient to allow analysts to determine
the
cost
impacts
from
changes
specifications, technology and
Data
can
design
methods
and
operational capabilities.
be collected from
cost data (e.g.,
in
shipyard data).
individuals or
from
existing
In either case, data gathering
can be broken down into three steps;
(1) find the right experts or cost data files
(2) ask the right questions or retrieve necessary data
(3) transform the responses or data into data
compatible with the model input data
The
processing
individuals
biased
involved
is illustrated
in
retrieving
cost
data
in Fig. 4.3. The data collected
by the respondent's perception of the event in
from
may be
question;
it may be irretrievable based upon the respondent's understanding
of the question,
Once
or; it may even be forgotten.
access
to
cost data files
is
available,
the
main
problem with existing data is the conversion of the data. Much of
the
effort
in
developing
transformation
of
the
shipyard
NCA
cost
cost
data
model
into
involved
SWBS
the
compatible
groupings (see Ref. 23).
The
most important consideration for the maintenance of
historical
database
is to have a process
consistently collect data.
to comprehensively
an
and
If personnel other than cost analysts
are required to document return costs,
it is imperative that the
forms used be simple and short.
repeated personal contact
Also,
provides continuing reminders and gets the job done!
4.1.1.1
Establishingq User Needs
Information
methodologies
on
user
needs
can be
(i.e., CER, Parametric)
divided
into
and model features
costing
that are
important to the analyst. Table 4.3 provides a partial listing of
some
his
of these desirable features.
own
ranking
for this list,
Since each analyst
a cost model can
be
will have
developed
which is cognizant of those priorities.
In
order
requirements
throughout
to
establish the cost
within the USCG,
various
estimating
a questionnaire
departments at HQ in
267
and
was
Washington,
modeling
distributed
D.C.
The
RESPONOENT'SINFORMATIONPROCESSING
INTERVIEWER'SDESIGN OF OUESTIONS
Behavioural Information Processing Model
(Ref. 19)
Figure 4.3
268
Input Data Availability
Is the input data requi,-ed by the model available
during
the conceptual/feasibility design phase?
Evaluation Capability
DesLiagn
Does the model consider factors appropriate
for design trade-off studies?
Ease of Use
Is the model easy to use with minimal training?
Ease of Calibration
Can the model be calibrated to the stated situation
with available data?
Data Base Validi ty
Is the model database relevant to and valid for
the situation addressed?
Currentness
Is the model based on current information which
reflects the latest technology?
bi 1i ty
Accessi
Is the model easy to access at a low cost?
A1i
pP
_Ralg?._of
cabi
i tey
Is the model applicable to a wide variety of programs?
Ease of Modification
Can the model be easily modified to
incorporate new data?
Cost Model User Features
(Ref. 19)
Table 4.3
269
survey,
Table 4.3,
The
in Appendix C,
found
as well as asking for specific costing
intent
different
addresses all of the features
of
the survey was to develop an
costing
recommended
cost
model
selection
or
requirements.
awareness
requirements thoughout the CG,
in
of
the
that
any
would
be
so
selections
responsive to these needs.
Unfortunately, of the twenty surveys distributed at USCG HQ,
one
only
was
making
returned,
sampling
representative
it
impossible
to
obtain
There
of the costing requirements.
one reason that the information was not available for
only
study,
and
that
was the lack of personal follow-up
after
a
is
this
the
questionnaires were distributed. A more effective procedure would
been to meet personally on a one-on-one basis and let
have
answer
the survey questions orally.
In this manner,
not have been intimidated by a lengthy survey,
of
their time would be lost,
them
they would
a minimum
and any misunderstandings
amount
arising
could be settled immediately.
4.1.2
Resource Reqirements
Personnel and monetary resources are closely related through
the costs of salaries and training.
have
several
individuals
who have been
estimating for several years.
minimum
pool
number.
ensures
associated
with
cost
Two people should be regarded as a
A constant nucleus of people provides a
stable
which tends to
increase
the
management's) confidence in their
ability,
and
that new staff members receive extensive formal and
on-
of knowledge and experience,
analysts'
Most organizations generally
(and
the-job training.
2 7C)
Monetary
costs
resources are not only required for the
associated
(e.g.,
with the development of a
software,
equipment purchases,
viable
manuals,
front-end
cost
model
training) , but
also with recurring operations and support costs (e.g., salaries,
training,
model updates,
resLilt from
effects
of
equipment
repair).
Model updates
can
programmatic changes and/or CER changes due to
new
materials
and
fabrication
processes
the
(e.g.,
robotics, zone-outfitting, etc.).
4.1.3
Organizational Considerations
For cost estimating to function properly,
must
be
(a) committed to maintaining the
responsive
to
organization
the
results.
As part
of
the
organization
capability,
its
and
commitment,
must allow the costing group to communicate
(b)
the
within
the organization as the need arises.
Effective
encountered
for
reporting.
both
can overcome 90%/of
information
tables)
management.
For
can
greatly
example,
popular
affect
financial
problems
and
analysis
the
CASHWHARS (see Ref.
text,
receptivity
spreadsheet
programs
because results can be obtained and
immediately as inputs are changed.
USCG
gathering.
the
The presentation of estimating resilts (i.e.,
graphics,
becoming
communication
of
are
displayed
Thus, programs similar to the
34) give management fast feedback
for
program trade-offs.
Historically,
although
costs
risk analysis,
probabilities
have been presented as a point
with its range of costs and
(see Section
2.3),
271
defines
costs
val Lue,
associated
in
a
more
(and statistically)
intuitively
illustrates
correct
costs can be presented
how
Fig.
manner.
4.4
management's
combining
desire for "a" cost figure with risk analysis. Such presentations
can
overcome management bias and resistance
and
provide
still
all the necessary data for decision-making.
can
Management
a
its commitment to sustaining
show
the cost team
Primarily,
estimating capability in several ways.
cost
must be included in the early stages of design, where costing has
on
maximal
impact
reduces
financial
design.
Early planning
uncertainty
cost
of
As
in the project.
resources
the
design
progresses,
cost analysts must be allowed to interact on a real-
time
with
basis
to
design engineers
relationships
This will allow the
costs and design engineering.
between
establish
cost
impacts of design changes to be known quickly.
If the need for a cost estimating team has been established,
management must support the development of a time-phased plan
make it a viable,
activities
support
for
(e.g., see
development,
and model
demonstrate
training
and
ef. 58) .
To gain management's confidence,
to
This involves prioritizing
independent group.
database
to
repeatably
that
the cost team must be able
the
estimates
produced
are
reliable. Because this can only be done using actual return data,
the
rate
that management trust is built-up is highly
upon the product cycle.
dependent
For a very short cycle, this process may
take only a few months. For shipbuilding in the USCG, the product
cycle
is
several
years,
and so many of the
makers have yet to go through a ship acquisition
present billets.
272
current
decision
program in their
PROBABILITY
RESERVE IS
SUFFICIENT
100
Nominal cost
$330 million
+
60
0
33
RESERVE (SM)
60 percent confidence that 10 percent reserve is sufficient
Point Estimate with Cost Reserve Probability
(Ref. 19)
Figure 4.4
273
COST MODEL SELECTION
4.2
a cost model can be divided
approach to selecting
The
into
four basicsteps:
(1) determining needs
(2) selecting candidate models
(3) choosing the best model(s)
(4) reconsidering the choice
The previous sections in this chapter have discussed some of
needs that may require the development of a cost
the
and
capability
Chapter
several existing models have been
The choice
3.
of the best model(s)
discussed
in
both
should consider
accuracy)
(i.e.,
(see Table 4.3) and quantitative
qualitative
estimating
considerations.
Most
can support
models
limitations
of
the
models
a cost estimating
are
they
understood,
the
if
activity
used
are
consistently, and are calibrated for the particular organization.
From
the
various
of
limited information available on the accuracy
costing techniques discLussedin Chapter 3,
there is
the
no
evidence to suggest that the existing USCG cost estimating
model
for new ship costing in the CG. However,
it is
is not appropriate
generally
recommended
model be utilized:
(e.g.,
see Ref.
one as the primary
19),
that more than one
estimating
resource
and the
(i.e.,
other(s) for comparison and validation of primary outputs
an independent reasonableness check).
CER
and model development can be done in-house,
to an outside source,
to
or purchased on a time-sharing
contracted
basis.
the front-end resource requirements for in-house and
Due
outside
source development, time-sharing models are becoming increasingly
274
The
RCA
PRICE parametric cost model is the
most
commonly
It is primarily
category.
used model in the time-sharing
a
models.
back-up
as
particularly
attractive,
used as
(i.e.,
back-up model for validating bottoms-up cost estimates
the
NAVSEA Code 017 model) and generating internal cost
targets
attractive features of this model (see
Section
(Ref. 19).
3.1.5)
The
include
it
detail,
is
applicability.
its ability to analyze systems at any
easily
However,
calibrated, and
as
with
any
has
unlimited
costing
system
technique,
and quantity
accuracy is dependent upon the quality
of
level
its
of input data
available.
At
the
present
time,
the current USCG
cost
model
(see
Section 3.1.4), combined with the RCA PRICE parametric model as a
back-up seems appropriate for ship estimating in the CG. However,
as suggested
reconsidered
analyzed,
any model choice
in the four steps above,
periodically
organizational
due
to
changes
structure,
experience.
275
in
database,
should be
projects
personnel
being
and
CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
5.1
CONCLUSIONS
discusses some of the technical and
study
This
that
considerations
can be
capability
are applicable
ship
ship
new
to
design.
The
for
the
with the costing requirements
overlap
can
techniques
The
any organization.
within
established
during the feasibility phase of
construction
estimating
must be examinedbefore a cost
techniques
estimating
cost
managerial
conceptual/exploratory, as well as preliminary phases.
of a mission profile
translation
The
into the
final
ship
design necessitates a ship optimization process. Synthesis models
one method used to provide the alternative designs
are
to
this
produce
optimization.
Any method
generally
required
the
has
following common elements;
(1)
selection
of criteria,
or measures
of merit,
that
will indicate which alternative ship design is the
opti mum
(2)
development of CERs
(3)
construction
Measures
of
merit
of LCC models
can
be
rated
according
to
military
effectiveness, LCCs, or techical design characteristics. The CERs
require
the determination of ship design,
operational
factors
total ship costs
occur
which have an effect
from inception
on costs.
to retirement
in three distinct phases;
R&D,
construction
and/or
LCCs are the
and are assumed
to
investment or acquisition
and O&S. This study is primarily concerned with so-called initial
276
investment
costs associated with the delivery of the
the
owner.
These include material,
labor,
for
hull,
outfit,
auxiliary
fabrication.
n.achinery
Economic
and
vessel
to
overhead and profit
procurement
and
and production learning effects can
have
significant cost impacts on final vessel costs.
The
need for early design definition for new ships has
put
increased demands for greater costing accuracy at lower levels of
design
detail. The accuracy
skill of the analyst,
of any estimate
is dependent upon the
the quality and quantity of data available
and the time available to make the estimate.
Results seem to
be
independent of the costing technique used.
5. 1
Fig.
estimating
as being supported by a pyramid structure.
database
design
costing
effort.
transform
of
Cost
Several
applicability
models
cost
provide the
trends
criteria
into
means
cost
current
may be selected to
necessary
or
any
to
projected
determine
single most important factor that determines the
is experience.
the
observation
skill
Experience can be defined as the
knowledge (skill) gained from a period of activity that
The
The
of a cost model for any organization.
the analyst
training,
cost
and acquisition defines the foundation for
historical
estimates.
The
illustrates the development of high quality
of practice,
includes
and personal participation.
key ingredient of experience is allowing the time
necessary
to gain knowledge.
Organizations can assist analysts in gaining this experience
by ensuring continuity.
This means supporting long-term planning
which includes hiring full-time cost analysts,
277
providing lead-in
Cost Estimating Building Blocks
(Ref. 60)
Figure 5.1
278
for new
training
of
documentation
additional training
personnel,
costing
techniques
and
required,
as
allowing
access
to
necessary documentation within the organization.
a skilled cost analyst requires an extensive
Even
from
and;
re-calibrate
reflect
departures are made from
large
the
the
trends
The more good
any
data
On the other
the more accurate the estimate will be.
available,
if
CERs to
the
the additional information.
by
established
hand,
requires
collect data; validate it; incorporate it into their
ability to:
database,
Any costing group
to base estimates.
which
database
sized
database,
errors will increase substantially.
involved with data collection is the proprietary
The poblem
for obtaining data from agencies
true
particularly
If outside data is required,
your own.
this
is
outside
of
Of course,
of the majority of the information.
nature
it typically
must be paid
for.
Within
organization,
any
in SWBS format).
are
way to get information
Return
content
or they can be easily
models
Cost
characterized
by
technical
design
estimating
new
used
their
data
ship
during
rapid
costs
useful
from
individuals,
and cost data forms may be intimidating in their
surveys
and/or
gathering
Generally, personal interviews
to them (e.g.,
best
data
and in a form that is
must be sent to the costing group,
the
of
to inadequate communications.
reduces
typically
the problem
length
"buried".
studies
feasibility
turn
around
time
and
The
five
models
requirements.
construction
since
costs
that
were
are
limited
for
examined
displayed important commonalities in two areas: SWBS weight group
279
incorporating the
non-cost parameters for calculating costs, and
on costs. It is important to
influence of different technologies
regardless of the technique used to
that all estimates,
realize
calculate them, are extrapolations from experience.
cost
Different
of
view
the various model
of
right
or
and
design
(e.g.,
assumptions
can
be
to give feasibility cost estimates to within 10-20%
of
All
standards).
manufacturing
expected
cannot be regarded as
any differences can be rationalized from the point
since
wrong,
estimates
the models examined
of
return costs. However, figures as high as 40% have been suggested
2).
see Ref.
(e.g.,
An
the
with
1ts
structure,
of
adoption
available parametric cost model is the
commercially
associated
for
consideration
important
commonality
and
language
approach.
Disadvantages include the proprietary nature of these models
difficul ty
the
outputs.
RCF
The
flexibility,
of
in
getting
FRICE
management
model
appears
terms of its ability
have
to estimate
level of detail with a minimum of required
and
in
their
the
most
confidence
to
a
costs
at
any
data.
Combat systems, electronics and machinery were found to have
the
These
cost leverage for construction costs.
highest
are also very dependent upon the standards to which a
ship
is
designed
cost
controllable
scheduling.
single
and
factor
Costs
by
ship
in any ship
increase
batch productions.
influenced
fabricated.
Possibly
dramatically for
In addition,
quantities and
280
time
particular
the
acquisition
costs
biggest
program
multi-lot
versus
production learning
between
is
is
construction
starts.
There are a number of assumptions made in any cost estimates
various
being
economic,
the
more
managerial,
obvious.
scheduling,
and technical
The purpose of risk
analysis
quantify these effects to produce a range of possible
opposed
to
ones
is
to
costs,
as
the more traditional point or most likely
estimate.
Programmatic and schedule effects appear to have the largest cost
uncertainty (i.e., risk).
estimate
costs,
people
In the final analysis,
do,
models don't
and so there is a highly
variable
uncertainty associated with the cost analyst.
For
for
a cost estimating group to be able to provide
management
operates
decisions,
be:
must
the
committed
organization
to
within
maintaining
guidance
which
its
it
expertise;
responsive to its results, and; allow the group to operate within
the organization
as the need arises
(e.g.,
receiving
return
cost
data).
The
primary
explainable
design
cost
phase
discarded;
benefits
of
developing
a
estimating capability during
are that it shows the:
lost causes
credible
the
and
feasibility
that need to be
the winners that must be pursed more vigorously, and;
the marginal cases requiring more careful examination.
5.2
RECOMMENDATIONS
This
study
has examined a variety of cost models
intended
for
use during the feasibility phase of ship design and some
the
factors that must be considered to incorporate
technique into an organizational structure.
found
here
address
costing
The recommendations
issues of both a technical
281
any
of
and
managerial
nature
5.2.1
General Recommendations
majority of tl:isstudy deals with the documentation
The
models now in use for new ship costing.
cost
instructive
It would
most
investigate the sensitivities and accuracies
to
manner.
models in a more systematic and quantitative
these
be
primary difficulty
of
of
The
of this type would be dealing
with an exercise
with any proprietary model inputs and outputs.
This
inflation,
study discussed the effects of scheduling,
input uncertainty,
technology, plus a variety of other issues in
a rather general sense.
examination
of
any
A more exhaUsive and again, quantitative
of the topics
mentioned
within
would
be
enlightening to the cost estimating community.
Weight
is a primary
cost driver
for the majority
of
CERs.
For the models considered here, SWBS weight groups are defined as
elements with similar trends.
cost
fabricating
techniques
producibility
methods)
applicability
of
(i.e.,
on
CER
the effect of
However,
zone
outfitting
and
development
SWBS cost elements has yet
to
and
other
even
be
on
a
better
the
evaluated.
Current shipyard accounting practises generally do not use
Perhaps
new
procedure for grouping cost elements
SWBS.
can
be
found.
Only cost estimating procedures used in the USN and the USCG
were discussed here. It may prove extremely
seful to look at how
other Navies and Coast Guard equivalents have incorporated a cost
estimating capability into their ship acquisition
difficulty
with
this
suggestion
282
would
be
the
programs.
high
The
travel
expenses,
since
only
by
personal
information be passed along.
contact
Typically,
would
any
useful
this contact would have
to occur 3 or 4 times over a 4-6 month period to be effective for
establishing
a
satisfactory working relationship
between
the
persons involved.
Because of the amount of information collected from
sources
and
the
number of individuals that were
various
contacted
in
regards to this work, it represents a "first cut" at establishing
some
degree of cooperation within the cost estimating community.
Perhaps
more
dialogue
work
between
towards
in this area could establish
some
estimators for selected topics.
developing a dialogue is to participate in
A
sort
good
of
step
professional
organizations working in related areas of interest.
Typically,
a cost analysis requires inputs from economists,
engineers,
managers,
important
that
accountants,
analysts
learn
to
plus other estimators. It is
communicate
with
these
individuals on a personal basis. Developing people skills will go
a
long
way
to enabling a cost analyst to obtain
desired
information
properly
prepared
and
establishing
necessary
credibility
or
through
presentations (e.g., the use of graphics
to
explain and examine trends) and discussions.
It is important that neither the cost analyst nor management
regard
cost estimating as a separate function.
thought
of
in
a
systems
integration
Costing must
approach,
since
be
cost
estimators interface with all the functional-type areas mentioned
above
and operate within the framework defined by the needs
resources of their organization.
283
and
5.2.2
US Coast Guard Recommendations
The cost survey (see Appendix C) distributed throughout USCG
HQ
was
not successful in determining the cost estimating
needs
within the Coast Guard. A similar survey should be carried out on
a
personal interview basis before any further work in this
is carried
area
out.
USCG cost estimating operates in the Ship Building Branch of
the
Office
of
Engineering.
estimating group (i.e.,
The
limited
one individual),
size
of
this
cost
presents several long-
term problems, some of which are addressed below.
Experience
capability
be
is
estimating,
impossible.
With
only
continuity
one
within
individual
involved
this
for
area
in
the
can
cost
CG
is
The only solution to this problem is the addition of
The actual
needs and projected
number can only
resources
the cost
Augmenting
personnel requires
be documented,
be determined by the future
for cost estimating.
estimating
group with
the development of a training
end, it is recommended that
the existing
additional
program. To this
CG costing
methodology
perhaps in a similar manner to that found in
NAVSEA Cost Estimating
The
estimating
and continuity is the method by which experience
maintained.
personnel.
an essential element of any cost
Manual
the
(see Ref. 16).
CERs used for costing
in the CG are based on a
limited
database for a wide variety of vessel types. Currently within the
CG,
the
CFPM (Critical Path Method)
contractors
Network
program
to supply material and labor cost data broken down by
SWBSweight groups. This program should ndoubtably
However,
requires
it
would
be extremely
284
useful for funding
be continued.
to
be
made
available
for
various
projects
to
expand
and
improve
the
database.
Improvements
major
could
functional
include
a detailed shopping
components and high cost
list
equipments
for
such
as
electronics, which can be non-weight related. There could also be
a
formal
the
tracking
VAMOSC
general
of CG operating
system
and support
employed by the USN
costs
(see
similar
Ref.
21).
Since
and administrative (G&A) and shipyard support costs
make up a substantial part of the acquistion costs,
to
can
it would
be
advantageous that they be more accurately known.
Feasibility
applications
should
7) and
cost
have
proven
(e.g., see Ref. 27).
be
models:
studies
capable
model
for
of
Patrol Boat model
using
vessel
model
synthesis
15), Goodwin's Cutter model
(Ref.
8).
In
(Ref.
addition,
outputs should be available for life
calculations
CS
The USCG cost estimating
of interfacing with a variety
the USN's ASSET (Ref.
Tuttle's
useful
cycle
the
costing
the CG developed CASHWHARS spreadsheet
(see
Ref. 34).
Cost
exists
method.
estimates
a capability of cross-checking them using an
The
development
many
gain credibility and consistency
there
independent
recommendation in this area is for the simultaneous
of another
cost model.
The RCF PRICE model
organizations as a means to provide analysts
with a reasonableness check on cost estimates.
uses
if
is used in
and
In fact,
managers
Ref. 27
the RCA PRICE model to investigate the cost of several USCG
patrol
boat
quantifying
concepts.
the
Also,
for
an
cost impacts of various
285
increased
design
emphasis
and
on
technical
innovations, the RCA PRICE model may prove valuable (Ref. 46).
Ideally,
(1)
fashions
the
would operate in the following
PRICE model would be run simultaneously to obtain
(the differences
check
knowledge
pass
CG cost model would be run to get first
(including G&A and shipyard support functions);
results
RCA
CG cost estimating
should
be easily
cross-
a
explainable
(3) the contractor
of each model);
(2) the
with
would have a
expenditure program to enter the shipyard costs,
cost
and; (4) return
from the shipyard would be tracked to obtain a history
costs
a
of
the expenditures and to re-calibrate or validate the cost models.
This
constant
becomes
an
feedback
integral
is necessary
part
of
construction project.
286
the
to
team
ensure
that
managing
any
costing
ship
REFERENCES
1.
Reeves, J.M.L.,
D.D. Findley and M.A. Logue, 1980. ASSET
Theory Manual - Cost Analysis Module. David Taylor Naval
Ship Research & Development Center, Technical Director's
Technology Application Team, Technical Contribution No. 4,
111 pp.
2.
Carstensen, R.V., 1975. Cost Considerations for Electrical/
Electronic System Installation Design. Marine Technology,
12, 378-389.
3.
Anon., 1986. Building Canadian F'atrolFrigates, Marine
Engineering/Log, 91(4), 33-40.
4.
Anon., 1986. A New Approach to Warship Design, Marine
Engineering/Log, 91(4), 52-59.
5.
Hope, J.P. and Stortz,
V.E., 1985. Warships
and Cost
Constraints. Naval Sea Systems Command. 26 pp.
6.
Mills, J., Feasibility Studies and Ship Synthesis Models.
Lectures 7,8,9. Unpublished lecture notes, Professional
Summer at MassachuLsetts Institute of Technology. Cambridge,
MA,
112 pp.
M.J., 1975. Ship Synthesis Model for Coast Guard
Cutters. Ocean Engineering Thesis, MIT, 278 pp.
7.
Goodwin,
8.
Tuttle, J.G., 1981. Synthesis Model for Ocean Going Boats
between 5C) and 150 feet in Length. Ocean Engineering Thesis,
323 pp.
9.
Maritime
Administration
(MARAD), 1986. Relative
Cost of
Shipbuilding, U.S. Department of Commerce, Washington D.C.
10.
11.
International Maritime Associates, 1977. A Study of Ship
Acquisition Cost Estimating in the Naval Sea Systems
Command. Washington D.C., NTIS AD-A046 978.
Shipbuilders Council of America, January 1986 Statistical
Summary,
12.
Flanagan,
Washington
D.B.,
D.C., 2
pp.
1969. A Cost Estimating
Procedure
Recommended for Coast Guard Cutters. Unpublished Report,
Naval Engineering Division, Office of Engineering,
US Coast Guard.
13.
Turner, D.K.,
1978. Marginal
Cost Factors
for High
Performance Ships and their Impact on Subsystem Design.
Ocean Engineering Thesis, MIT, 192 pp.
287
14.
Brennan, A.J., J.D. Burroughs, D.H. Wacker, 1976. HANDE - A
Tool for Integrating the Hydrofoil Preliminary Design Cycle,
SNAME, Pacific Northwest Section, Salishan, Oregon.
15.
Greenwood,
R.W. and Fuller,
of a
A.L., 1986. Development
Common Tool for Ship Design and Technology Evaluation.
Presented at SNAME (New England) Symposium "Marine Computers
1986, Boston, MA, 24 pp.
'86", 17-18 April
16.
NAVSEA
Ship Cos t Estimating.
Entry-Level
Shi
Proceedings
of
A Reference
Document
for the
p Cost Estimator. Naval Sea Systems Command,
Washington, D.C ., 30 September 1986.
17.
The International Society of Parametric
Analysts (ISPA) , Fifth Annual Conference, St. Louis,
Missouri, April 26-28, 1983. Vol. II, No. I, 502 pp.
of The International Society of Parametric
Analysts (ISPA), Sixth Annual Con ference, San Francisco,
California May 15-17, 1984. Vol. III, No. I, 751. pp.
18.
Proceedings
19.
Proceedings of The International Society of Parametric
Analysts (ISPA), Eighth Annual Conference, Kansas City,
Missouri,
May 12-15,
1986. Vol. V, No.
I, 607 pp.
20.
PRICE Systems User's Reference Manual, RCA PRICE Systems.
21.
Visibility and Management of Operating and Support Costs Ships (VAMOSC - Ships), Total Support System Report, Fiscal
year 1985, DD-I&L(A
R)1422(SHIF'PS 5200), March,
1986.
22.
Accelerating Utilization of New Materials. Prepared by The
Committee on Accelerated Utilization of New Materials,
National Materials Advisory Board, Division of Engineering National Research Council. Publication NMAB - 283, National
Academy of Sciences, Washington, D.C., May 1971.
23.
Cost Model U.S. Naval Vessels (Destroyer-type) Final Report.
Prepared for Office of Naval Research, Contract No.
N00014-80-C-0950.
1981.
24.
Shipbuilders
Prepared
by Gibbs & Cox, Inc., August
Council of America, Statistical
Quarterly,
Third Quarter, 1980.
25.
Salmon,
C.L. and A.J. Locke. The Application
of PRICE to
Ships and other Large Structures. J. of Parametrics, Vol.
II, No. 3, Winter 1982-83, pp. 11-17.
26.
Jones,
R.R., D.L. Mayo,
and D.J. Clark.
the RCAPRICE Hardware Parametric
Vehicles,
66-89.
The Application
of
Cost Model to Air Cushion
J. of Parametrics, Vol. 5,
No.
2,
June 1985, pp.
27.
Jeffers, M.F., and R.R. Jones. A Comparative Cost Analysis
of a Hybrid Ship Concept for the Coast Guard WPB Mission.
Ship Systems Integration Department, Departmental Report,
DTNSRDC/SSID/87/DRAFT, January 1987.
28.
Walsh,
S.P.,
1985. An Improved
Method of Risk Analysis
for
the Naval Ship Design Process, Ocean Engineering Thesis,
MIT, 88 pp.
29.
Hutchison, B.L., 1985. Application of Probabilistic Methods
to Engineering Estimates of Speed, Power, Weight, and Cost.
Marine Technology,
30.
Vol. 22, No. 4, pp. 358-372.
Byington, J.A., 1985. The Estimating and Administration of
Commercial Shipbuilding Contracts. Marine Technology, Vol.
22, No. 3, pp. 286-292.
31.
Black, R.L. and J.J. Wilder, 1984. Determining Cost
Uncertainty In Bottoms-up Estimating. J. of Parametrics,
Vol. 4, No. 2, pp. 51-64.
32.
Black, R.L. and J.J. Wilder, 1982. Probabilistic Cost
Approximation When the Inputs are Dependent.
Parametrics, Vol. 2, No. 1, pp. 3-8.
J. of
33.
McNichols, G.R. (editor), 1984. Cost Analysis. Operation
Research Society of America, Arlington, Virginia, 332 pp.
34.
Sappern,
A.S.,
S. Kimbrough,
C. Pritchett
and T. Coe, 1986.
CASHWHARS - Wharton Cash Flow Analysis Model, Version , 1
July 1986, USCG R&D Center, Avery Point, Groton, CT, 44 pp.
35.
Moy, J., J. Vail and S. Williams,
1973. Marginal
Cost
Analysis of Major Ship Design Parameters for DG Shipborne
System Trade-Off Studies. Prepared for Naval Ship Systems
Command, PMS 305, Report No. 6112-74-3.
36.
Judge, S.D., 1985. Minimizing Ship O&S Cost. SNAME Hampton
Roads Section Paper, December 5, 1985.
37.
Mauro, C.V., 1985. The FAST-E Cost Estimating Model. J. of
Parametrics,
38.
Vol. 5, No. 1, 87-102.
Fetchko, J.A., 1968. Methods of Estimating Investment Costs
of Ships, Department of Naval Architecture and Marine
Engineering, University of Michigan, Ann Arbor, Michigan,
June 1968, 201 pp.
;9.
Landsburg, A.C., 1982. Interactive Shipbuilding Cost
Estimating and Other Cost Analysis Computer Applications.
Presented at the IFIP/IFAC Conference "Computer Applications
in the
Automation
of Shipyard
Operation
and Ship Design",
US Naval Academy, Annapolis, MD, 7-10 June 1982.
289
40.
Landsburg,
A.C., A. Jenks, S. Lee and E.B. Schimler,
1986.
Analysis of Japanese and Korean Shipbuilding Prices, SNAME
Chesapeake Section, June 18, 1986.
41.
Jenks, A. and J.E. Larner. A Tanker
Owner's
Perception
of
Newbuilding Costs and Prices in Japanese, North European,
and United States Shipyards 1971 to 1981. Tanker Department,
Exxon International Company, Florham Park, NJ.
42.
Carreyette, J., 1978. Preliminary Ship Cost Estimation.
Transactions
of the Royal Institution
of Naval Architects,
Vol. 120, 235-246.
43.
McNeal, J.K.,
1969. A Method for Comparing
Costs of Ships
Due to Alternative Delivery Intervals and Multiple
Quantities, SNAME Transactions, Vol. 77, 71-103.
44.
Shafer,
W.J.E.,
1983. Cost Estimating
Relationships
for US
Navy Ships. IDA Faper P-1732, Prepared for Office of the
Under Secretary of Defense for Research and Engineering
Special Ass tant for Assessment.
45.
Leopold, R., O.P. Jons and J.T. Drewry, 1974. Design-to-Cost
of
Naval
Ships. SNAME Transactions,
Vol. 82, 211-237.
46.
Spaulding, K.B., 1984. The CONFORM Program - An Update,
Naval Engineering Journal, Vol. 96, No. 3, 94-110.
47.
Rawson, K.J., 1973. Towards Economic Warship Acquisition and
Ownership. Trans. Royal Institution of Naval Architects,
Vol. 115, 43-56.
48.
Keane, R.G., W.C. Sandberg, 1984. Naval Architecture for
Combatants, A Technology Survey. Naval Engineers Journal,
Vol. 96, No. 5, 47-64.
49.
Kehoe, J.W., K.S. Brower, H.A. Meier, 1982. The Impact of
Design Practices on Ship Size and Cost. Naval Engineering
Journal,
Vol. 94, No. 2, 68-86.
5C.
Dart, C.E. 1970. Cost Estimating - Ship Design and
Construction. University of Michigan. Department of Naval
Architecture and Marine Engineering. No. 082, 61 pp.
51.
Summers, L.S., 1973. The Prediction of Shipyard Costs.
Marine
Technology,
Vol. 10, No. 1, 8-15.
290
52.
Kazek, M.S., 1985. A Comparison of Offshore Patrol Vessels.
Marine Technology, Vol. 22, No. 4, 351-357.
53.
Whitlock, R.D., 1982. Estimates
Curves. Journal of Parametrics,
54.
Advanced Naval Vehicles Concept Evaluation (ANVCE) Final
Report, CNO (P-96), December, 1979.
55.
Daschbach, J.M., K.F. Matta and W. Frank, 1983. Regression
in Parametrics: Estimation versus Prediction Errors. Journal
of Parametrics, Vol. 2, No. 3, 3-8.
56.
Moy J. and J. Vail,
at Completion Using Beta
Vol. 2, No. 2, 15-18.
1974. CV Ship Acquisition
Cost Model,
Final Technical Report. NAVSEC Report C-6113B3-75-15,
October
1974.
57.
Starrett, C.O., 1982. Parametric Cost Estimating - An Audit
Perspective. Journal of Parametrics, Vol. 1, No. 4, 3-5.
58.
Cost Analysis Capability at NUSC. Report of the Work Group
Session 15-17 August, 1984. Encl (3) to MAT-01F41 memo 5050
Ser 01F41/410
59.
Jeffers, M.F., The Costing and Design Systems Office at
David Taylor Naval Ship Research & Development Center. PRICE
Presentation to NAVSEA Cost Engineering Task Force, 21
August,
60.
of 28 Sep 1984.
1986.
Unpblished
material.
Costing and Design Systems Office,
Code 1204, David Taylor Naval Ship Research and Development
Center.
61.
Gaffney, R., 1980. Avionics Software
Parametrics, Vol. 1, No.
, 26-30.
62.
Unpublished material. Cost Estimating and Analysis Division,
SEA 017, Naval Sea Systems Command.
63.
Unpublished material. Ships Cost Analysis Division, NCA-2,
Naval Center for Cost Analysis.
64.
Breiling, R.W., 1984, Parametric Cost Estimating and the
PRICE H Model. Presentation to Graduate Program in
Technology Strategy and Policy. oston University, October
4, 1984.
291
Costing.
Journal of
APPENDIX A
SWBSWeight Group Keys
for Costed Military
(Ref.
292
1)
Payload
SWBS
Weight
Group
SWBS
Group
Name
410
Command and Control Systems
440
Exterior Communications
450
Surveillance Systems (surface)
460
Surveillance Systems (underwater)
470
Countermeasures
480
Fire Control Systems
490
Electronic Test, Checkout and
Monitoring Equipment
498
Command
Surveillance Operating
Fl uids
710
Guns and Ammunition
720
Missiles and Rockets
750
Torpedoes
780
Aircraft Related Weapons
790
Special Purpose Systems
F21
Ship Ammunition
F22
Ordnance Delivery Systems (Ammo)
F23
Ordnance Delivery Systems
F24
Ordnance Repair Parts (Ships Ammo)
F25
Ordnance Repair Parts (Ordnance
Delivery Sys. Ammo)
F26
Ordnance Delivery Systems Support
Equipment
F60
Cargo
293
B
AFPENDIX
ASSET
KN Factors
for
SWBS Groups 100 - 900
(Ref.
294
1)
KN VALUE
_
1.000
SUGGESTED APPLICABILITY
IN ANVCE
INTERPRETATION
__
Mild/HT steel displacement
FFG7
MCM
VSS
hull with aluminum deckhouse
1.570
Mild steel plus 25% HY-80
monohull with aluminum
deckhouse
SWACVN
2.191
Conventional aluminum hull
CPICX
LCPC
PCI
SWA4
ACVI
HYD7
LSES
SESCV
ACV3
HYD2
HOC
LCAC
PHM
SES3
SESCVN
5.586
DATA BASE
CASES USED FOR KN
HT aluminum hull
MONO3
DE 1040
SWATH DE
DEI 0S3
PF 109
DDG 2
DLGN 35
DLG 12
DLGN 36
DLGN 25
DLGN 38
CPIC
PGG I(P)
PG 92
PGG
PGH I
DBH
PCH I
2K SES(R)
AGEH I
2K SES(B)
DE 1037
(T)
CASES NOT USED
DLG 16
SWATH MCM
JEFF A
HULL STRUCTURE (SWBS 100) GROUP KN VALUES
295
DLG 26
PGH 2
JEFF B
PG 93
PHM I
ARCTIC SEV
KN VALUE
INTERPRETATION
-- ---------1.000
CODOG/CODAG
power plant,
SUGGESTED APPLICABILITY
IN ANVCE
CPICX
LCPC
ACV3
HOC
PHM
HYD2
LSES
SES3
DATA BASE
CASES USED FOR KN
CPIC
PG 92
PGG (T)
PGH I
PHM I
2K SES (B)
PGH 2
DBH
PCH I
2K SES (R)
DDG 2
DLG 12
DLG 16
MONO3
PF 109
SWATH MCM
SWATH DE
SESCV
JEFF A
DLGN 35
DLGN 36
high speed marine propulsors
and straight drive train
1.502
GT/CODAG power plant, high
speed marine propulsors and
right angled drive train
1.979
Steam turbine power plant,
low speed CP propeller and
long shafts
2.345
GT power plant, low speed CP
propeller and reduction
gears, with special arrange-
HYD7
PCI
-
FFG7
VSS
MCM
SWA4
ments as in SWATH
3.280
GT power plant with complex
drives
ACVCI
LCAC
SESCVN
3.436
Nuclear steam pressurized
water reactor power plant,
low speed FP propellers,
straight drive train
SWACVN
DLGN 25
DLGN 38
CASES NOT USED
DE 1037
DLG 26
AGEH I
PROPULSION (SWBS 200) GROUP KN VALUES
296
DE 1040
PG 93
JEFF B
DE 1053
PGG (P)
ARCTIC SEV
KN VALUE
SUGGESTED APPLICABILITY
IN ANVCE
INTERPRETATION
2.036
3.719
__
__
-_
1.000
DATA BASE
CASES USED FOR KN
Conventional 60 HZ power,
steam or diesel generator
drive
CPIC
FFG7
MON03
SWACVN
Conventional 60 HZ power,
light diesel or GT generator
drive
HYD7
HOC
LCPC
PCI
Very low 60 HZ or mixed
SES3
SWA4
ACVI
PHM
MCM
LCAC
SESCV
DE 1037
PF 109
DLG 16
PG 92
DLGN 25
DLGN 38
DE 1040
DDG 2
DLG 26
PGG (T)
DLGN 35
SWATH MCM
CPIC
PGG I(P)
PGH I
AGEH I
PGH 2
DBH
PCH I
2K SES (R.
PHM I
JEFF B
SESCVN VSS
60/400 HZ, GT generator drive
12.684
All 400 HZ, GT generator
drive
CASES NOT USED
PG 93
ELECTRIC PLANT (SWBS 300) GROUP KN FACTORS
297
JEFF A
DE 1053
DLG 12
DLGN 36
SWATH DE
KN VALUE
INTERPRETATION
SUGGESTED APPLICABILITY
IN ANVCE
1.000
Simple control systems,
minimal electronics
CPICX
3.153
Modest control systems,
sophisticated electronics
FFG7
SWA4
6.906
Complex control systems,
sophisticated electronics
weight critical ships
ACVI
ACV3
HYD2
HYD7
HOC
LCAC
LSES
PHM
SES3
SESCV SESCVN
LCPC
PCI
MCM
MONO3
SWACVN VSS
DATA BASE
CASES USED FOR KN
DE 1037
DLG 12
CPIC
DE 1040
DLG 16
PG 92
DE 1053
DLG 26
PG 93
PF 109
DLGN 35
SWATH MCM
PGG (P)
DLGN 36
SWATH DE
DLGN 25
DLGN 38
PGH I
PHM I
JEFF A
PGH 2
AGEH I
JEFF B
PCH I
DBH
ARCTIC SEV
CASES NOT USED
DDG 2
2K SES(B)
COMMAND AND SURVEILLANCE
298
(SWBS 400) GROUP KN FACTORS
PGG I(T)
2K SES(R)
KN VALUE
SUGGESTED APPLICABILITY
IN ANVCE
INTERPRETATION
1.000
Steam propelled displacement
ship
1.528
GT propelled displacement
4.161
5.370
DATA BASE
CASES USED FOR KN
DE 1037
DG 2
DLG 26
DE 1040
DLG 12
DE 1053
DLG 16
CPICX
FFG7
LCPC
MCM
MONO3 PCI
SWA4 WACVN VSS
PF 109
PG 92
DLGN 25
CPIC
PGG (T)
DLGN 35
PG 93
PGG (P)
DLGN 36
Fully submerged hydrofoils
HYD2
HYD7
HOC
PCH I
AGEH I
DBH
Air cushion vehicles
ACVI
LSES
ACV3
SES3
LCAC
SESCV
JEFF A
2KSES(B)
ship
ARCTIC SEV
2K SES(R)
CASES NOT USED
PGH I
JEFF B
AUXILIARY SYSTEMS (SWBS 500) GROUP KN FACTORS
299
PGH 2
PHM I
KN VALUE
INTERPRETATION
1.000
Conventional displacement ship
1.857
Weight critical
ship
DATA BASE
CASES USED FOR KN
SUGGESTED APPLCABILITY
IN ANVCE
CPICX
MCM
SWA4
LCPC
FFG7
MONO3 PCI
SWACVN VSS
ACVI
HYD7
LSES
SESCV
HYD2
ACV3
LCAC
HOC
SES3
PHM
SESCVN
DE 1037
PF 109
DLG 16
PGG (T)
DOGN 35
SWATH DE
DE 1040
DDG 2
DLG 26
PGG (P)
DLGN 36
PGH I
PHM I
2K SES(R)
PGH 2
AGEH I
2KSES(B)
DE 1053
DLG 12
CPIC
DLGN 25
DLGN 38
PCH
ARCTIC SEV
CASES NOT USED
PG 93
DBH
OUTFIT AND FURNISHINGS (SWBS 600) GROUP KN FACTORS
300
PG 92
JEFF A
SWATH MCM
JEFF B
KN VALUE
1.000
3.401
SUGGESTED APPLCABILITY
IN ANVCE
INTERPRETATION
Conventional
Weight critical
armament
displacement
ship, light
ship
CPIC
MCM
SWA4
ACVI
HYD7
LSES
SESCV
FFG7
LCPC
MONO3 PCI
SWACVN VSS
ACVC3 HYD2
HOC
LCAC
PHM
SES3
SESCVN
DATA BASE
CASES USED FOR KN
DE 1037
DDG 2
DLG 26
PGG I(T)
DE 1053
DLG 12
CPIC
DLGN 35
PF 109
DLG 16
PG 92
DLGN
SWATH .MCM
PGH 2
DBH
SWATH DE
PCH I
2KSES(B)
PGH I
PHM I
38
CASES NOT USED
DE 1040
DLGN 25
JEFF A
2K SES(R)
ARMAMENT (SWBS 700) GROUP K N FACTORS
301
PG 93
DLGN 36
JEFF B
PGG (P)
AGEH 1
ARCTIC SEV
KN VALUE
--
INTERPRETATION
SUGGESTED APPLICABILITY
IN ANVCE
DA'rA BASE
CASES USED FOR KN
;--
__
1.000
Foliow ship before
2.881
Follow ship, 1970 and after
12.888
Lead ship, unsophisticated
weapons
CPICX
MCM
LCAC
26.064
Lead ship, sophisticated
weapons
ACVI
HYD2
LESES
PHM
SESCVN
VSG
ACV3
HYD7
MONO3
SES3
SWA4
DE 1037
DODG2
PG 93
SWATH MCM
DE 1040
DLG 12
PG 92
DE 1053
DLG 16
DLGN 25
LCPC
PGG I(T)
PGH I
AGEH I
JEFF B
2D SES(B)
PGG (P)
PGH 2
DBH
ARCTIC SEV
SWATH DE
PCH I
JEFF A
2K SES(R)
FFG7
PF 109
PHM 1
1970
HOC
PCI
SESCV
SWACVN
CASES NOT USED
DLG 26
DLGN 38
INTEGRATION/ENGINEERING
302
(SWBS 800) GROUP KN FACTORS
CPIC
DLGN 35
KN VALUE
1.000
4.254
INTERPRETATION
Simple tooling, limited trials
Complex tooling, extensive
trials
SUGGESTED APPLICABILITY
IN ANVCE
CPICX
LCPC
ACVCI ACVC3
HYD2
HYD7
LCAC
LSES
MONO3 PCI
SES3
SESCV
SWA4 SWACVN
DE 1037
FFG7
HOC
MCM
PHM
SESCVN
VSS
DATA BASE
CASES USED FOR KN
DE 2040
DDG 2
DLG 12
CPIC
PGG (P)
DLG 16
PG 93
AGEH 1
DLG 26
PGG 1(T)
PF 109
DLGN 36
PHM I
JEFF B
DLGN 25
DLGN 38
DBH
2K SES(B)
DLGN 35
SWATH DE
JEFF A
CASES NOT USED
DE 1053
PGH I
ARCTIC SEV
PG 92
PGH 2
2K SES (R)
SHIP ASSEMBLY AND SUPPORT SERVICES (SWBS 900) GROUP KN FACTORS
303
SWATH MCM
PCH I
APPF'FENDIX C
COST ESTIMATING SURVEY
The survey is designed to examine the perceived needs of any
individual who must answer the question; "How much will this
cost?" This is meant to include all levels of management involved
in any type of ship acquisition program, from concept design
through to production,
and into operations
and support,
as well
as official cost estimators.
I hope to accomplish two tasks with this survey. The first
is to develop an awareness of the different cost estimating needs
that occur throughout the various phases of a ship's life. The
second is to attempt to match these needs with the
characteristics of different cost estimating techniques. In other
words, to improve the responsiveness of cost estimating models to
user expressed needs.
questionnaire
The
hand
column
costing
have
is divided into two columns.
is for any individual who perceives
information and the le-t hand column is for
used
any
sort of cost estimating
3 (I 5
The
right
a need for
technique.
people
ship
who
NEW SHIP COST ESTIMATING SURVEY
NAME
..........................
...
RANK .........
PHONE ..............
POSITION ............................ DEPARTMENT .............................
PLACE OF EMPLOYMENT .......................
·
·
·
·
·
COST MODE[L
PERFORMAN CE
YOUR FUNCTION ...................
~~~~~~~~U·ff··l··ll
~
·
·
·
~
·
l
ll··1·····l···l·
USER
I
MPORTANCE
Name of cost model (e.g., ASSET,
II
RCA PRICE,...)
Under the
following
headings
indicatethe cost-related
informat
ion
that
is incorporated into your cost
model and how this is done.
Under the following headings,
1indicate the cost-related information
you feel should be incorporated into
a cost model and any ideas of how
this can be accomplished.
Technology
(e.g. ,technology available, state-of-the-art desi gn
. . .
·
· · · ·
. . .
·
·
· · ·. · ·
· ·
mwll·wll·.mmmmm····!·~a.#··mm··!·
·m.m.
m··m····l··w··laam···l·m···m·l·m·····~··
influences, . . .
. .
.m...mmm..·
. .
.
.
.
. .
m.m....m...
. .
.
.
.
Design
ship configuration,
(e.g., weight, margins, design standards & practices,
component/suLbsystemselections,...)
IM····w·mlmmmmmmmm·&C
Manufacturing
(e.g., construction method, degree of automation, too
learni ng,.
i ng,
. . .
..
.
.
.
.
.
.
.
.
.
.. ...
. .
.
.
.....
...
..
.
.
.···m·
. . .
Programmatic
(e.g.,
type
of
a.........
........
..........
·
·
··
·
·
·
·
i· ·
·· ·
·
·
· ··· ···
·
·
·
,·
·
· · i
· m··
· ·
·J··,,··
,·w·
·
m·mm
ng j
. . .. .
orders I
contract, competitions, scheduling, cost of change
..................................... .................
..
. . .. .
. . ..
·m
·
·
.
.
.
.
.
.
.
· ·
·
.
.
.
.
.
.
.
i/imii
. . ..
.
·
iili·i
. . .. .
COST MODEL
PERFORMANCE
USER
IMPORTANCE
Economic
(e ?.g., escalation, inflation rates, discc)unting,...
. .
. ..
.
. . ..
.
. .
. ..
.
.
..............
. .
. ..
.
. . ..
.
. .
. .. .
.
..
· ·
·........
·
·
·
·
·..
·
Cost Breakdown
(e.g., nonrecurring, direct,... )
. .
. .. .
. . .. .
.
. .
. .. .
. . ..
. .
.
.
.
.
.................
.
.
.................
Operations & Support
(e.g., maintenance scheduling, reliability, m odu lari ty,...)
. .
. .. .
................
...
. .
. ..
.
·...
.
....
·· mmm·.
·
.
.................
Other
I
.....................................
I .......................................
I
I
I
For the following features, use
performance ratings from 1 to 10 (1
for total disagreement;
10 for
For the following features, use
importance ratings from 1 to 10 (1 for
"nice to have"; 10 for extremely
extremely important)
I
mportant)
........
Minimal training is required to use the model ..............
..... ...... The model can be calibrated using return data ..........I.....
..... . The model is based on data reflecting the latest technology .......
.
:
The model is able to
..... .............. (a) generate internal cost targets ...................
.....t........... (b) prepare planning/budgetary estimates ................
..................... (c) perform trade-off studies .......................
..... .............. (d) validate outside industry bids ...................
The model is able to provide estimates starting in the
....
I...................
conceptual
design
phase
.................... .....
...
The model is easily accessible at reasonable cost ...........
ft.......
.....
..
The model is developed in-house and can be easily modified .......
I
The model gives an accurate estimate
of
I
.(a. ................... (a) construction costs ..........................
.
...................
....
(b) programmatic costs .....................
I.....
..
.
......................
(c) life cycle costs ........................
I
The model is available
to agencies/individuals
I
i..................... with a need to know ..........................
1
Other
f ·
·
·
·
·
··
·
·
·
·
·
·.
·
·
·
·
·
·
·
·
·
·
·
·
· ·
·
· ft
· ___
USER
IMPORTANCE
COST MODEL
PERFORMANCE
Indicate an approximate accuracy (as
a percentage)
for your cost model
during concept/preliminary design for
construction costs
(a)
life cycle costs
other
(c)
(d)
..
Is
(a) construction costs ......
(b) programmatic costs ......
......
(c) life cycle costs
programmatic costs
(b)
t
Indicate
your required accuracy (as
a percentage) during concept/
perliminary design for
(d) other
.........
:here an area where the model
wel 1?
perf orms especially
...................
..............
...................
..............
Is there an area of cost estimating
parti cular
that you feel needs
attention?
........
0. ...
....e
,
....
............
............
What is your definition
of well?
.....
...................
Is t here
perf orms
............
..............
an area where the model
poorly?
Whatis the reason(s) that this is
so important?
..............
..............
What
is your definition
For what phase(s)
acqu isition
appl icable?
...................
.
.
.
.
of a
of poorly?
For what phase(s) of a ship
acquisition
program are you most
ship
is this model
program
interested in receiving costs for?
....................................
· · m·
mlmwmm.·e
....................................
...................
...................
..............
3(-)8
.
APPENDIX
D
SWBS GROUP BREAKDOWN FOR THE
NCA TWO-DIGIT COST MODEL
(Ref.
1)
Cc6T GUUOP:
Sd S N.
IA
DESCRI PION
I
- --111
1
113
SHELL PLATING, SURF. SHIP AND SUBMARINE PRFSS.
HULL
ILNNERBO1
SHELL APPENDAGES
STANCHIONS
I'T. FRAMING, SURF. SHIP AND SL4ARINE
LI
T
114
115
116
PRESS.
IlI.L
117
121
122
123
124
131
132
133
134
135
141
142
14 3
144
145
149
166
TRANSV. FWAMING, SURF. SHIP AND SMARINE
HUI.L
LfCI'IJUDINAL SI'HUK'UM1AI,HULIIH'ADt
TRANSVERSE STRULXRAL BUIIHEALk
'TRuNKS ND ECIOSURES
SYS'I'TE
BLULKIIEADSIN 'IORPEllU tlI'EC'I'IN
MAIN LECK
2NL DECK
3RD EC'K
1 4'1i ACK
5' 11t DECK AND L*.CKb L-L w
I 1S' PLA'FtrU)t
ND PIATFI
31)WPLA'I'tU
4''11 PIA'rt'O1
PIA'I'CF@
15i
I ILA'S
S1t
s
C'OSTGRfJUP: _
N.)S
_
_
IF-(:RI l'I(N
I
'1n) Fi{Rr
151
DECKHOUSESTRU'IJ
152
lSF
153
2ND DECKHOUSE EVEL
154
155
156
157
158
159
164
3RD DECKHOUSE LEVEL
4'TH Lt;CKHUSE LEVEL
'LEAVEL
5'THlCKHUSE
b'IT DCKHOUSE LEVEL
7Trl UECKHOUSELVEL,
MTtH
DECKHOUSELEVEL AND AOVE
LAI.ISTIC I'IAT{M(;
[ECKOUSE EVEL
310
[LVE[,
PRESS.
cOST GRmUP:
!(
.
ND.
SWS
-
-- -L--
_
_
_
_
_
I---
_
_
_
_
Mt2SCRIrli
I------1
1
182
PROPULSION PLANT FOINTIONS
183
ELECTRIC PANT -r IIDATIONS
184
CMAM
185
AUXILIARY SYS'I'ItS
186
OUIFI r AN
1
187
CUS
GROUP:
S
N
Nt.
-I
f
I
1
FDATICS
F'lNISIINGS
RrATI4S
AJmwwrFUNT
ID
I' 'l(IN
EtlIV.
A
FOImN;S,
STRUL'iRAL CASrNrG,
WEI 1EMS
STACKS AND MKS (MBINED STAC'KAND)MAS')
SEA CSTS
16,'
165
.JNDATIONS
t
I*-'
I
162
AN) SURVEILIANCE FOUNDATIONS
1
S(AR
LX1)E
I
167
168
169
1
IULL SI1UCTIAL CLL6URES
C)SlUES
DECKUSE bSTrHULI'IJHAL
SPECIAL PURI4SE CLlUHkES A) sL,''RWnJi;S
171
1MASTS, lOWERS, 1VrRAOw
172
KINGWSIS AND sUP*r
311
FWVtES
ca'r GFUP:
E.
-bS
2A
_
_
I
I
r
I*.SCHI
I
_
I
PR[PULSION BOILERS
CAS CENERATIDRS
MAIN PROPULSION BATTERIES
MAIN PRPULSION FUEL CELLS
PRPULS ION SEAM JB~ [NES
PROPULSION STEAM ENGINES
PKPULSION II'ERNAL CBI3ULSTICNENGINES
PROPULSION GAS TURINJES
221
222
223
224
231
232
233
234
235
236
237
241
242
ElICTRIC PPUI SIC1J
1
1
1
POUISION
SEIF--CCWNTAINED
1
253
254
255
!
PIUPULSION CWLI'IES AND CUPLI NGS
MAIN STEAM PIPING SYS'"
JLz'IDRS
C(NDENSFRS AND AI
FEE) AND LCD0*NSA'I'ESYS'IlT
-Sc-).
---- ------
c~zcr GI&JP:
A3
SWS N) -
I*.SC'RIPION
243
PFDPULS ION SAFrING
244
PRPULSIO
245
PIRPUISORS
1
246
CT'
GI4_IP:
S4US NJ.
_
251
259
_
SYSTEMS
AUXILIARYPULSITON LVICES
PIOlUIS (IN RFIXI'I(N CARS
SHAFT BEARINS
PROPULISORSIfOU
AND FIKIIS
2C
DESC' I I'ION
I
I
7
1
1
I
CBLUSTIOI; AIR SYSTEM
UI'AKES (INNER CASING)
312
C6'
GINUP:
2D
).
SWBS
DESCRIPIrILON
252
256
258
261
262
264
!
C6T Cl EIP:
SYSTFPI
PIROPLISIOCNOIR')L
CIC(JIATING AND C(XtiNG SEA WAI'FR SYS1EM
H..
I
sP.
I DRAIN SYSTEM
P'JEL SERVICE SYSTlEIS
MAIN PRORII.SION 11JUE OIL SYSTE]U
1BE Oi)1. FllI., TANSPER, A%%)RHIFICATI(ON
A
SWi3SN.
IESC'RIPFr(N
I
311
SHIP SERVICE K)WER
NERA'rON
312
3 14
FMF;
I(M'Y }$'
LRA 6H
[A4Eu (ONVEIk'IN WI1[[
34
341
ssr
111fBE0II.
[ED ESEI:. SUtlq)!el' SYSTE'MS
342
34 1
TIRDBINE
:;bl'l")I'r
711'
.'YS'I'.IIS
Cor GWPP: AI-SI3S
).
|
Fl:RPI'(N
313
321
I
1
3ATIERIES A
SERVICE FACl, I''IrS
SHIP SERVICEt iWEtH CAbk
322
323
324
I
IME
Y
CAS:AIrY
9./I'CHEA
PowE.R CABIE SYSr
iJH
CABLE SYSIAW
AND PANUE5S
331
IUlTING
DIS'TRIBUION
332
I,IGHTING FIJUES
I
313
COiT GIP:
S
4A
N.
..
I
[ESCRIPrN
-I
_______
I
421
422
423
424
426
427
.
1
j
428
NAVICATION
QaNfL
P
431
SWITH0AR3S
432
TELEPHONE SYSTet4S
4 33
4 34
/
FO
IIT0RING
I.C.
SYSTEMS
IUAO
ING SYSTEMS
E24lTERtAII#METAD TRAINING SYSTEMS
4 35
4 36
O)ICE 'IUBES AND MESSAG PASSING SYS['EMS
ALARI, SAFETY, AND 'ANING SYSTIS
4 37
INDICATING, ORDER, AN) METERING SYSI'S
438
INIT'ATED (XTKUL SYSTEMS
4 39
443
47 3
474
475
RECODING; AND TE'lVISION SYSTEMS
VISUAL AND AU)DIBLESYSrEMS
'IOPEDO DECOYS
DECOYS((mTER)
L)XAL SI;NG
j
476
491
492
491
494
495
MINE OITE EASURES
EILECRNIC TESr, clECKrI.,
AND MWIITKRING
UrEN
I
FtIITXr fIL
AND NSTRLN4r [LAN)ING SYSTEMS
IJN C3BAT IIATA PECXESSIt; SYS"IS
METIEX* G(AI, SYS]S
SPECIAL, FPRI~JH
E INEILLI[(;Ei'E SYSTEMS
I
I
(-D''GRUP:
S4WtS ).
_
_
_ _._
I
IESCRIPrI(N
412
i
1
413
414
415
417
441
!)IGIAL
UATAWI'1tHCI
1
NII'EHFACEtJIiIN
IGTrAI. IATA lMINICA'rI(*;
1 (A.MN) A
I rl)a). ANAL-X; .'14I'1
HA)IO SYS'r24S
411
LATA DISPLAY (JP
ArA PX'FSSING c1P
1
442
UNuLwUA'rESYS:rS
444
445
446
451
452
4533
454
TEIM4T1UY
7
SYSI'11;
'I1Y AN) FSIMILE
SYSTE4S
StUH IrY EIU I1wT
SYSl'EMS
SUfA(':
SEAI' HAIt
AIR StAA'II
UiAR (21))
AIR SFARCH FADAR (JD)
AIRCHAIT' CINI L APPIACII
FIIR
455
456
459
461
462
463
464
465
471
472
481
482
483
484
489
.
NoEL-EW ICAI/ELTELMNIC NAVIGATION AIS
ELECTRICAL VIGATION AIDS (INkL NAVIG. L(;rrS)
EIlXTR
IC NAVIGATIONSYSrt-S,
ADIO
ELETNIONIC VIGATION SYSTEMS, UCIurlCAL
EIEC'ICAL
NAVICATIO SYSTEMS
INERIrAL
VIGA'ION SYSTmS
1
g
1
II*NIrIlA'IIlJN
SYSI.S
M IRSi
(IFF)
MUL'I'IPLE MI: RAIAR
SPACE VEIItL'IE EL1tIIC
TRACKING
AL'i VE R
PASSIVE SINAR
!MULTIPLE MODE
CLASSIt'CA'I'l(IN 9UNAR
BA'THI'HEWOAPH
ACriVE L (INCI. C}MINATION ICIVEPASSIVE)
PASSIVE; EL
GUN FIRE OIWrDL SYSTEMS
MISSILE FIRE XOtCN~LSYSTEMS
UNElWATEH FIRE
'll
SYSTEMS
fInX;Ai'rE) FIRE aWlUL L SYSrtMS
WEAF)N SY:.ruS
11'ICIti
314
OaTG CGRUP:
5A
1
e3S W).
---
-
I
---
--
_7-
ITESTRIPrl(N
511
CEATIG SY
I
511
Ct"lIPARWIT/
512
ITING
SYSIEM
VENTILATION SYSTW
513
514
MACIINERY SPACE V2TIIATI'N
516
SYSiTE:
AIR CTrITlDIlING
REt RIGl-ATI'N SYSTrEM
ILERS AND CMIER HEAT
AUXII,IAHY
1
517
Cwr GiirJP:
SYSTEI
J'KHES
58
---
SS I). I
DECRIPrl(N
521
522
AND FUISIIING (SEA WATER) SYSTEM
FIRiA[N
SPR I NKLER SYbSl'E
52 3
WASIfUX)
524
526
AUXILIAHY SEA WATER SYS'I:M
SCUPPERSAND IXK LIAINS
527
528
529
FIEMAIN _YiIATED SERVICES - (HE
PIJMBI NG DRAINAGE
I)HAINA(E AO 3AIJAS'ItNG SYSTE4
5 31
5 32
PIANT
I)ISTIIJING
XCI).ING WA'IEH
533
534
5 35
536
541
542
F frA8LE. WATER
AUX. SEA AllD IRAINS WITlHIN MACHINERYBOX
AUX. STF
Am1 RAINS OUISIDc MACHINERYBOX
AUXII,IARY t'ThSH WA'rR cXof.lNK
SYSTE
SIIIP
lEl, ANM)FtJEL (XI1'TSArING
fW24E FUELS
AVIATION AN{)(t,NIAI.
1
1
543
SYb'ti
AVIA'l(iN A
u-ER AL
PuHH6E
I
HI(A'rPING OILL,
544
545
549
551
LIQUID CAFUD
TANK HEATING
SPECIAL FELL AND)LLURICANIS, 10N)ILING A)
CTIPRSSED AIR SYSlEMS
552
S53
(CIPRESSEL GASES
0o2N SYST:1
554
IP 8BU
'IE
555
556
557
558
565
593
594
SInVA.R
EXTXrIN;ISiiINL; SYSTIES
IIYIJRAULIC FWII) SYSTI
LIQUID GASES, CAlML
SPECIAL PIPING SYSIEWS
AND IEEL SYSTE4S (SURFACE SHIPS)
'IN
.
NVIKOEJrAL R)UA'ION (XNrR)L SYSTEIS
ISUUIINE ESCUE, SALVAGE, A)
315
SURVIVAL SYSTEMS
--I----GaEJP:
CT
4SNO1C.
5C'
*CIr
I
561
562
568
1
SiTERING AND DIVING (fl",
RII)ER
MANEUVEING SYSTIMS
CLIT GItUP:
50
57
IF RE
9-- -571
--A
I
SYSTiMS
YS
I-c;CRREPLENISlIt4MI'-Al-SEA
1
i (N
-PrIl
SYSr-M,
I
572
SIilP SIORES AN
ElUIE
573
574
581
CAW) IDte)LINGSYSliS
582
WXDRIN, AND lING
r HIANLING SYST'EIS
VhlrICAI. REPLkNISIEMEF_ SYSTIIS
A'IWIf hiA.IN;
AND SITJW_ SYS'TMS
(WATr, fAT
584
SYSI4MS
WHAI)I,IJG ANI) S'lWAC
SYI'rIMS
]
OR, GATE, RAMP,
MFUHANICAI, (YPERATED
TItkNrAI3E SYS'I'Im
tlEVATING A) HwRACrING (AR
AIRCAIF
lA?,I.IN;,
.SEHVICING AND SLM
~85
MISCEIJANE'XlS M'IANICAL HA ,IING SYSEIMS
uMIHuMJAND IVt
SUP1J~' Am P'rr(7'1C
SYS-ErmS
AND HAN)LING tH LIIRIcIATE
'11111I;, LAI,'lIN
S YS'iNS
IIANI)1.IN; SYS'2S
tIXJ DIVER AN SMESIBIL
589
592
595
596
S9
I
SALVAG S4JPIHT SYS1TIS
316
COST GIlJP:
6A
S ND.
PSCRIPFrIIN
A-
_-----605
1o['
----
_-----
-
F Ad) VEtfIN P'ING
b611
HJU. FTrtNS
612
IAIIS,
H
61 3
RIC(IN;
625
AIR)HR PR'IS,FIXtl)
SrANCHIcNS, A)
L'IFELINES
AND CANV7S
R(ILIGHM'S,
A)
crlsr CR tip:
Sis
N).
I
£SqrR Prl
N
I
-
621
b22
Nl9-STHLUC URAL BIJULHEALE
FILJR PArfI
A) (ATINGS
I
623
[AIdRS
624
t]N-STRL'VURAI.
637
51IF'AliI N(;
C0T GlUP:
CIJ{H
S
6c
N).
Di3S I
IJFSHIP'I(IN
602
603
604
631
IILL
ESIGNATING A) MARKING
L*AFT MARKS
DCKS, KEYS, A
TAGS
PAlmrING
632
ZINC OTING
6 33
CAT1OiOIC PlIEX'TION
634
6 35
D)CK NVERING
IIIIIL INSUIATION
6 36
639
HUILL DWMPING
HADIATI(IN SiEIING
317
WNXkS
CFT GaCUP:
SWBS N).
I
654
655
56
b64
1
1
I
1
671
672
i
C~T6fT
CRP:
638
641
642
643
644
645
iESCRIPrFIC
SPACES
L'ILI'Y
LALURY SPACES
TRASH DISR.3AL SPACES
STATI(NS
AUE (ITl)L
'TESAREAS (INCU)ING
ORKSHOPS, LABS,
I'X)LS, EUyI NI'r)
665
B3S ND.
6D
RF(:AB.LE
LXKErI AND SPECIAL SWA.C
I.MS
IC;UUIMS AND ISSUE
6E
IPrl(N
!IFXR
I
PE:FHIGERATEDSPACES
E)MESSING SPAC'ES
OFI'CER BRTING
MJNCCIMISSI.Nl(EDOFFICER bLERHING A) MESSING
[
SPACES
I
AN) MESSING SPACES
ENILISTED PERSCONEL BtEING
SPAC'ES AND FIXJitS
.SANITAHY
SPA('ES
LEISUR4EAN) (lCMITY
1
I
651
CMIISSAY SPACES
652
65 3
661
662
MEDICAI, SPAC(S
L*NrAL SPAC'b
OFFICES
MACFIINERYCINT).,
663
ELECrJICS
318
CEEIRTS
F-URISIIMS
Cl)IfM L CEIrERS FURNISHINGS
:
C6Tr GIXS:
SawS N).
GCNERALARRANCE4
701
7 11
712
713
721
722
- WEAPONRYSYSrE
IAN)II NG
ITIN
AMMNL
All' I l N $~ITAGE
IAtICHING; IVI1CEtS (MISSILES AN) RHCKETS)
723
724
72)
72b
728
7 29
731
72
733
741
742
743
751
752
753
754
761
762
DRESCPrIcN
I
I
1
I
1
I
I
I
E CAPSUlt
MISSIIE,
IU'KET, ANLi(UIL'
ItAN)lING
M1ISSILE
MISSIIE
MISSI
MISSILE
SYSTI
E
AND K('Kbl' Sn'&IMAC
IiYIMAULIC.S
MISS.L
II.H
LALN
'rNG
(1IIIt'SATIKH
UI
tK)L
ItPLHATlHE (JINIWLL
MISSI1 E tIEA'rIl , (CX),lIN[;,
rD Am(2MI
TESr
MISSIllE K1I'I!m[NG,
MINE IAL'lIIlNG Il:VICES
,I1A
LINGK
MINE
MINE SIRtWL
IUEMIIIClIWCE LAL 'HING DILVICES
IEfll CHARGE YAN)LING;
DEPI C1HAE SIfML;
'IORPEDOI3ES
'TIRPEDO IANDLING
1r>PEDO SIi 7#A(Z
SiIA~ INE TIRHPEJO EJETI(N
U1ALI,
AIMS AND PYROTECHNIC [AILMNClNG EVICES
MALU,ARMS AND PYRfIECHN IC HANDLING
9AI1, AI4S ANDPYFIYIEIJNIC SIrdE
CA.) MIINTI(fIi
CAR) MUNITI(IN5 IiANDL[N
'AGE
CAR ) MUNI''I()NS S
7631
770
772
773
HAM.,INI
782
Altl'IW-I' iLA';i) WNJS
783
AIRClAIT RELATED WEAPNIS SWAGE
SPECIAL WEAI 5 IMIANLING
SPECIAL WEAPM STOMGE
MISCEIANLU.S (I4U1AtCE SPACiS
792
793
797
I
319
APPENDIX
ANVCE SHIP
E
AND SWBS GROUP CERS
SAMPLE
(Ref.
532C)
54)
Vehicle
ACV 1
ACV 3
A'L(N)
AVP
VhcI
Definition
I
_ _
Noaunal 1000-tonne ACV
Nouinal 3000-tonneACV
Air loiter aircraft
Nuclear-powered,air loiter aircraft
Advancedpatrol aircraft
CPIC
OLX
Coastal patrol interdiction
FAB
FMI 7
Pully air-buoyant LTAvehicle
HYD 7
HYD 2
HOC
Naounal, 700-tonne,
LCAC
LCPC
LSES
Landing craft,
Landingcraft,
RMNO3
Nomnal, 3000-tonne, advanced-technology
monohull
Long-range cargo aircraft
IANVCE Proi- BaselineIEx'stina
ect Design .,.sDesign Des xcn
X
X
X
X
I
X
X
craft
X
(USAF)
X
x
X
PER-class, guided-mussilefrigate
X
high-speed hydrofoil
X
X
Nounal, 2000-tonne hydrofoil
Hydrofoil, ocean cortatant
X
air cushion
vehicle
planrng craft
X
X
x
Large SES
X
x
x
PC I
PiM
Nmtunal, 1000-tonne planlrq
SAB
SES 3
SES CV
SES CVN
Semlair-buoyant LTAvehicle
Ncounal, 3000-t ne SES
S/L(L)
Large, sea loiter aircraft
S/L(S)
S/L(V)
S
4
Small, sea loiter aircraft
Ship-based,V/STOL, sea loiteraircraft
Nonunal,4000-tonneS9
frigate
MNir-ceountrmre asure
STH shiD
Nuclear-powered,
SH
carrier
i
craft
X
Guided-missile, patrol hydrofoil
SWA
SI
x
M
CVN
vS
WIG(H)
WIG(O)
I wJl~rc
ZI X
ZPG X
SES carrier
Nuclear-pcwered SES carrier
X
X
X
X
X
X
X
X
X
X
V/S$L, support ship
x
Hard-ernd-plate
WIG vehicle
Out-of-round-effect,WIG vehicle
ChFed~rrllat
Patrol,,.,
i Patrol,
vei;l
L.
Lnr.
vehicle
321
.
ls
X
X
I
I
x
I
I
I
!
-
l
:.
---
- -
- l
-
I
- " .
I Special-I
I Air Vehicle Groups
Surface Vehicle Groups
Purpose I
I
Small,
Multi-I
ILarge,
I
I1I
13000-Torne I 1000-TonnelAircraftIMission Air-I ASW
I Vehicles
I
IAircraftl
craft
ICarriersl
I Class
I
Class
I
I
S/L(S)I
I
S/L(v)
l SWA
I
S/L(L)
I
AVP
I WIG(H)
I
WIG(O)
I
I
I
I
HYD
SES
2
3
I
i
HYD 7
PC 1
ISES
SWA 4
I
I
WIG(S)
MONO 3
I
FAB
I
ACV
322
I
I
I
A/L(N)
CVN
ISWA CVN
I
I
I
I
SAB
A/L
3
1
I
I
I
I
I
ISES CV
ACV
LCAC
LCPC
I
MC
I
IU
Z
650
U)
0.
40
030
I.-
0
10
10
100
1,000
GROUP 1 WEIGHT (TONNES)
323
10,000
NONNUCLEAR PROPULSION
140)
1200
X
2
400
200
1.000
10 000
100.000
GROUP2 POWER(MHPI
1.0.000
NUCLEAR PROPULSION
ZXI
O
0
z
i
\K\
i
o
NOTE COWNICO
COST OLI DOES
NOT mCLuOt COST
O rLiC'OR
0*! I Tr41S
O. O".
S*S'
TO theE _uCL
U
'I
0
100
I
!ELAtEO
---
1.000
10.000
GROUP2 WEIGHT (TONNES)
324
100,000
a600
zz
0
D500
o,)
zc00
I.-
w2 0 0
0.
I-
0
u 100
1
10
GROUP 3 WEIGHT
325
100
ITONNES)
1000
I -
tCA4
7m F-
- -
--
.wI'sI
-
--
I- - - - :n
1~~~~~~~
cn
_
.
i
-
KEY
~- -
X SIMPE CONtRa
SYSTEMS ItNIMAL ELECTRONICS
2 MOOEST CONTROL SYSlftS
SOPHISTICATED ELECTRONICS
3 COMPLtE CONTROI SSTEMS
SOPHISTICATED
ELECTARONICS WIFICRITICAL
SHIPS
3
-
0
I0
z
(u,
O0
I
w
Z
z
z
0
)-
IVn)
0
U
I
100
-
£C
.....
0 1
_IIY~iS
'MY0
H
-C
LCPC
i
.
10
GROUP 4 WEIGHT(TONNES)
326
2
St S
I
- w%}
iIJEJ
S^4
.
_
100
_
r -CV b 1S CVN
-
-
-
S.. -- C..
SWACffi
,
1000
a;
zz
O
I-
0
a
w
0
z
o
w.
I-
(t
u-
1
10
100
GROUP5 WEIGHT (TONNESI
327
1,00
10.000
z
oz
u) 1
a
z4
u)
0
z
a
0I-
U
00
GROUP6 WEIGHT ITONNES)
328
KEY
1. CONVENTIONALDISPLACEMENTSHIP
2. WEIGHT-CRITICALSHIP, LIGHT ARMAMENT
Z
F
---
LCAC
|Y-
WIGISISS
CV.
-
CVN
(fl
0o
O
I
z0
WA 4
16
--
JMCM
1
L
Cpc
SWA CVN
10
100
GROUP 7 WEIGHT
329
(TONNES)
1000
10(
ca
z
co
0
f0
1
e Li
0.1
_01
1
1000
100
10
SUMMATION OF GROUPS 1 THROUGH 7 COSTS ($ MILLIONS)
330
o lO1
-J
2j
a
0g1)
.
L)
(0
C'
0
cc
0
1
o
,IAl
1
SUMMATION
10
100
1000
OF GROUPS 1 THROUGH 7 COSTS ($ MILLIONS)
331
APPENDIX
F
PRICE H INPUT GLOSSARY AND DATA SHEET
(Ref. 20)
GLOSSARY OF PRICE H VARIABLES
BASICINPUTS
AUCOST
CATGRY
CMPID
COST
CPF
DCOST
DESRPE
DESRPS
DFPRO
DLPRO
DSTART
DTCOST
ECMPLX
EREL
INTEGE
INTEGS
MCPLXE
MCPLXS
MECID
MODE
MREL
NEWEL
NEWST
PCOST
PEND
PFAD
PLTFM
PRCOST
PROSUP
PROTOS
PSTART
PTCOST
QTY
QTYNHA
RATOOL
TARCST
TCOST
Average recu.ringunit productioncost (Historical
Data).
Thru-put cost category.
Equipmentelectronicclassification.
PurchasedItem cost.
Two,three, or five digit code for the manufacturing
process used in production.
Thru-put developmentcost.
Decimal fraction of electronicdesign repetition.
Decimal fraction of structuraldesign repetition.
Month/Yearof completionof first prototype(Not to
Includefield tests).
Month/Yearof completionof development.
Month/Yearof start of development.
Total developrn-nt cost (HistoricalData).
Engineeringcomplexityfactor.
Electronic reliabilityadjustmentfactor.
Level of integrationand test requirements
applicable o electronics.
Level of integrationarn lest requirements
applicableto structuraLmechanical
areas.
Empiricalinput of the electronicsmanufacturing
complexity.
Empiricalinput of the structural'mechanical
manufacturingcomplexity.
Equipmentsmechanicalstructuralclassification.
Numericaldesignationof PRICE input mode.
Mechanicalreliabilityadjustmentfactor.
Decimalequivalentof uniquenew electronicdesign
required.
Decimalequivalentof unique new mechanical/
structural design required.
Thru-put productioncost.
MonthtYearof completionof production.
Month/Yearof productionfirst article completion
(Does not include field testing).
Equipmentsdesignedoperatingenvironment
(SpecificationLevel).
Prototypemanufacturingcost, to includespecial
tooling and test equipment(HistoricalData).
Prototypesupport empiricalfactor.
Number of prototypesto be developedfrom
DSTARTto DLPRO.
Month/Yearof release to production start of
productioncycle.
Total productioncost (HistoricalData).
Number of productionunits to be built from
PSTART to PEND.
Numberof equipmentsrequiredfor integrationto
the next higher assemblylevel (System).
Average monthlyproductionrate for which tooling
is to be costed.
Target design-to-unit-production-cost
(Amortized
over all productionelements).
Total thru-pIt cost.
333
Proportionof an equipmentsvolumeoccupiedby
the electronicspackage.
Envelopevolumeof an equipment (Cubicfeel).
Electronicpackagingdensity(Weighlof electronics
per cubicfoot).
Weight of mechanicatlstructuralelements
(Pounds).
MechancaVstructural
density (Weightof structure
per cubicfool).
Weight of equipment(Pounds).
Economicbase year.
Technologicalbase year.
USEVOL
VOL
WECF
WS
WSCF
wr
YRECON
YRTECH
GLOBALINPUTS
COSTU
DDATA
DESIGN
DMULT
DPROJ
DRAFT
DTECIM
DTLGTS
ECNE
ECNS
ESC
NFACS
NSHIFT
PDATA
PMULT
PPROJ
PSF
PTECIM
PTLGTS
RSRCE
SYSTEM
TECDEL
Units for cost measurements(e.g.dollars,
thousands of dollars,etc.).
Level of data requirementsfor the development
phase.
Empiricalfactorcontrollingthe levelof engineering
design.
Developmentcost multiplier (tor additionsto
manufacturingcost level).
Level of projectmanagementfor thedevelopment
phase.
Empiricalfactor controllingthe level of drafting.
Level of technologicalimprovementfor
developmentprograms.
Level of specialtools and lest equipmentrequired
for prototypemanufacturing.
Levelof productionengineeringchangeactivityfor
electronicssections.
Level of productionengineeringchangeactivityfor
structuralsections.
Escalationcontrolvariable.
Numberof productionfacilities (Lines)to build
QTY equipment.
Numberof productionshifts used to build OTY
equipment.
Level of data/documentation
requirementsduring
the productionphase.
Productioncost multiplier(For additionsto
manufacturingcost level).
Multiplierof productionprojectmanagementcosts.
Prototypeschedulefactor (Definesthe sequenbal
manner in which the prototypeswll be
manufactured).
Level of technologicalImprovementfor production
programs.
Multiplierof specialproductiontools and equipment
costs.
Indicatorof resourcesthat are availablefor
production.
Multiplierof systemsengineeringcosts in the
developmentphase.
Numberof years of technological delay (Lag).
ILTJE
Input Data
Worksheet
m
Filename:
Shemt_ of
Basic Modes
t:
Title:
Fdu4tm
snt'w
Genral A
GeneralB
Mechanical/
oad
v-
OaTY
PoToG
WT
VOL
MODE. HSMNT
O-mvi/lit
NHin AmMV
NM
EItrotu
FrMm
Swctwel
S cd0tmna
Lew
Yaw
tie.0l
Ya l
Tectmn
orYNNA
INTo
I TES
PLYTFM
YRECON
YRTECH
Sru gtwS
Mi
Dht
U
Rpil
S
RMa.dltr
Swlql
Iftioe"
fDqur
CovplI-t
tt
I d
NW/ISW
FestyvPft
-~
Volm
_
ItA
If g
CPL5
Swuctu
NW,.
hmufiduiMIWI
Comeilbl
E1V.sOfI
Dsq I
ROa
Ectf
RaI·LJ.ilV
NEWEL
DESIAE
EEL
W.S "V
Structurl
IVelu
NE
rC Ft/ Flrtuu
Ebctronics
Development
WECF I
USEVOLMFLXsE
OIo_.0et
Son
In lotoYp
C omlte
Oiwlopmnw
opem
C
E.iwarw
C.t
Tol
&
Tyt Eq.s
ty
etowlp
Atyr
OST^RT
OFPRO
OLPRO
ECMPLX
OTLGTS
FROSUP
hobcto
ComfeN
PRICElmprovemnt
Tooll
T
to
E5M
I,
Rau/Mont
Tre
Ftor
PIF
FTLGS
RATOOL
Fdoect.u
S1/t
F.r AItcil
Dhlr Y
Production
FSTAT
PF^D
Actual
A-la Urm
Cost Data
AUCOST
EN
PhodguiLt
D
To*t
hosr
PTCOST
PRCOST
:.o
leopmm
oTCOST
(Mode7 only)
Notes:
_
_~~~~-
.
I ELECTRONIC
ITEM
2 MECHANICAL
ITEM
S MODIFIEDITEM
7 ECIRF
-
GC 1613 1/85
o( 1985RCACorporation
-
[CE0I
334