tl.

',''^

. l

SHIP PRODI.ICTION
SECOND EDITION

BY
R I C H A R D L E E S T O R C H ,C O L I N P . H A M M O N ,
HOWARD M. BUNCH, & RICHARD C. MOORE

{}illlltilililllltillill
CORNELL

MARITIME

PRESS

CENTREVILLE, MARYLAND

,?t132 ,l

Copyright O 1988, 1995by Cornell Maritime press, Inc.

All rights reserved'No part of this book may be used or reproduced
in any manner whatsoeverwithout
written permissionexcept in the caseof brief quotations embodied
in critical articles and reviews. For
information, addresscornell Maritime pres, Inc., centreville,
Maryland 21677.

y'/vl

tf6

,5Br
I e95

Library ofCongressCataloging-in-publication
Data
Shipproduction / byRichardleeStorch
p.
cm.
-i-:.

-i,

. [etal.l. _2nded.

Includes bibliographical referencesand index.

(hc)
ISBN 0-87033-461-1
1. Shipbuilding. I. Storch, Richard Lee, 1947vM146.S84 1995
623.8'3'068-dc20

Manufactured in the United States of America

First edition, 1988.Secondedition, 1995.

C'

C O N T EN T S

LIST OF PI,ATES, FIGURES, AND TABLES

PREFACE
I.

xix

INTRODUCTION
The Shipbuilding Process,3
Shipbuilding Terms and Defrnitions,6
Ship Definitions, 6
Ship Types,6
Shipyard Facilities, 6
Organization, 7
Trades,11
ClassificationSocietiesand RegulatoryAgencies,1A
Status of the Shipbuilding Industry, 14
The World Shipbuilding Market, 15
The U.S. Shipbuilding Industry, 18
The U.S. Shipbuilding Market, 21
ResourceMarkets,2S
Industry Organization and Performance,25
TechnologyDevelopmentsand Productivity, 28
References,29

II. SHIPBUILDING MANAGEMENT THEORY

Introduction, 31
Shipbuilding EconomicTheory, 31
The Shipbuilding Model, 33
Graphical Analysis of the Shipbuilding Model, B7
Summary of the Shipbuilding Model, 43
The Shipbuilding Model: An Example, 44

2l$o

XI

31

Vi

CONTENTS

Group Technology,45
Group Technology,an Overview,46
Group TechnologyDefined,47
Classificationand Coding,49
Group Technologyand the Shipbuilding Model, 51
Work Breakdown Structures, 53
Systems-OrientedWork Breakdown Structure, 54
Product-OrientedWork Breakdown Structure (PWBS).54
References,59
ilI.

PRODUCT-ORIENTED WORK BREAKDOWN STRUCTURE

Introduction, 60
Planning for Production,61
Design and Material Definition, 62
Production, 63
Controlling,66
Costing,67
Zone Construction Method, 67
Hull Block ConstructionMethod (HBCM). 68

60

Zone Outfitting Method, 76

Zone Painting Method, 89
Pipe PieceFamily Manufacture, 93
Work Lots, 94
Material Receiving,95
Pipe PieceFabrication, 97
Pipe PieceAssembly,99
Pipe PieceJoining, 103
Testing and Coating, 103
Palletizing, 104
References,104
ry.

METAL MANUFACTIJRING AND CONSTRUCTION PROCESSES

Hull Materials, 105
Structure of Metals, 105
PhaseDiagram ofSteel, 107
Propertiesof Steel,110
Types ofShipbuilding Steel, 112
Aluminum and Aluminum Allovs. 112
Metal Processes,
116
Straightening ofPlates and Sections,116
Cutting, 117
Forming, 120
Welding, 123

105

coNTENTS

vii

Welding Processes,126
Soldering and Brazing, 134
Distortion Removal, 135
Outfit Processes,136
Machinery, 137
Piping, 138
Heating, Ventilation, Air-Conditioning (HVAC), 140
Electrical, 141
Accommodations,142
Deck Fittings,142
CombatSystems,142
Material-Handling Processes,143
Material-Handling Equipment, 143
Material-Handling System Design, 148
Rigging, 148
Staging/Scaffolding, 149
SurfacePreparation and Coating, 150
SurfacePreparation, 150
Shop Priming, 156
Coating Systems,156
Flame-SprayedCoating, 157
Quality Assurance,158
Steel ProcessQuality Assurance,158
Outfit ProcessQuality Assurance,159
Test and Trials, 159
VesselDelivery Certificates, 160
References,160
V. SHIPYARD LAYOUT
Historical Perspective,161
Pre-World War II, 162
World War II-1960,162
1960-Present, 163
Shipyard Facilities and Siting, 168
General Yard Layout, 17O
Building Positions,171
Material Handling, 173
WarehousingFacilities, 174
Production Facilities, 175
ProcessLanes,189
Physical ProcessLanes, 189
ProcessLane Loading, 193
References,194

i
I

161

Viii

coNTENTS

VI. SHIP DESIGN AND ENGINEERING

Introduction, 195
Design Cycle, 195
Information Flow, 197
GT Shipbuilding Impact on Design/Engineering,19g
Design Stages,200
Basic Design, 200
Functional Design, 203
Transition Design, 213
Work Instruction Design, 218
Mold Loft Interface, 222
Design and Engineering for Production,222
General Principles, 227
Standardization,22S
Logic of Design,229
Design Changes,240
CADICAIII,244
Hardware and Software, 245
CAD Outputs,246
CAD/CAM Potential,247
Models,248
References,2S0

195

VII. PLANNING, SCHEDULING, AND PRODUCTION COIYTROL

Introduction,252
Planning and Production Organi zations, 252
Integration with Other Shipbuilding Functions, 252
Networks,253
Activity Durations, 257
Planning Overview, 257
Shipyard Planning Overview, 257
Project Planning and SchedulingOverview,2Sg
Planning and Scheduling,27I
Hull Construction Planning, 272
Basic Planning and Scheduling,279
Detail Work Instruction Planning and Scheduling,284
Shop-LevelPlanning and Scheduling,298
Production Control, 298
Man-hour Expenditures, 298
Production Progress,298
Productivity Indices, 299
Zoneto System Transposition,302

252

CONTENTS

IX

Material Control, 302

Material Defrnition, 303
Material Classifrcationand Procurement,305
Warehousingand Palletizing, 314
References,318

VIII. ACCI]RACY CONTROL

Introduction, 319
Statistical Principles, 319
Overview,324
Planning,326
Preliminary Planning, 328
Detail Planning, 329
Standardization,332
Executing, 335
Self-Check,336
When and What to Check, 336
Information for Check Sheets,337
Evaluating, S40
Regular Analysis, 340
Urgent Analysis, 343
Control, 344
Applications, 349
Design, 349
Mold Loft, 349
Production Control, 353
Parts Fabrication, 353
Subassembly,35S
Block Assembly,355
Standardization,355
AccuracyStandards,355
Excessand Shrinkage Allowance Standards,356
Measuring,358
References.358

319

DL SHIP COI{VERSION, OVERHAUL, AND REPAIR

Introduction, 359
Life Cycle of a Ship, 359
Size and Scopeof the Industry,360
Nature of the Industry,361
Marketplace Characteristics,362
Selectingan Approach,366

359

CONTENTS

Facilities, 367
Planning Approach,368
ManagementApproach, 384
Repair and Overhaul, 386
UnscheduledVoyageRepairs, 386
Planned Maintenance,391
Overhaul, 394
Conversionand Modernization, 396
Deactivation,403
Scrapping,404
RecentInnovations in Ship Repair, 404
Maintenance ManagementApproaches,405
Delivery of Services,406
TechnologicalTrends, 406
References,408
GLOSSARY

409

INDEX

421

ABOUT THE AUTHORS

436

P L A T ES , F I G U R ES , A N D T A B L E S

Plates
All plates follow page 84.
Fig. 3-27. Outfit units. (a) An engineroom
tank top unit. (b) Pumps, valves,pipe
pieces,and foundationsincorporatedas a
unit. (c) Pneumatic tubing and supports
form a small unit. Fig. 3-28 (a & b).
Typical machinery outfit units (the same
problem area). Fig. 3-29. Machinery
outfit unit assembledon a common
foundation (courtesyAvondale Shipyard).

the building dock. Fig. 3-52.Erection of

the block shown in figures 3-49 through
3-51.

iv. Fig. 3-53.On-blockoutfitting of one-half

of a superstructuredeck. Fig. 3-54.
On-block outfitting of another superstructure deck block. Zone, problem
area, and stageare the same as in figure
3 - 5 3 . F i g . 3 - 5 5 . G r a n d - b l o c kj o i n i n g o f
the aft half of the superstructure.
Fig. 3-56. Erection, keel laying plus
1 1 w o r k d a y s .F i g . 3 - 5 7 . E r e c t i o n ,k e e l
laying plus 13 workdays.

I I . Fig. 3-35. Engine room tank top components

fitted on-block. Fig. 3-36. Outfitted
engine room blocks erectedover previously landed outfit units (courtesy
Avondale Shipyard).Fig. 3-37. Downhand outfitting on-block,in an upsidedown block. Fig. 3-38. On-blockoutfitting and painting, on the ceiling and
on the deck. Fig. 3-39. Two complex
blocks, outfitted together, upside down,
to easeerection.Fig. 3-40. On-board
outfitting performed prior to landing
the main engine.

V. Fig. 3-58.Erection, keel laying plus 15 workdays.Fig. 3-59.Erection, keel laying plus
19 workdays.Fig. 3-60.Erection, keel
laying plus 22 workdays. Fig. 3-61.
Erection, keel laying plus 22 workdays.
Fig. 3-62.Erection, keel laying plus 24
workdays. Fig. 3-63.Erection, keel laying
plus 24 workdays.

III. Fig. 3-47. On-blockoutfitting of a forecastle

deck upside down. Fig. 3-48. On-block
outfitting ofa forecastledeck right side
up. Fig. 3-49.On-blockoutfrtting of a
side shell and engine room flat upside
down. Fig. 3-50. On-blockoutfitting of a
side shell and engine room flat right side
up. Fig. 3-51.On-blockoutfitting of the
block shown in figures 3-49 and 3-50
being completedright side up next to

VI. Fig. 3-64.Erection,keel laying plus 27 workdays. Fig. 3-65.Erection,keel laying plus
28 workdays.Fig. 3-66.Erection, keel
Iaying plus 29 workdays.Fig. 3-67.
Erection,keel laying plus 29 workdays.
Fig. 3-68.Operation and test (vessel
Iaunched 43 workdays after keel laying,
deliveredsevenmonths after starting
fabrication).

XI

Xii

PLATES,FIGURES,AND TABLES

Figures
1-1.
1-2.
1-3.
l-4.
1-5.
1-6.

Basic ship dimensionsand definitions, 7

Ship types, 8
Inboard profiles of major ship types, 9
Representativeshipyard layout, 12
World shipbuilding market cycles,16
LocationsofU.S shipyards,20

2-I.

Productionplan for traditional ship

construction,33
Sample ship production progresscurve,
34
Shift in the average cost curve when
required output changes,39
Shifts in the average cost curve with
changesin labor quality, 40
Bottlenecksincreasecostsand reducethe
optimalrate ofproduction,40
Production plans beforeand after
total output adjustment, 42
Manpowerprofilesin an environmentof
changingrequirements,45
Comparisonof conventionaland group
processbatch flow, 47
Company functions affected by group
technology,50
The Boeing Company group technology
classification system, based on the E.G.
Brisch concept,52
Example of U.S. Navy hierarchical
subdivisions,55
Product work breakdown structure
components,56
Elements of the three-dimensional
PWBS matrix, 57
Impacts of time, units of resources,
and quality of work circumstanceon
productivity value, 58
Optimum PV requires balanced
influencesof T, N, and Q, 58
Iterative developmentof work
packages,S8

2-2.
2-3.
2-4.
2-5.
2-6.
2-7.
2-8.
2-9.
2-70.

2-11.
2-12.
2-13.
2-14.

2-15.
2-76.

3-7.
3-8.
3-9.
3-10.
3-11.
3-12.
3-13.
3-14.
3-15.

3-16.

3-17.
3-18.
3-19.
3-20.

3-21.
3-22.
3-23.
3-24.
3-25.
3-26.

3-1.
3-2.
3-3.
3-4.
3-5.
3-6.

Industrial projectmanagementcycle,62
System and zoneorientations in the
management cycle,62
Design process,64
Design outfit specialtygroups,64
Processlanes for simultaneoushull
constructionand outfrtting, 65
Integrated schedulesfor hull construction,
outfrtting, and painting,66

3-27.
3-28.
3-29.
3-30.
3-31.
3-32.

Hull block constr-uctionmethod

(HBCM) manufacturing levels,69
HBCM classifrcationby product
aspects,71
Parts fabrication problem area and
stageclassifications,72
Part assemblylevel classification,T2
Sub-blockassemblylevelexamples,73
Block assemblyproblem area and stage
classifrcations,75
Semi-blockand block assembly,showing
the bottom center block ofa cargohold, 76
Block assemblyand grand-blockjoining,
showing the top of a wing tank,77
Semi-blockand block assembly,showing
a bottom wing tank side shell with
hopper and the transversehopper in a
cargo hold, 77
Block assemblyand grand-blockjoining,
showing a corrugated transverse bulkhead and a deck center betweencargo
holds, 78
Block assemblyand grand-blockjoining,
showing a cant block, 78
Block assembly,showing an upper deck
and an engine room flat,79
Semiblock and block assembly,showing
a bulbous bow, 79
Semi-blockand block assembly,showing
a forecastleand the upper deck in a
fore body, 80
Grand-blockjoining, showing a forecastle
and the upper deck in a fore body, 80
Block assemblyand grand-blockjoining,
showingthe bottom of an engine room, 81
Block assembly,showingthe side shell
of an engine room, 81
Zone outfitting method (ZOFM) manufacturing levels,83
ZOFM classificationby product aspects,
85
Problem area subdivisionsfor design and
material requirements for component
procurement,86
SeePlate I.
SeePlate I.
SeePlate I.
Engine room outfit unit, 87
Ahatch coverand coamingunit, 87
Very complex piping units, for the deck
ofa product tanker,87

PLATES.FIGURES.AND TABLES

Mast units, 87
On-blockoutfrtting in a pipe tunnel
under a cargo hold, 88
3-35.
SeePlate II.
3-36. SeePlate II.
3-37. SeePlate II.
3-38. SeePlate II.
SeePlate II.
3-39.
3-40.
SeePlate II.
3-41. Zone painting method (ZPTM) manufacturing levels, 90
ZPTM classificationby product aspects,
3-42.
91
3-43. Typical paint systemsapplied in
accordancewith ZPTM, 92
A center deck and center transversebulk3-44.
head grand block, upside down, 94
3-45. A grand block, including the side shel^,
top side tank, transverse
hopper, and side transverse bulkhead
ofa cargohold, 94
3-46. The same grand block as in figure
3-45,94
.)-zt
, .
SeePlate III.
3-48. SeePlate III.
3-49. SeePlate III.
3-50. SeePlate III.
3 - 5 1 . SeePlate III.
3-52. SeePlate III.
3 - 5 3 . See Plab fV.
3-54. See Plate fV.
3-55. SeePlate IV.
3-56. SeePlate IV.
3 - 5 7 . See Plate IV.
3-58. SeePlate V.
3-59. SeePlate V.
3-60. SeePlate V.
3 - 6 1 . SeePlate V.
3-62. SeePlate V.
3-63. SeePlate V.
3-64. SeePlate VI.
3-65. SeePlate VI.
3-66. SeePlate VI.
3-67. SeePlate VI.
3-68. SeePlate VI.
3-69. Pipe piece family manufacturing method
(PPFM) manufacturing levels,95
PPFM classfficationbyproduct aspects,96
3-70.
Problem area subdivisions for pipe
3-77.
piece fabrication, assembly, and joining
levels,97
Tlpical classificationsfor PPFM, 98
3-72.
Tlpical decisionlogic for determining
3-73.
pipe piecefamilies, 99

3-33.
3-34.

3-74.
3-75.
3-76.
3-77.
3-78.
3-79.

Xiii

Typical PPFM classifications,100

Typical PPFM classifications,101
Typical PPFM classifrcationsof
penetrations,102
Pipe pieceassemblywork stations, 102
Pipe piececoating, 102
Pipe piecepallets and pallet storage
areas,103

Arrangementofatomsincubiclattices,106
Dendritic crystal growth schematic,107
Grain boundariesof dendritic
crystals, 107
4-4.
Iron/iron carbidephasediagram, 108
4-5.
Stmctural changesin0.407ocarbonsteel
during slow cooling,109
4-6.
Stress-straincurve,111
4-7.
S-N plot offatigue test results, 111
4-8.
Diagram illustrating plate being
straightened in roll-straightener, 116
4-9.
Diagram illustrating the straightening
of thin plates (sheetmetal), 116
4 - 1 0 . Heavy-duty shears,117
4-tt.
Band saw for sheet metal cutting, 117
4-r2.
Burning torch, 118
4 - 1 3 . Profile cutter, 120
Flame planer, 120
4-t4.
4-15(a).Rolls used for shell plate curving, 120
4-15(b).Rolls used for forming round components,120
4-76. Pressused for bending plate, 121
4-17. Pressbrake, 121
4-18. Gap press,121
4-19. Frame bender, 122
4-20.
Curvature from line heating, 122
4-21,. l,ocal reactionsiiom line heating, 122
4-22. Convergenttype torch tip, 123
4-23. Typical line heating operations,124
4-24. Butt weld, 125
4-25. Fillet weld, 125
4-26. Metallurgical zonesin welding, 125
4-27.
Tlpical penetrationsfor various welding
processes,125
4-28.
Singleand multi-pass welds, 125
4-29. Gas welding process,128
4-30. Oxyacetylenegas welding equipment,
t28
4-31. Electric arc welding processes,129
Manual shieldedmetal arc welding, 130
4-32.
4-33.
Manual shieldedmetal arc welding
process,130
Manual shielded metal arc welding
4-34.
positions,130
4-35. Gravity welding machine, 131

4-7.
4-2.
4-3.

XIV
4-36.

PLATES,
FIGURES,
ANDTABLES

Automatic shielded metal arc

welding, 131
4-37.
Submergedarc welding, 131
4-38(a). Submergedarc welding, semiautomatic, 132
4-38(b). Submergedarc welding, automatic
equipment, 132
4-39.
Stud welding, 132
4-40.
Gas tungsten arc welding, 133
4-41.
Gas metal arc welding, 133
4-42.
Plasma metal inert gas welding, 134
4-43. Electroslagwelding, 134
4-44.
Electrogaswelding, 134
4-45.
Classificationoffairingmethods, 135
4-46.
Distortion removal following
stiffener welding, 136
4-47. Distortion removal, 137
4-48. Shipyard machine shop, 138
4-49. Pipe shop band saw, 139
4-50.
Plasma arc pipe cutter, 139
4-5I.
Pipe bending machine with numerical
control, 140
4-52.
Palletized pipe pieces,141
4-53.
Sheetmetal duct work, 141
4-54.
Sheetmetal cutter with numerical
control, 141
4-55.
Pressbrake for sheet metal cutting and
forming, 141
4-56.
Turret press, 142
4-57.
Conveyorused at surfacepreparation
and coatingwork station, 144
4-58.
Roller and chain conveyorin a subassemblyline, 144
4-59. Conveyorin a pipe shop, 144
4-60. Top-running bridge crane, 145
4-6t.
Under-running bridge crane, 145
4-62. Magnetic clamp for crane steel
handling, 145
4-63.
Gantry crane, 145
4-64. Jib crane, 146
4-65. Hammerheadcrane, 146
4-66. Mobile crane, 146
4-67. Block and tackle, 146
4-68. Flatbed truck moving a boiler, 147
4-69.
Forklift truck, 147
4-70. Heavy-lift crawler vehicle, 147
4-7L. Pallets, 148
4-72. Riggersinstalling a rudder, 149
4-73. Staging, 149
4-74. Mill scalelayer composition,152
4-75.
Blast cleaningfacility, 153
4-76.
Hull bottom cleaner,153
4-77.
Self-propelleddeck cleaning unit, 154
4-78.
Portable pressureblaster, 154
4-79. Push mower type vacuum blaster, 154

4-80.
4-81.
4-82.
4-83.

5-1.
5-2.
5-3.
5-4.

Centrifugal flagellator, 155

New constructionpainting process,157
Painting facility, 158
Piping unit hydrostatic test before
erection,159

Productionorganizationproperties,162
First-generationshipyardlayout, 163
Second-generation
shipyard layout, 164
High-capacity,in-line processshipyard,
165
5-5.
In-line processshipyard in futl
production,165
5-6.
Self-propelledtransporter, 165
5-7.
Third-generationshipyard layout, 167
5-8.
Fourth-generationshipyard layout, 169
5-9.
Proposedfourth-generationnewbuilding
yard layout, 170
5-10. Longitudinal-slopedbuilding ways, 171
5-11. Graving dock,t72
5-12. Side-launchways, 173
5-13. Floating dry dock with removable
wing wall, 173
5-I4.
Floating dry dock with heary-lift
translation system, 173
5-15. Synchrolift, 174
5-16. Level building launch facilities, 174
5-17. Warehousefacility, 175
5-18. Flow diagram of N/C steel marking
process,176
5-19. Typical layout ofa steel curved panel
shop and panel line, 178
5-20. Automatic stiffener welding, 179
5-21(a). Large shipyard steel panel line in
operation,179
5-21(b). Larger shipyard steel panel line
schematic,179
5-22(a). Shipyardsteelpanelline schematic,179
5-22(.b).Shipyard steel panel line in operation,
180
5-23. Panelline blast and coatfacility, 180
5-24. Typical layout of a block surfacepreparation and coatingfacility, 181
5-25. Erection site painting, 181
5-26. Schematicof a semiautomatedpipe
shop, 182
5-27. Pipe storageand retrieval rack, 182
5-28. Pin jig, 183
5-29. Typical shipbuildingjigs and fixtures,
lB4_87
5-30. Adjustable mobile scaffolding,188
5-31. Erection work unit, 188
E-32. Processlane facility layout, 190
5-33. PlatenAlock categories,191
5-34. Flat block platen. 192

PLATES,
FIGURES,
ANDTABLES
6-1.
6-2.
6-3.
6-4.
6-5(a).
6-5(b).
6-5(c).
6-5(d).
6-5(e).
6-6.
6-7.
6-8.
6-9.
6-10.
6 - 11 .
6-12.
6-13.
6-14.
6-15.
6-16.
6-17.
6-18.
6-19.
6-20.
6-21.
6-22.
6-23.
6-24.
6-25.
6-26.
6-27.
6-28.
6-29.
6-30.
6-31.
6-32.

6-33.

Product-orienteddesigrrprocess,196
The pallet concept,198
Information interchangeto integrate
HBCM, ZOFM, andZPTM, 199
Intensity ofgroup technology
engineering effort, 20 1
The design and engineeringcycle,
marketing to contract, 202
The design and engineeringcycle,
contract to Go meeting, 203
The design and engineering cycle,
Go meeting to K meeting, 203
The design and engineering cycle,
K meeting to ML meeting, 204
The desigrr and engineering cycle,
ML meeting to keel laying,204
Preliminary block plan, 205
Functional desigr responsibilities,206
Structural key plan, 208
Structural key plan, 208
Structural key plan, 209
Structural key plan, 209
Diagrammatic of part of a seawater
ballast system,210
System plan for part offorecastle deck
mooring system,211
Material ordering zones,2l2
Material list by system (MLS), 212
Transition designresponsibilities,2I4
Compositedraft of an engine room, 215
Separationof types of outfrtting materials on compositearrangements,217
Compositearrangementsshowing
hull outfrt, 218
Workinstructiondesignresponsibilities,
2t9
Coordinationof design and material
defrnition,220
Stagesfor outfitting components,221
Mold loft-generated key plan, 223
Mold loft-generated yard p),an,223
Mold loft-generated work instruction
224
Mold loIt-generatedwork instructionplan,
224
Mold loft-generated material list, 225
Mold loft-generated material list for a
pallet,226
Mold loft-generated schedule,226
.Information booklet title page,227
Improveddesignapproach,228
Design and productioninformation
useful for improving design at various
stages,229
Classificationofstandards.230

6-34.
6-35.
6-36.
6-37.
6-38.
6-39.
6-40.
6-4I.
6-42.
6-43.
6-44.
6-45.
6-46.
6-47.
6-48.
6-49.
6-50.
6-51.
6-52.
6-53.
6-54.
6-55.
6-56.
6-57.
6-58.
7-t.
7-2.
7-3.
7-4.
7-5.
7-6.
7-7.
7-8.
7-9.
7-10.
7-lf.
7-I2.
7-13.
7-14.
7-15.
7-16.
7-17.

XV

Standard structural assemblydetail, 230

Standardmachineryarrangement
module,231
Standard piping layout module, 231
Standard outfit unit module, 231
Use of geographicalanangement of
material,233
Pipe passagesreservedfor multiple pipe
runs, 234
Pipe passagesreservedfor multiple pipe
runs,234
Pipe runs parallel toYYZ axes,235
Commonsupportsfor pipe, 235
Pipe supportedon commonedges,235
Componentsaligrredside by side,236
Outfrt units include pipe piecesfor
other systems,236
Small independenttanks built as outfrt
components,2ST
Block joints avoid interferencewith
components,237
Blocks are designedto be stable right
side up and upside down, 238
Shelfplatesfacilitate pipe penetration
installation, 238
Logic for overlappingdesigrrstages,239
Standard format for purchaseorder
specifrcations,241
Preventativesand countermeasuresfor
designchanges,243
An integrated CAD/CAM system,245
CAD-generatedkey plan,247
CAD-generatedworkinstruction,248
CAD-generatedschedulereview, 249
CAD/CAM future, 249
CAD/CAM network, 250
The managementcycle,253
Build strategy process,254
Build strategy processflow, 255
Network example,256
Productivity indices,258
Different levels of detail in planning, 259
Interrupted work flow, 260
Level work flow, 260
Schedulingobjectivesfor design and
material defrnition, 262
Shipbuilding master schedule,262
Designschedulingsequence,263
Desig:ndepartmentmaster schedule,264
Ship designmaster schedule,265
Ship designgroup schedule,266
Drawing issue schedule,267
Designpersonnelschedule,263
Man-hour control curve, 269

XVi

PLATES,FIGURES,ANDTABLES

7-18.
7-79.
7-20.
7-2I.
7-22.
7-23.
7-24.

Schedulinghierarchy, 271
Milestone schedule,272
Outfitting master schedule,272
Monthly schedule,273
Weeklyschedule,2T3
Planning and schedulingcycle,274
Shop organization by zonelproblem
area.lstage,276
Communication and feedbackchannels.
277
Typical work unit, 282
L-type grand blocks,284
Block defrnition consistent with
integrated HBCN, ZOFM, and ZPTM,
285
Hullconstructionproduction
planning flow, 286
Hull constructionproduction scheduling
flow,287
Man-hour budget determination system,
293
Manpower expendituresfor hull
construction, 299
Manpower expendituresfor machinery
outfitting,299
Manpower expenditures for electrical
assembly(noncable),299
Manpower expendituresfor electric
cable assembly,299
Production progress for hull
construction, 299
Typical cost centers.300
Productivityofpartsfabrication, 301
Productivity of subassemblyand block
assembly,301
Productivityoferection, 301
Productivity of machinery outfitting, 301
Productivity of electrical assembly
(noncable),301
Productivitycontrol group, 303
Relationship of material to desigrr,
procurement, and production, 304
Flow of material-related information in
desigr,306
Critical path model of material lead time,
307
Material control classifications,307
Material control identification codes,309
Structure for material cost classification,
310
Functional flow of the requisition
process,311
Typical classifrcationstandards,313
Relationshipof material lists to
design and procurement,314

7-25.
7-26.
7-27.
7-28.

7-29.
7-30.
7-31.
7-32.
7-33.
7-34.
7-35.
7-36.
7-37.
7-38.
7-39.
7-40.
7-4I.
7-42.
7-43.
7-44.
7-45.
7-46.
7-47.
7-48.
7-49.
7-50.
7-51.
7-52.

Tlpical breakdown of work to pallets, 315

Functional flow of warehousing and
palletizing processes,316
7-55(a).Pallets stored outdoors,317
7-55(b).Pallets stored outdoors,317
/-bb.
Subcontractordeliveriestreated like
pallets,317
7-57. Palletizing flow for material issue,317
7-53.
7-54.

8-1.
8-2.
8-3.
8-4.
8-5.
8-6.
8-7.
8-8.
8-9.
8-10.
8-11.
8-12.
8-13.
8-14.
8-15.
8-16.
8-17.
8-18.
8-19.
8-20.
8-21,.
8-22.
8-23.

8-24.
8-25.
8-26.
8-27.

Normal distribution of variation from

designdimensionofcut flat bars, 320
Normal distribution of longitudinal
spacing,321
x and R control charts for variation of
dimensionsof N/C burned parts,321
Normal distribution of random samples
taken from a normal population, 322
Random sample data to be entered on
an x and R control charL,322
Merged variation of fitting flat bar
betweenlongitudinals, 323
Relationshipofstandard ranges and
tolerancelimits, 325
Erectionjoint gap rework types, 325
Accuracycontrol cycle,327
Accuracycontrol planning process,328
Sourcesofvariation by stageofconstruction, 329
Stages at which accuracy control
measurementsare taken, 330
Vital points and dimensions,331
Variation merging equation
example,332-33
Relationshipsamong accuracystandards and stages of construction, 336
Accuracycontrol checksheet,338
Sight line templates for checking
curved plate accuracy, 339
Vital dimensionson a flat block which
incorporatessomecurved shell, 339
Accuracy control measurements during
erection,340
Accuracy control information flows, 341
Analysis of rework at block weldings, 343
Normal distribution of erection joint
gap,343
Shifting an erection joint gap distribution to control the type ofrework
required, 344
Hull erectionrework analysis and report,
34447
Histograms of variations of different
characteristics, 348
Regular control items, 350
Position dimensiondiagram, 351

PLATES.
FIGURES.
ANDTABLES
8-28.
8-29.
8-30.
8-31.
8-32.
8-33.

9-1.
9-2.
9-3.
9-4.
9-5.
9-6.
9-7.
9-8.
9-9.

Block overall dimensionreport to be

used to plan erections,352
Longitudinal end position report to be
used in plan erection,352
Mold loft error list and frequencychart,
354
Weld shrinkage analysis basedon work
sequences,355
Alternative assemblysequences,356
Excessdistribution must be determined
for eachalternative assemblysequence,
357
Maintenance/repairlife of a ship, 360
Breakdown of U.S. ship repair
market, 361
Nature of the industry, 361
U.S. Navy shipbuilding and repair
budgets,364
Approach selectioncriteria,367
Representativelayout for a small repair
yard, 369
Representativelayout for a larger repair/
conversionshipyard, 370
Systems-orientedplanning network for
repair,372
Stages,374

9-10.
9-11.
9-12.
9-13.
9-14.
9-15.
9-16.
9-I7.
9-18.
9-19.
9-20.
9-21.
9-22.

9-23.
9-24.
9-25.
9-26.
9-27.
9-28.
9-29.

XVII

Classificationby product, 375

Aggregateschedulefor a zone-oriented
conversion,37U77
Work packageidentification sheet,378
Compositedrawing, 379
Functional zonerepresentation,3S0
Geographiczonerepresentation,380
Variablezonerepresentation,38l
Using AIM, 383
Unscheduledvoyagerepairs, 387
Example of bottom damage, 388
Example of bottom damage,389
Planned maintenance,392
Rigging neededto remove a bull gear
from the main reduction gear during a
repair cycle,393
Overhauls,395
Conversionof an older tanker for
continueduse,397
The conversion/modernization
decision,
397
World merchant fleet age distribution,
399
Conversionand modernization,400
Initial zone breakdown, 402
Analysisofvesselsbroken up in 1992,
405

Tables
1-1. The world's leading shipbuilding nations, 17
1-2. U.S. shipyard new constructionand repair
facilities, 19
1-3. Hourly compensationfor shipyard workers
worldwide(U.S.$), 21
4-1. Requirements for ordinary-strength hull
structural steel, 113
4-2. Requirements for higher-strength hull
structural steel, 114
4-3. Properties of high-strength hull structural
steelsdevelopedfor Navy ship
construction,115
4-4. Applicability ofcutting processesto
materials, 118
4-5. Characteristicsof fusion welding processes,
127
4-6. Weaponssystemscomponentsof combatant
ships, 143

4-7. Factors presentin the marine environment

that causecorrosion,151
4-8. Relative location of metals in the electromotive series,151
5-1. Samplepercentageofcoveredarea by
production stage for a product-oriented
shipyard, 175
5-2. Recommendedcutting methods for steel
parts,177
6-1. Comparisonof intensity of engineering
effort for a large commercialvessel
constructionproject,200
7-1. Example of data required to develop a
network.256

P R EF A C E

In 1981 the leaders of the National Shipbuilding Research Program (NSRP) recognized the need for greater focus on shipbuilding education;the result was the establishment of an educationpanel (SP-g)within
the Society of Naval Architects and Marine
Engineers Ship Production Committee-the
implementation arm of the NSRP. The frrst
project authorized by the panel was a textbook on advanced conceptsof ship production. The first edition of the book was
published in 1988. In 1993 the decisionwas
made to revise the book, and to update its
contentsto be consistentwith the technology
changesthat have occurredin the years since
the first edition was released.This book is
the revised, secondedition.
Signifrcant changeshave been made in
the second,revised edition from the text found
in the original version. Chapter I, which contains the history of shipbuilding and statistics relating to production and market share,
has beenupdated to have more current information. Chapter II has been expanded to
more directly incorporateinto the text material relating to shipbuilding economictheory.
In Chapter fV there has been new material
added that relates especially to fabrication
processes.Throughout this sectionthere has
been an updating of all processinformation
to more accurately reflect the state of technology that currently exists. Antiquated information has been removedfrom Chapter V

(Shipyard Layout) and new material has been

inserted that more accurately reflects the
conditions of current ship production facilities. Chapters VI (Ship Design and Engineering), VII (Planning, Scheduling,and Production Control),andVIII (AccuracyControl)
have all beenupdated to reflect the impact of
the computeron all of theseimportant areas.
Finally, Chapter IX (Ship Conversion,Overhaul, and Repair) has been totally rewritten;
this chapter now reflectsthe experiencesofa
major consulting firm that has extensively
supported shipyards in the improvement of
their conversionand repair operations.
Importantly, the illustrations and tables
have been completely reorganized and improvedthroughout the secondedition. It is this
feature that will be the most quickly recognized in a comparisonof the two editions.
The authors of the first edition were Professor Richard Lee Storch, University of
Washington; Dr. Colin P. Hammon, consultant; and ProfessorHoward M. Bunch, University of Michigan. A fourth author, Richard
C. Moore,University of Michigan, was added
to support the preparation ofthe secondedition, especiallyin the revision of Chapter IX.
Appreciationcontinuesto be expressedto
thosewho were heavily involvedin the preparation of the frrst edition, and whoseeffort is
still recognizablein the secondedition: Mr.
Robert Baseler, Mr. William Blaiklock, Mr.
C.P. Buckley, Mr. Robert Slaughter,Dr. WilXIX

XX

PREFACE

Mr- \{aymond
\iarn Zeig\er, Mr' 'IoYrrr \)cart'
and Nlr' George
I'au\ Vickers,
i.tt""V,-fanr.
supwho
contributors
Additional
O'I(eefe.

"D e"P rocess Ana\y srs vi a Accrrr acy Contro\ "'

and"'Line lteatrn{'
zorre brrtfrtting,"
*i*-iot
*-"t" o""d extensive\ in the first edition' and'
trave been carried, over into ttre second" Ttle
logic developed in those works permeates the

portedpreparationofthe secondedition were

Mr. Douglas Peel; Mr. Robert E' Luby, Jr';
text in both editions.
and Mr. William J. Swahl; all were involved
A final expression of gratitude is exphotoin the revision of Chapter IX' New
\en{ed. to \[s. Parne\a Cotren. research assogtaD\s see\\rr \\e sets\'\ e\r\\qt!\3r e \ee\
(\lh
of Michigan Transportatr'on
crate .\Jr\i\ersitl
' lesse
supp\\ed.\; \t ga\\s S\\p\rr\\{\r'g
Research Institute. Her patience and forBtasher), Avond.ale Shipyards (Mr' Steven
b eann g in de a\in g with the au\,tror s \i as \'Tu\)
Maguire), and Newport News Shipbuilding
remarkable.
(Mr. Robert P' Leber).
Preparation of the revised textbook was
A special expression of indebtedness is
a project managed and cost-shared by Unireaffirmed to Mr. L.D. Chirillo. His manu"Outfrt Planning," "Product
versity of Michigan for the National Shipscripts,including
"Integrated
building Research Program.
Work Breakdown Structure"'
Hull Construction, Outfitting, and Painting,"

SHIP PRODUCTION

CHAPTER ]

INTRODUCTION

Pr ocess
1 . T h e S hipbuilding
Shipbuilding is an industry that produces
products (ships, offshorestructures, floating
plants, etc.) for customers (private owners,
companies,governments,etc.).In most cases,
the product is built to order and customized
to the specificrequirementsof the purchaser.
This applies even in caseswhere a similar
seriesof ships is being built. The entire processis likely to vary somewhat,dependingon
the customer involved, but it generally involvesa number of specificstages.Thesemay
be summarized as:
.
.
.
.
.
.

developmentof owner'srequirements
preliminary/concept design
contract design
bidding/contracting
detail design and planning
construction

The first stage in the shipbuilding processis the formulation of the product requirementsby the customer.For example,a shipping line may forecastthe need for a means
of transporting 250,000automobilesper year
betweenJapan and California; a state transportation agencymay need to ferry 150,000
passengersper day acrossan inland waterway over 10 routes averaging 30 trips per
route; an oil company may need to transport 10 million tons of crude oil per year
from the Caribbean to the northeast United

States; or the U.S. Navy may need a ship

capable of delivering supplies to support a
battle group anywhere in the world on short
notice. Alternatively, a shipping line owner
may forecast an increasein container trade
between the United States and the People's
Republic of China of an unspecified nature
and amount. The defrnition of the use or mission of a new ship may be narrow or broad,
but the end product should reflect the owner's
needsand intended use.
Oncean owner has identifredthe needfor
a new ship and defined operational requirements, the next stage involves preliminary
defrnition of the basic characteristicsof the
vessel. This preliminary or concept design
stage can be done internally by the owner's
staff, by a design agent hired by the owner,
or by the staffs of one or more shipyards.
Common practice in the United States (but
not overseas)has been to use a design agent
for preliminary design. A notable exception
is the U.S. Navy, which has a large internal
preliminary design section. Owners with
considerableexperiencewith particular ship
types may, in order to satisfy specificoperational requirements,approacha shipyard directly. The aim is to develop a design that
will meet the requirements while taking advantage ofthe building experienceand capability of a parbicular shipyard to minimize con-

i
SHIPPRODUCTION

struction time and cost. The end product of

this stage is a general definition ofthe ship,
including dimensions, hull form, general arrangement, powering, machinery arrangement, mission systems defrnition (such as
cargo capacity and handling equipment, combat systems, or habitability), capacities of
variable weights (such as fuel oil, water, crew,
and stores) and preliminary defrnition of major systems(such as structural, piping, electrical, machinery, and ventilation [FIVAC]).
Based on the general description of the
ship to be built, as determined by the end
product ofthe preliminary design stage,more
detailed information is required to permit
bids and/or contracts to be prepared. This
information, called the contract design,must
be of suffrcient detail to permit the preparation of cost and time-to-build estimates by
shipyards interested in the shipbuilding project. As in the preliminary design stage,this
work can be performed by the owner's staff,
by design agents, or by shipyard personnel.
Preparation of contract design packages by
the owner's staff is uncommon.
Following completion of the contract design stage, a specifrcshipyard is chosen to
build the vessel. Unless an owner has involved a shipyard in the preliminary and./or
contract design stage and thus is negotiating
a contract based on a mutually agreeable
design,competitivebidding basedon the contract design and specificationsis common
practice. Due to the high cost of a new ship,
contractsare generally very long and complex.
The most significant factors are cost, delivery
date, and performancerequirements.
After the bidding processis completeand
a contract has been signed, the frfth stage of
the shipbuilding process,detail design and
planning and scheduling, proceeds. Shipbuilding involves the purchasing of tons of
raw materials and many thousands of components, the manufacturing of thousands of
parts from the raw materials, and the assembly of these parts and components.Therefore,

complex and very detailed planning is required. Detail design and planning must answer the questions of "what, where, how,
when, and by whom." Determiningwhat parts,
assemblies,and systems are to be built and
what componentsare to be purchasedis primarily detail design. Where and.how are facility-use questions that include determination of the location within the shipyard and
constructiontools and techniquesto be used.
Considerationsof subcontractingand in-house
manufacture versus purchasing are also answeredhere. Thesequestionsare resolvedas
part of planning. When determines the sequencingof all operations,including purchasing and manufacturing, as well as need times
for information (design, planning, approvals,
etc.). This is the schedulingfunction. Finally,
by whom relatesto the utilization of the shipyard work force. Clearly, there is considerable interdependence among the answers to
these questions.The successofany shipyard
or shipbuilding project is directly related to
the answers to the questions or to the detail
design and planning process.
The frnal stage of the shipbuilding process is the actual construction of the vessel.
Ship constructioncan be consideredto occur
in four manufacturing levels. The first is
parts manufacturing, using raw materials
(such as steel plate and sections,pipe, sheet
metal, and cable) to manufacture individual
parts. The purchasing and handling of componentscan be consideredto be a part ofthis
lowest manufacturing level. The next manufacturing level involves the joining of parts
and./orcomponentsto form subassembliesor
units. Thesesmall collectionsofjoined parts
are then combined in the third manufacturing level to form hull blocks. Hull blocks are
commonly the largest sectionsof ships built
away from the final building site. Erection,
the final manufacturing level, involves the
landing and joining of blocks at the building
site (such as launching ways, graving dock,
or dry dock).The actual construction phase of

INTRODUCTION

shipbuilding is primarily involved with assembly, whether of parts, subassemblies,or

blocks,to form a completedvessel.An important part ofthe construction phase is verifrcation that the ship complies with the contractual requirements. Consequently, the
vessel is subjected to a series of tests and
trials prior to delivery to the owner.
Shipbuilding can therefore be viewed as
a processthat begins when an owner perceivesa need for a vesselto perform someset
of functions, that proceedsthrough a number
of stages of paperwork (design, contracting,
planning, etc.),and that culminatesin a massive collection and joining of parts and components to manufacture the desired vessel.
Productive shipbuilding is highly dependent
on careful consideration, control, and performance in each of these stages.
The shipbuilding industry is centuries old,
paralleling the history of man. Shipbuilding
techniques have changed in response to
changesin vesseldesign,materials, markets,
and construction methods. The organization
of shipbuilding companieshas also changed to
match this progression.
Throughout its early history, shipbuilding, like most early industries, was craft oriented. As such, it was almost exclusively
dependenton the skills of the craftsmen doing the work. Little planning was performed
prior to beginning the construction. As owners became more specific in defining the
desired characteristics of a new ship, shipbuilders were required to do more planning.
Nevertheless, prior to the use of iron and
steel for ships, little more than a scalemodel
or a simple drawing of a proposedship was
used to guide construction.
As industrial processesbecame more
complexand efficient, shipbuilderskept pace
with changing technology. Shipbuilding began to be subdividedinto specialties,such as
hull construction, machinery, outfitting, and
painting. More recently, the developmentof
mass production techniques and welding

both had profound impacts on shipbuilding.

As late as the 1960sand 1970s,shipbuilders
continued to try to employ mass production
or assemblyline approaches.Sincethen, a different approach to shipbuilding has emerged
and has proven to be better suited to the
economicand technical condition of the industry. This approachis basedon the application of group technolory to shipbuilding.
(Group technologyis defrned in Chapter II,
Section3.2.)
The goal ofthis textbook is to describethe
principles and practicesof shipbuilding employing group technology. As in any industry
that has existed for many years, the system
used is a mix of old and new techniques. The
system described in this book is drawn from
many sources, combining pieces of current
practice from many places. Undoubtedly, no
shipyard anywhere in the world operatesprecisely as is suggestedherein. What is presented includes many parts that in combination producea systemaimed at optimizing
productivity in shipbuilding under current
economicand technologicalconditions. It is
presented in a way that will provide the
reader an opportunity to gain some knowledge of all aspects of the industry as it is
currently evolving.
The book is organizedto present a general overview and basic principles, followed
by in-depth descriptionsofindividual pieces
of the system. The remainder of Chapter I
presentsa seriesof basicdefinitions. A much
more comprehensiveglossaryof terms is presentednear the end of the book. Additionally
a briefreview ofthe status ofthe shipbuilding
industry is presented.Chapter II deals with
two major topics: first, an overview of the
economictheory of shipbuilding and, second,
the general principles of group technology.
Chapter III is a detailed discussion of the
application ofgroup technologyto shipbuilding, calledproduct-orientedwork breakdown
structure. Chapter IV is a descriptionofthe
metal manufacturing and construction pro-

I
SHIPPRODUCTION

cessesemployed. Chapter V discussesthe

physical plant of a shipyard, indicating how
the processesare combined and organized
using group technology.Ship design and engineering,from the shipbuilder'sperspective,
are discussedin Chapter VI. Chapter VII
then presentsthe planning, scheduling,and
productioncontrol functions,including material control. Since these are of major importance to the overall productivity ofthe ship-

building system,this chapterprovidesa summation of the preceding material and is a

focal point ofthe text. Chapter VIII describes
the final critical pieceof the system,accuracy
control. Finally, Chapter IX discussesthe
importance of the ship conversion,overhaul,
and repair aspectsofthe industry, and indicates the applicability of the system, as presented for new construction, to work on
existing vessels.

2. Shipbuilding
Termsand Definitions
Shipbuilding is the constructionofships, and
a shipyard is the placewhere ships are built.
Shipbuilding is a construction industry
which uses a wide variety of manufactured
componentsin addition to basic construction
materials. The processthereforehas many of
the characteristicsof both construction and
manufacturing. Shipbuilding requires many
workers having various skills (or trades;,
working within an established organizational structure at a specificlocationin which
necessaryfacilities are available.The goal of
a privately owned shipbuilding companyis to
earn a profit by building ships.
2.1.ShipDefinitions
A ship, although a complex combination of
things, can be most easily classified by its
basic dimensions, its weight (displacement)
and./orload-carrying capacity (deadweight),
and its intended service. Figure 1-1 defrnes
a number of basic ship dimensionsas well as
typical shipboardregions.Somespecificdefrnitions are dependent on the vesseltype or
selvice, but in general, most definitions are
applicablefor all ship types.
2.2.ShipTypes
Ships can be subdivided into a number of
classesbasedon their intended service. The
primary classesare (1) dry cargo ships, (2)

tankers, (3) bulk carriers, (4) passenger

ships, (5) frshing vessels,(6) industrial vessels,(7) combatantvessels,and (8) others(see
Figure 1-2).No simple classificationof ships
by types is likely to be all-inclusive,but this
general breakdown is sufficient to indicate
general trends. Figure 1-3 showstypical inboard profiles of ships within each of these
classes.[1]
2.3. ShipyardFacilities
A shipyard generally contains several specific facilities laid out to facilitate the flow of
material and assemblies.There is no typical
shipyard layout, partly becausemany shipyards were initially constructedin the nineteenth or early twentieth century. These
yards have grown according to the availability of land and waterfront as well as in response to production requirements. Figure
1-4is a representativelayout. Typical important features are listed below:
r a location on land for erecting a ship,
along with an associatedmeans for
getting the ship to the water, such as
a graving dock,launching ways, or a
floating dry dock
r piers for storing ships afloat to permit
work to continue following launching
. shops for performing various kinds of
work, such as

INTRODUCTION

SUPERSTRUCTURE
RUDDER

MAIN DECK

FORECASTLE

. PROPELLER
LENGTHBETWE
LENGTHOVERALL

Fig. 1-1.Basicshipdimensions
anddefinitions.

-steel marking, cutting, and forming

shop
-steel assemblyshop
-surface preparation and coating
shop
-pipe shop
-sheet metal shop
-machine shop
-electrical shop
. storage,marshaling, and outdoor (blue
sky) work areas
. offrcesand personnelsupport buildings
(cafeteria,sick bay, etc.).

Associated with each of these general types

of facilities are specific pieces of equipment
that are related to the work carried out in
that location. This subject will be discussed
in detail in Chapters IV and V.
2.4. Organization
Shipyard workers are organized within departments or sections that are responsible for
some aspect of the operation of the company.
Although each company is likely to have some
variation in its organization, the usual subdivision is into six functions. These are (1)

SHIPPRODUCTION
BULK
CARRIERS

OIL
TANKERS

PASSENGER
SHIPS

COAL
SHIPS

COASTAL
FERRIES

INDUSTRIAL
VESSELS

TRAWtERS,
SEINERS,
ETC,

FLOATING
DRY

CABTE
LAYERS

00cKs

II

ROLL.ON
ROLL.OFF
(RO/RO)
t

CARGO
LINERS

LIQUEFIED
GAS
CARRTERS
l

I
ORE
SHIPS

CRUISE
SHTPS

HARBOR
FERRIES

COMBATANT
SHIPS

OTHERS

AIRCRAFT HYDROFOIL
CARBIERS
CRAFT

II

FACTORY
tcEFLOATING
SURFACE
SURFACE
BREAKERS CRANES COMBATANTS E T T E U I D
SHIPS
(PROCESSORS)
SHIPS

(SES)

CHEMICAL OIUBULI(/ORE
CARRIERS (080)sHrPS

BARGE
CARRIERS

DREDGES
FLOATING
WORK
PLATFORMS

SALVAGE/BUOY
VESSELS

AUXILIARIES

SUBMERSIBLES

SUBI\,IARINES

II
I
REFRIGERATED
CABGOSHIPS

LUMBER
CARRIERS

I
TENDERS

PILOT
CRAFI

Fig. 1-2.Shiptypes.
administration, (2) production,(3)engineering, (4) purchasing, (5) quality assurance,
and (6) projectmanagement(contractadministration).
Administration includes the chief executive officer and stafl payroll, accounting,personnel, labor relations, safety, and job estimating.
Productionis the department responsible
for the actual construction.Consequently,all
trades workers are in the production department. Planning, scheduling,and production
control functions may also be in the production department. Generally,75 to 85 percent of the shipyard's employeesare in this
department.
Engineering is responsiblefor the preparation of information about the construction
project to be used by productionin constructing the vessel.Engineering functions include
preliminary design,detail design,production
engineering,and, sometimes,bidding of new

jobs. Many shipyardssubcontractsomeofthe

designworkto outsidedesignagents.Production engineering,which is playing an increasingly important role in shipbuilding, may in
some casesoccur in the production department, and it includesplanning. The distribution of responsibilitiesbetweenplanning and
production engineering varies widely, depending on the specifrcsof the shipyard organization.
The purchasing department is responsible for obtaining the materials to be used to
build the ship. Theseinclude raw materials,
subcontracted work, and components, and
may also incorporate tools, transportation,
manufacturing, and safety equipment.
Quality assurance is often a separate
function within a shipyard that is responsible
for documenting for the vesselowner, regulatory agency, and classification society that
the vessel being constructed satisfies applicable rules, regulations, and contractual re-

INTRODUCTION

C O N T A I N E RS H I P

H O L DN O . 4

HOLDNO.2

ROLL.ON / ROLL- OFF

q.-

s-

\rJL

MACHY
SPACE

TANKNO.5

T A N KN O . 4

T A N KN O . 3

TANK NO. 2

,l

TANKNO.1 D . T .
NO.

F
TANKER

z
t ruo.o
| ruo
l,!L| i HoLo
i frflL! f,!:?i L3:3i uor-o
BALLAST

BULKCARRIER

Fig. 1-3.Inboardprofrlesof majorshiptypes.

s^Y.lt

10

SHIPPRODUCTION

HARBOR
FERRY

FISHING
VESSEL

DIESELOIL SIORAGETANK

Fig. 1-3 (continued).

INTRODUCTION

11

MACHYSPACE

coMBATANT(SURFACE)

AIR-CONDITIONING
BLOWER/ HEATER
DIESELROOM
ELECTRICAL
EQUIPMENT
ROOM

t I
-r -*f F 1**rf

SPACE
EOUIPMENT
ELECTRICAL

HYDROFOIL
Fig. 1-3 (continued).

quirements. A separateprojectmanagement
or contract management department is
charged with overseeingthe progress of a
particular shipbuilding project. It monitors
compliancewith budgets,schedules,materials usage, and the overall progress of the
shipbuilding program. Within this department, it is commonto have shipbuilding superintendents who are responsiblefor each
construction project. This department, too,
has generally been the lead group in the
implementation of statistical processcontrol
conceptsinto the yard's activities.

2.5.Trades
Within the production department, actual ship
construction work is subdivided into various
trade skills.[2] Theseare:
. air-conditioningequipment mechanic:
installs, maintains, and repairs refrigeration and air-conditioning equipment
. blaster: preparessurfacesfor coating
by abrasiveblasting (may be combined
with painter)
. boilermaker: assemblesboilers, tanks,
and pressurevessels,using power tools

12

SHIPPRODUCTION

PARKING LOT

BOUNDARY

o
tr

{r
Y
F
a

a
U
J

(r
U
z
U

I
L

F
a

o
(r

ENGINEERING/ MOLD LOFT

ADMINISTRATION
O F F I C EB U I L D I N G

o
F
a
U
L

I METAL I
I SHOP I
UNIT
ASSEMBL

6",/ **

K9

:*

FLAT BLOCK
ASSEMBLY BAY

O U T F I /TT E S T& T R I A LP I E B

Fig. 1-4.Representative
shipyardlayout.
and hand tools (may also refer to all of
the steelworking trades)
. carpbnter:fabricates and assembles
woodenstmctures, gratings, keel blocks,
and shorings,and is often responsible
for launching (may also perform joinery
work)
. chipper/gnnder:grinds and chips weld
splatter, high spots,burrs, weld slag,
and rust from metal surfacesof ships
to improve their appearanceor prepare
them for painting
o electrician (inside):installs and maintains wiring, fixtures, and equipment
for shipyard facilities
o electrican (outside):installs and repairs
wiring, fixbures, and equipment for all
electrical services aboard ship
r electronicsmechanic:works on various
types of electronicequipment to put it
in repaired operating condition
r electroplater:sets up, operates,and
maintains metal plating baths to deposit

metallic plating for protective purposes

or decorativepurposes,and to build up
worn surfaces
r insulator: installs insulation in designated areas and on piping aboard ship
o joiner: installs finished panels,floor
grouting, and tiling in shipboard living
quarters
. laborer: performs a variety of shipyard
tasks, such as carrying, digging,janitorial duties, etc.
r loftsman: lays out lines of a ship to full
scaleon the mold loft floor and constructs templates and molds to be used
as patterns and guides for layout and
fabrication ofvarious structural parts of
ships (computer-assistedlofting, which
has replacedfull-scalelayout, may be
doneby loftsmen or by engineers)
o machinist (inside):sets up and operates
machine tools as well as fits or assembles
parts to make or repair metal parbs,tools,
or machines

-_-__--_-

INTRODUCTION

o machinist (outside): installs ship machinery such as propulsion machinery,

auxiliary motors, pumps, ventilating
equipment, and steering gear
r molder: heats and pours molten metal
into hollow forms to produce all manner
of metal parts (may also be called
foundryman)
. ordnanceequipment mechanic:repairs
machinery and mechanical equipment
such as cranes,pumps, motors, and
conveyorsystemsassociatedwith
weaponssystems
o painter: mixes and applies paint or
other coating materials for protective
and decorativepurposesby means
of spray gun, brush, roller, or immersion (may also perform surface preparation)
o patternmaker: plans, lays out, and performs machine operations and benchwork to construct, alter, and repair
three-dimensional wood patterns and
core boxes for use in making molds for
foundry castings
o pipefrtter: fabricates,lays out, installs,
and maintains ship's piping systems
such as stam heating, water, hydraulic,
air pressure,and lubrication systems,
using hand tools and shop machines
(may also perform pipe welding)
o pipe welder: installs, repairs, and maintains shipboard piping systemsby means
of applicablewelding processes
o rigger/crane operator: installs and
repairs riggrng and weight-handling
gear, attaches hoists and handling gear
to rigging, and operatescranes and other
mobile material-handling equipment
to lift, move, and position machinery,
equipment, structural parts, and other
heavy loads aboard ship (may also be
called operating engineer)
. sheet metal mechanic:fabricates,
assembles,installs, and repairs sheet
metal

13

o shipfitter: Iays out and fits up metal

structural parts (such as plates, bulkheads,and frames) and maintains them
in position for welding
. shipwright: constructs or repairs wooden
ships or ship sections,sets and adjusts
supports upon which a ship is built or
docked,and aligns blocks that are to
be joined to form a ship (may be combined with carpenter)
e tacker/burner: cuts steel plate by
burning, washeswelds, and applies
temporary welds to position metals
for final welding (may also be called
boilermaker)
o welder: makes or repairs structures or
parts, using gas or electric welding
equipment, soldering equipment, gas
or electric cutting equipment, etc.
(may also be called boilermaker)
Work practi@s,work responsibilities, and trade
divisions vary between shipyards. However,
all actual construction work within a shipyard
falls into the domain of trades such as those
Iisted above.
2.6. ClassificationSocietiesand Regulatory
Agencies
The principal maritime nations have established classificationsocietiesin order to improve vessel safety. Although the original
purpose was to ascertain the relative risk of
insuring specificvessels,classifrcationsocieties now function more as safety regulatory
bodies,and are often closelycoordinatedwith
a government regulatory agency. Classifrcation societies generally publish rules for the
design,construction,and maintenanceofvessels.The U.S. societyis the American Bureau
of Shipping (ABS). Classifrcationsocietiesof
other major shipping nations are:
. England-Lloyd's Register of Shipping
. France-Bureau Veritas
. Germany-Germanischer Lloyd

14

SHIPPRODUCTION

. Italy-Registro Italiano Navale

r Japan-Nippon Kaigi Ngokai
. Norway-Det Norske Veritas
o USSR-Register of Shipping of the USSR
Classifrcation simplifies the structuring and
determination of insurance rates. It also facilitates documentation and safety regulation
by governments.An operatoris not required to
have a ship classifred.If it is not, however,
certain government agencieswill have to be
satisfred that the ship meets safety criteria
before it can be documented and a safety
construction certificate issued.An ABS representative will generally be on site for major
commercial shipbuilding or conversionprograms. Their purpose is to ensure that the
vesselis built in accordancewith the classifications for which the owner has applied.
Items within their cognizanceinclude the
structural strength ofthe ship, the installation of certain equipment, such as anchors
and hawsers, and reliability of machinery.
(Mention should be made of ABS's approval
of plans being required prior to production
commencing.Proper understanding of ABS's
plan approval requirements can help reduce
this sometimeslengthy process.)
Ships are classified according to their
intended uses, such as oil tanker, liquefied
natural gas carrier, ore carrier, trawler, tug,
etc. They are also classifiedaccordingto operating environment. Special classesmay
apply to seagoing ships operating on restricted routes or for specialpurposes,and a

specialclassappliesto ships, such as ferries,

which operateonly in shelteredwaters (harbors, rivers, etc.). Specialclassesmay be assigned because of unique design features.
Oceangoingvesselsmay also be classified
according to the level ofice strengthening.
In order to maintain the initial classification, steel ships are usually surveyed annually. The condition of hull closure appliances,
freeboardmarks,and auxiliary steering gear is
inspected.This is sometimescombinedwith
the required docking survey. The ship is inspectedin dry dock for corrosion,chafrng,or
distortion of the shell plating, stern frame, or
rudder. Specialsurveysare alsoconductedat
periodic intervals. This inspection varies with
the age ofthe ship. In addition to regular periodic surveys,the societiesrequire that damage repairs be accomplishedaccordingto society rules.
In addition to classification society inspection,government agenciesinspect merchant vessels to assure that they are designed, built, and maintained according to
published rules. Efforts have been made to
avoid overlapsbetween classificationsociety
and regulatory body inspections.Primary areas of regulatory body safety involvement
include stability, fire protection, machinery
and electrical system safety, hazardous and
polluting substancecontrol, lifesaving equipment, and navigation and communication
equipment. In the United States, merchant
vesselsafety inspectionis carried out by the
U.S. CoastGuard.

3. Statusof theShipbuilding
Industry
I

Perhapsin no other industry (with the exception of shipping, to which shipbuilding is

closelytied) are there as many diversepolitical factors that affect industrial organization
and economicperformance. The U.S. shipbuilding industry is now essentially a closed
system,dependenton a single buyer, the federal government,for its existence.Operating

through the Navy and Maritime Administration (MarAd), the federal government
has provided a market through direct purchase and differential subsidies. The government also provides a market indirectly
through tax incentives, financing, and cabotage and cargo preference legislation. (Cabotage is the reservation of all domestic and

]
l

t
I

INTRODUCTION

coastaltrade to operatorsof U.S. built ships.)

Political rather than economicconsiderations
therefore provide the dominant stimuli that
shape the structure and performance of the
industry.
Industries such as construction are naturally isolated from direct world competition
by the cost of transport, the nature of the
product, and the production process.Some
other manufacturers are naturally isolated
from foreign competition by persistent buying habits and a high degreeofproduct differentiation, as U.S. auto makers were prior
to the 1970s.In such a climate, an industry
may remain competitive even when production costsare high.
The market for ships is a worldwide market. Ship operatorsand builders freely cross
national borders to obtain a price or technological advantage. In the absenceof external
factors,the shipbuilding industries of nations
with a comparative advantage,based on resourcemarkets, cheaplabor, or superiortechnology, would be expected to prosper. The
economiesofnations thatlack suchan advantage would be expectedto allocate resources
to other activities. This simple economicmodel
does not frt the realities of world shipbuilding. This is true primarily becausemost trading nations perceivethe shipbuilding industrial base to be a vital national resource<-rn
other than strictly economic grounds. The
national policy of most trading nations reflects a primary requirement to maintain some
degreeofcontrol over the resourcesneededto
move their products to foreign markets. They
have therefore instituted policiesaimed at
assuring the existence of a nationally controlled fleet. In addition, someof these nations
have supported their shipbuilding industries in order to provide the means of replacing the national fleet should it be threatened militarily or economically.Other nations
have supported a shipbuilding industry as a
"hub" industry for economic
development.
Many coastal nations also view a national

15

shipbuilding industry as essential for defensepurposes.

3.1.The WorldShipbuilding
Market
Cycle swing is an important characteristicof
the product market and one that has a predominant effect on industry structure. This
is partly inducedby national policy. The volume of U.S. Navy new constructionand overhauls is determined by the political process,
basedon perceivednational defenserequirements. The market for merchant ships is
similarly affectedby national policy.
The market for U.S. ships operating in
the U.S. intracoastal trade is affectedby U.S.
economicconditionsgenerally.However,these
effectsare superimposedon the world market
demand for ships. The U.S. economytends
to be generally synchronouswith the world
economy,and requirements for construction
likewise follow the world economy and the
world market for shipping.
The international market for merchant
ships has historically beencyclical,with nine
major cycles between 1890 and 1992 (see
Figure 1-5).With the closingof the SuezCanal
in 1959 and the oil embargo in 1973, the
range ofthe peaksand troughs reacheda new
extreme. A cycle is also apparent in naval
ship construction,involving delivery and decommissioningdecisions.The first half of the
1970swere boom years for the world's shipbuilders.Between1970and 1975,the world
fleet deadweight tonnage increased by 70
percent (gtosstonnage of 58 percent). Much
of this tonnagecomprisedsupertankersbuilt
to carry crude oil from the Middle East to
Europe and Japan. In 1973,the world order
book was equivalent in deadweight tonnage
to 50 percent of the world fleet.[3] Even U.S.
shipbuilders,who have had a negligibleshare
of the world market since 1959,receivedorders for 47 merchant ships in 1972,up from
13 in 1970.
After the Arab oil embargo in 1973, the
entire world shipbuilding market contracted

------l

16

SHIPPRODUCTION

M GRT
(TO A LOG SCALE)

LAUNCHES
1975

35.0

14.0

1893 1903

1913 1923 1933

1943

1953

1963

1973

1983

1993

Fig. 1-5.Worldshipbuildingmarketcycles.
dramatically. The worldwide order book declined from a peak of 242.3 million deadweight tons (dwt) in Ig74 to 32.0 million in
the first quarter of 1979, and some tankers
were laid up as they were completed.Many
of the marginal shipbuilders who were able
to penetrate the market during the boom
years went bankrupt or were nationalized.
In particular, highly automated shipyardsin
Europe were either nationalized or restructured. Even Japan suffered nrany bankruptcies and a reduction in shipbuilding capacity
of 35 percent.
Although some recovery occurred in
the world shipbuilding market, its condition
throughout the 1980scan be describedas depressed.At the end ofthe decade,the industry was beginning to show signs of recovery,
with modest increasesoccurring in the early
1990s.

The Association of Western European

Shipbuilders estimatedl4] that presently installed yard capacityfor shipbuilding, worldwide, is about 18 million compensatedgross
tons. This translates into about 27 million
dwt of new tonnage per year. New building
requirementsfor the seven-yearperiod 19932000 have beenprojectedto be in the order of
330 million dwt, or about 47 million dwt per
year, average.[5]The obvious conclusion is
that during the remainder of the twentieth
century, there will be pressure to expand
world shipbuilding capacity.
As seenin Table 1-1 the bulk of the shipbuilding activities in recentyears occurredin
only a few countries. Twelve nations control
over 90 percent of the market; the Far East
yards have the lion's share, with Japan and
South Korea in the lead. The compositionof
the leading nations has remained essentially

INTRODUCTION

17

nations
Table1-1. The world'sleadingshipbuilding
(Rankingby percentage
sharein tonnageon order)
1992

1988

1985
Japan

43.3"/"

SouthKorea

32.57"

Japan

39.0%

SouthKorea

19.OY"

Jaoan

25.8"h

SouthKorea

21j%

Brazil

6.1"/o

Yugoslavia

5.0%

Denmark

5.4%

Taiwan

35%

Taiwan

3.9"/"

China

5.1"/o

China

3.47o

Brazil

3.5Y"

Taiwan

3.3%

Germany

2J%

China

3.3%

Romania

2.97"

Poland

2.0%

Germany

3.2"/"

Brazil

2.77o

sqg'n

2.9%

Poland

3.1%

Germany

2.4%

Yugoslavia

1.9%

Spain

3.1%

GreatBritain

2A%

Romania

1.7"/"

ltaly

2.7%

Poland

23%

Denmark

1.7%

Romania

23%

Spain

2.3%

u.s.A.

1.570 Denmark

1.97"

Croatia

2.17"

9.77"

AllOthers

9.0%

AflOthers

11.8"/"

AllOthers

Totaltonnageon order(Year-endrecords)
1985: 46,696,555dwt

1988: 38,536,093dwt

1992: 68,672,316dwt

"Newbuilding
Various
issues.
London.
Fairplay.
Supplementl'
Source:
the samesince1980.Japanbecamethe world
leader in commercial shipbuilding in 1956
and has continued to maintain that position.
South Korea, Spain, Taiwan, and Brazil
emergedas strongly contendingshipbuilding
nations in the 1970s.In the early 1980s,the
People'sRepublic of China also began a serious entry into the worldwide shipbuilding
market.
To maintain their market share, the
nations have followed a variety of marketing strategies, which have included subsidy
schemesand major initiatives to improve productivity. Since 1975 the work forces have
beenreducedover 60 percentin West Europe,
and over 70 percentin Japanesevards. These

reductions have been possibleby increasing

the amount of automation,standardizationof
designs,and rationalization of design, engineering, and production activities. During the
1990s,and possiblyinto the twenty-frrst century, it is expectedthat there will be changes
in individual country shares.While the Far
East yards are expected to maintain their
dominance,it is expectedthat North European shipyards will reverse their long-term
market share decline. There is also recent
evidencethat the East European yards will
establish a market position,especiallyin the
building of ships of relatively simple design.
It is expectedthat China will also increase
position, and the United States is putting

18

SHIPPRODUCTION

into place a major government-supporbed

initiative to reestablishitself in the world commercial market.
3.2 The U.S.Shipbuilding
Industry
As a major trading nation and world power,
the United States has had a policy of maintaining a shipbuilding industrial base capable of expansionin time of war. This policy
and the meansof meeting national objectives
have kept the industry alive, but not necessarily healthy. Although U.S. shipbuilders
have been preeminent in the production of
combatants,they have not been competitive
in the world merchant ship market sincethe
SecondWorld War. Government-induceddemand for merchant and navy ships has been
highly cyclical,both from the viewpoint of the
industry and individual shipbuilders. This
has resulted in risk-aversiontechniquessuch
as labor intensiveness.The labor force can
be reducedrapidly, whereas capital, oncein
place, represents a nearly constant cost regardless of the degree of utilization. Labor
intensivenessis therefore efficient from the
standpoint ofthe individual producer,but not
necessarilyefficient in terms of total national
utilization of resources.
The shipbuilding industry is part of a
political-economicsystem and responds to
other elements within that system. Of particular importanceto the objectiveof this text
is the relation oftechnology to the other elements of the system.Past national policy has
been directed at isolating the industry from
the world market. This policy has, in many
cases,resulted in a broad-basedbut low-technologyand high-costindustry. In an effort to
change this, several U.S. shipbuilders have
embarked on a courseof increasing productivity through improvements in management,
using basically state-of-the-arttechnology.The
successof this endeavor hinges in part on an
understanding of the interactions among market, political, and technologyfactors.

The U.S. active shipbuilding base is defined as privately owned shipyards that are
engagedin actively seekingconstructioncontracts for naval and commercial ships over
1,000tons. As of December81, 1998,there
were 39 shipyards (public and private) in the
United Statescapableofbuilding or drydocking a ship greaterthan 220 meters,and which
had an active employment of more than 200
persons.The list is shown in Table 1-2. As
seen,there are 14 new constructionyards, 17
repair yards with dry dock facilities, and
eight navy yards. Figure 1-6 displays the
location ofthe yards.
Beyondthe 39 shipyardsshownin Figure
1-6 there are about 560 additional establishments in the U.S. Census'sStandard Industrial Classification (SIC 3731 lshipbuilding
and RepairingJ).According to the Department of Labor, aggregateemploymentin the
private shipbuilding and repair sectorin 1992
was 123,000persons.[6]Additionally,the public shipyards also employed about b0,00060,000,making the total U.S. shipyard employment figure slightly over 170,000-the
largest number of shipyard workers in any
nation outside China, and possiblyRussia.
Approximately 80 percent of the total
labor forceof the primary active shipbuilding
baseconsistsof productionworkers. Employment levels have varied considerably over
time for all but a few of these yards. Typically, about 80-90 percent ofthe total active
shipbuilding baseproductionworkers are engaged in naly new construction, overhaul,
repair, or conversion.
In U.S. shipyards, between Bb and 60
percent of the direct costs associatedwith
building a ship are attributable to labor. In
foreign yards the labor percentageis significantly less: 20-35 percent.[Z]The major differenceis related to the types of ships being
built: the U.S. yards build naval ships primarily, and foreign yards concentrateon commercial ships. Also, someof the differenceis

INTRODUCTION

19

Table1-2. U.S.shipyardnewconstruction
andrepairfacilities
(Overalllength>220mand more than 200 employees,October1993)

YARDNAME

LOCATION

AK-WAInc
AlabamaShipyard
AtlanticDrydockCorp
AllanticMarine
Avondalelndustries
Bath lron WorksCoro
BathlronWorksCorp
BenderShipbuilding
& Repair
Bethlehem
SteelCorp
BethshipSabineYard
NavalShipyard
Charleston
Colonna'sShipyard
DetyensShipyard
GeneralDynamics
HalterMarine
IngallsShipbuilding
Intermarine,
USA
LongBeachNavalShipyard
MarelslandNavalShipyard
MaritimeContractors
McDermott,
Inc.
MelroMachineCorp
Nat'lSteel& Shipbuilding
Co.
NewparkShipbuilding
& Bepair
NewportNewsShipbuilding
NorfolkNavalShipyard
NORSHIPCO
NorthFloridaShipyards
PearlHarborNavalShipyard
PetersonBuilders
Philadelphia
NavalShipyard
PortlandShipRepairYard
Portsmouth
NavalShipyard
PugetSoundNavalShipyard
San Francisco
Drydock
SouthwestMarine
TampaShipyards
ToddPacificShipyards
TrinityMarineGroup

Tacoma,WA
Mobile,AL
FL
Jacksonville,
Mobile,AL
NewOrleans,LA
Bath,ME
Portland,ME
Mobile,AL
SparrowsPoint,MD
PortArthur,TX
Charleston,
SC
Norfolk,VA
Mt. Pleasant,SC
Groton,CT
MossPoint,MS
Pascagoula,
MS
Savannah,GA
LongBeach,CA
Marelsland,CA
Bellingham,
WA
MorganCity,LA
Norfolk,VA
San Diego,CA
Houston,
TX
NewportNews,VA
Portsmouth,
VA
Norfolk,VA
Jacksonville,
FL
PearlHarbor,Hl
SturgeonBay,Wl
Philadelphia,
PA
Portland,
OR
Kittery,ME
Bremerton,
WA
San Francisco,
CA
San Diego,CA
Tampa,FL
Seattle,WA
Beaumont,TX
TOTALCOUNT

Adaptedfrom

"Report

REPAIR
YARDS
NEW
WITHDRYDOCK
CONSTRUCTION
FACILITIES

U.S.NAVY
YARDS

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
14

17

on Surveyof U.S. Shipbuildingand RepairFacilities,1993j' MaritimeAdministration

related to productive effrciency.And cost of

materials has an impact, as doesthe national
wage rate.

In the 1940s, U.S. shipyards had the

highest labor rates in the world. Fifty years
later, however, the U.S. yards have signifi-

20

SHIPPRODUCTION

12,22,26,27

*.ro P

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

AK-WA Inc.
AlabamaShipyard
AtlanticDrydockCorp.
AtlanticMarine
AvondaleIndustries
Bath lron WorksCorp.
Bath lron WorksCorp.
BenderShipbuilding
& Repair
BethlehemSteelCorp.
BethshipSabineYard
CharlestonNavalShipyard
Colonna'sShipyard
DeytensShipyard
GeneralDynamics
HalterMarine
IngallsShipbuilding
Intermarine,USA
Long BeachNavalShipyard
Mare lslandNavalShipyard
MaritimeContraclors

Tacoma,WA
Mobile,AL
Jacksonville,
FL
Mobile,AL
New Orleans,LA
Bath,ME
Portland,ME
Mobite,AL
Sparrowspoint,MD
PortArthur,TX
Charleston,SC
Norfolk,VA
Mt. Pleasant,SC
Groton,CT
MossPoint,MS
Pascagoula,
MS
Savannah,GA
Long Beach,CA
Mare lsland,CA
Bellingham,WA

21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

21 5

McDermotl,Inc.
MetroMachineCorp.
Nat'lSteel& ShipbuildingCo.
NewparkShipbuilding& Repair
NewportNews Shipbuilding
NorfolkNavalShipyard
NOSHIPCO
NorthFloridaShipyards
PearlHarborNavalShipyard
PetersonBuilders
PhiladelphiaNavalShipyard
PortlandShip RepairYard
PortsmouthNavalShipyard
PugetSoundNavalShipyard
San FranciscoDrydock
SouthwestMarine
Tampa Shipyards
Todd PacificShipyards
TrinityMarineGroup

MorganCity,LA
Norfolk,VA
San Diego,CA
Houston,TX
NewportNews,VA
Portsmouth,VA
Nor.folk,
VA
Jacksonville,FL
PearlHarbor,Hl
SturgeonBay, Wl
Philadelphia,
PA
Portland,OR
Kittery,ME
Bremerton,WA
San Francisco,CA
San Diego,CA
Tampa,FL
Seattle,WA
Beaumont,TX

Fig. 1-6. Locations of U.S. shipyards.

cantly cheaper wage scalesthan almost all of

the European yards, and are about the same
level as those of Japan. Table 1-3 shows the
hourly shipyard wage scales for the major
shipbuilding nations for the period 1975-90,
and showsthe relative shift in U.S. position.
A part ofthe industry (about 200 yards)
concentrateson ship repair. As mentioned,

only about 17 repair yards can accommodate

vesselswith a length in excessof 220 meters;
about 55 more repair yards are capable of
dry-dockingvesselsof in excessof 100meters
in length. Many smaller repair yards, called
topside yards, have no dry-docking facilities
at all. Theseyards perform above-water ship
and barge repair, and generally employ less

INTRODUCTION

21

for shipyardworkersworldwide(U.S.$)
Table 1-3. Hourlycompensation
Country

1975

1980

UnitedStates
Canada

6.85
6.35

11.25 14.62 15.66

9.24 1 ' t . 9 4 1 6 . 9 3

HongKong
Japan
Korea
Singapore
Taiwan

1.05
3.92
0.54
1.20
0.57

Belgium
Denmark
Finland
France
Germany

8.75
7.04
5.73
5.13
7.22

15.75 10.71 N/A

1 1 . 9 0 8.60 19.37
9.15 9.14 23.19
1 0 . 2 3 8.86 16.43
14.02 10.91 23.45

Italy
Norway
Sweden
Kingdom
United

5.82
N/A
8.08
3.74

9 . 1 7 8.48 18.87
12.50 11.28 21.92
13.22 10.03 20.93
7.52
6.45 13.21

t.o/

6.75
1.65
1.95
1.54

1985 1990

2.32
8.12
2.13
3.20
2.46

3.78
15.04
6.52
5.16
7.50

FromU.S.Bureau
Nov.1991
of Labor
Statistics,
than 100 people.Topsideyards usually have
the capability of transporting men and material to the work site.
The U.S. ship repair industry continues
to be very active and competesvery aggressively in both domesticand foreign markets.
The U.S. shipyards benefit from their location in the United States (the world's most
important trading nation). Therefore, location, timeliness of repairs, and better pricing
can, and do, give the U.S. yard an advantage
over foreign competition.
The frnal element of the industry is the
group known as "second-tier"shipyards. These
yards are primarily engaged in supporting
inland waterway and coastal carriers. Their
focusis the constructionand repair of smaller
vessels,e.g., supply boats, tugboats, ferries,
frshing vessels, barges, and small military

and government-ownedvessels.This segment

of the industry had surplus capacity during
the 1980sand into the 1990s.But this excess
has disappeared,and the industry segment
is expected to have growth and firm prices
into the 21st century.[8]
3.3.The U.S. Shipbuilding
Market
The shipbuilding industry is unique among
all U.S. manufacturing and construction industries. First, since 1960, the industry has
played a negligible role in the world market
for merchant vessels. Second,the only domestic customers for ships has been either
the U.S. governmentor firms completelydependent on government policy. The navy is
the major customer for U.S. shipbuilding;
however, this sector is expectedto decrease
in relative importance. The countervailing

22

SHIPPRODUCTION

effect of the Navy construction program was

an important factor in keeping part of the
industry alive since the mid 1970s.The U.S.
Navy shipbuilding progr:am is the largest in
the western world. The Navy construction
program, bolstered during the 1980sby the
goal of attaining a 600-shipNavy, provideda
signifrcant boost to some U.S. shipbuilders.
With the end of the Cold War, however,the
Navy has sharply curtailed its fleet size requirement, and in early 1994 the fleet size
was set in the range of 350 ships. The result
has been significant reductionsin Navy procurement.
In the early 1990sthe Congressprovided
$2.1 billion to the Department of Defenseto
begin a fast sealift ship acquisition program.
A major component of the program is the
acquisition (through either new construction
or conversion)of additional sealift capacity
equal to 20 large, medium-speed(24 knots.)
roll-on/roll-off ships. Initial awards occurred
in 1993, with contracts being given to Newport News Shipbuilding, Avondale, and
NASSCO. It is expectedthat constmction/
modification of the ships will continue
through most of the decade.
In addition to the Navy constructionprogr&D, the Navy is the source of approximately 50 percent of the dollar value of overhaul, repair, and conversion work done by
commercial shipyards. This represents approximately 30 percent of the total Navy repair, overhaul, and conversionbudget. The
remaining 70 percent of this work is done in
the eight Navy shipyards. The 30/70 split is
based on an agreement between the Navy
and Congress and has remained relatively
constant over time.
Between 1950 and 1980,over 40 percent
of U.S.-built merchant ships receivedsubsidies, called Construction Differential Subsidies (CDS). This program provided a differential subsidy based on the difference between
foreign and domesticshipbuilding costs(not
to exceed50 percent of the domesticcost).

In 1981 the Reagan administration terminated thesesubsidies,and the commercialship

building market quickly disappeared.
The remainder of the domesticmarket is
made up mostly of directly or indirectly subsidized ships. U.S. cabotagelaws, called collectively the Jones Act after their principal
architect, reserveall U.S. inland and coastal
trade to vesselsthat have been built in tne
United States without CDS funds. For purposesof the law, Guam, Puerto Rico,and the
U.S. Virgin Islands are also consideredto be
coastal, or cabotage,trade. Since the cessation of the differential subsidy in 1981,all of
U.S. ships built for U.S. flag carriers, and not
subsidized,have been JonesAct ships. Since
1981 less than ten ships have been built in
U.S. yards-an averageof less than one ship
per year.
Reactingto the impact of the ExxonValdez grounding offthe coast ofAlaska, the U.S.
Congresspassedthe Oil Pollution Act of 1990
which requires ships and bargescarrying petroleum productsbe double-hulled.That legislation has stimulated the new construction
market, especially for double-hulled oceangoing barges and double-hulled barges for inland waterways.It is expectedthat there will
ultimately be an impact on new ship construction, too.
In 1992 the Clinton administration established a working group to developa proposal for a new maritime policy in an attempt
to reverse the decline of the U.S. merchant
marine, and to stimulate the redirection of
the shipyards from military construction to
commercialconstruction. The result was the
passageofthe DefenseConversion,Reinvestment, and Transition Act of 1992.The act had
specificelementsdirected at revitalization of
the shipbuilding industry, including (1) support for construction of ships that have
market penetration potential, (2) the establishment of National Maritime Technology
Centers to support technologytransfer, and
(3) cost-sharingof projectsthat will augment

INTRODUCTION

and accelerateinfrastructure changes.One of

the objectivesofthe program is to achievea
10 percent market share penetration in new
ship constructionby the year 2000.
Other government programs have included Title XI frnancing and two tax incentive programs, the Capital Construction Fund
(CCF) and the Construction Reserve Fund
(CRF). Title XI provided financing guarantees
for merchant ships built in U.S. shipyards.
The CCF and CRF assist operators in accumulating capital to build, acquire, and reconstruct vesselsthrough the deferral of federal
income taxes on eligible deposits.Thesegovernment assistanceprograms,however,have
been inadequate to sustain a viable market
demand.
3.4. ResourceMarkets
Shipbuilding resourcemarkets are very heavily in{luenced by the nature of the product
markets. Price and lead times of material
react to the cyclesand small size of the product market. Similarly, the quality and availability of the labor force is shaped,both directly and indirectly, by these same factors.
3.4.1. Labor Force.The shipbuilding and repair industry is labor intensive, comparedto
all manufacturing industries. Within the industry, repair is more labor intensive than
new constmction. Ship repair is involved with
smaller numbers of similar parts, and much
of the work is accomplished aboard ship,
where the opportunity for automation is reduced. Manpower skills used in new construction are basicallythe sameas for repair,
but the mix of skills is different. Shipbuilding
employs more of the structural trades (including welders, burners, and fitters), while
ship repair tends to use more of the outfitting
trades.
Shortage of skilled labor is a continuing
problem, particularly in the private sector.
Possible reasons for this are relatively low
wages, higher involuntary labor turnover

23

(Iayoffs),and few large-scaleapprenticeprograms, relative to either Navy shipyards or

related private industries. Because Navy
shipyard workloads are planned in advance,
and inflexible Civil Service regulations and
manpowerceilingsapply, Navy yard employment exhibits more stability than private
yard employment. Navy shipyards, in their
role as a support function to the Navy, must
be able to respondquickly to requirementsto
repair and install highly sophisticated systems on short notice.The Navy shipyard apprentice training program ranks with the
best in the country, and through specialization by eachyard, highly skilled work teams
are maintained for eachequipmenttype used
in the Navy. This investment in training is
an important factor in attracting and retaining quality personnel.
Private yards, on the other hand, must be
able to adjust the size of the work forceto the
workload. Higher involuntary turnover induces higher voluntary turnover, which is
exacerbatedby generally poorer work conditions and lower pay relative to other industries. Consequently,shipyards are reluctant
to institute costly training programs, lest
they end up training personnelfor other industries.
Faced with a labor shortage, the usual
initial response(adjustment) is to increase
overtime and recruiting effort. Of course,the
logical place to recruit is in the proximity of
other shipyards. However, studies of shipyard labor indicate that it is relatively nonmobile comparedto other industries.[9] Even
when a pool of workers exists within one
geographicarea, recruiting may be ineffective. In a unionized area, pay scalesare set
by contract and incentives for changing yards
are very limited.
Given a lack of mobility and the length of
time required to producea skilled craftsman,
additional skilled workers may not exist in
the locality and time frame in which they
are needed. Management may then con-

24

SHIPPRODUCTION

sciously turn to skill dilution in order to

reach numerical requirements.Skill dilution
may take the form of increasing the ratio of
trainees to journeymen, redefining what constitutes a journeS,rnan,or somecombinationof
the two.
If qualified mechanicsare not available
when demand increases,inputs to training
programs must generally be increased.This
is a reasonableshort-run solution to a temporary shortage.However, the training of a journeyrnan may take anywhere from three to six
years.Becauseofthe cyclesofthe market and
the length of market cycles, the period of
increaseddemand may be shorter than the
time required for training. In apparent responseto this phenomenon,only a few shipyards maintain ongoing apprentice training
programs. Notable exceptionsinclude yards
that participate in regional training programs and yards that have an isolated labor
supply and lower-than-averageturnover. They
are generally able to meet their own needsfor
skill training through formal apprentice programs. Many yards rely primarily on the
construction trades for basic skill training,
and provide only minimum shipyard-specific
training to new hires.
A labor shortagecan occurbecauseof an
inability to retain workers as well as an inability to hire qualifred personnel in the frrst
place. Turnover, which measures retention
(the sum of accessionsand separations per
100 employees),is historically higher for private shipyards than for Navy yards or other
industries, such as construction, which use
similar skills.
In general, turnover is lower for industries using more highly skilled workers.
Highly skilled labor is more costly to train
and replace. Therefore, employers will endeavor to retain these workers. However,in
the caseof shipbuilding and repair, which is
generally skill-intensive, this does not hold
true. Shipbuilding and repair exhibits signifr-

cantly higher turnover than other, similar

industries (not considering the Navy shipyards). This includes industries such as primary metal, fabricated metal products, transportation equipment,and aircraft. Additionally, wages in shipbuilding and repair are
generally lower than for other competingindustries, averaging about 15 percent less
when comparedto the aircraft, transporbation
equipment, and heavy constructionindustries.
Past studieshave determinedthat wagesare
higher in public than in private yards.
The shipyard labor market appearsto be
primarily influencedby the cyclesof the product market. This is evidencedby high turnover, both "quits" and layoffs, and a general
lack of formal apprentice training programs
relative to other skill-intensive industries.
Workers experiencehigh layoff rates, which
tend to result in high quit rates as well.
Many workers who expectto be laid off take
the initiative and migrate to other shipyards
or other industries.[10]In general,wagesare
low relative to other skill-intensive industries, and this is reflected in high industry
turnover. This situation is not necessarily
true in many local markets. For example,
formal training programs and higher wages
prevail in Navy shipyards. However, only a
few private yards invest heavily in training,
primarily becauseof high trainee attrition
and major fluctuations in workload.
3.4.2. Material.Depending on type, between
40 and 70 percentofthe total cost ofa ship is
material and subcontracted services. The
material market parallels the labor market,
in that it is heavily influencedby the characteristics of the product market. However, in
the short run, price and availability of material are more responsive to the effects of the
economicactivity in other industries. This is
due,in part, to the relatively small shipbuilding market in the United States,comparedto
the overall industrial base. Manufacturing

INTRODUCTION

lead times, another measure of the performance of the supplier base,have also generally exhibited the expectedresponseto economic conditions.Thus, lead times will also
be influenced more by general economicconditions than by the status of the shipbuilding
industry.
The shipbuilding supplier base, along
with other predominantly defense-oriented
industries, has declinedsincethe 1950s.The
U.S. industry is, in several instances, dependent on a single supplier. Examples rnclude suppliers ofanchors, anchor chain, and
activated rudders. In the case of quiet ball
bearings used in submarines and many surface ships, the single source is a Japanese
manufacturer. Reasons generally cited for
the declinein the supplier base are:
. cyclesand small size of the market
o costsassociatedwith government
regulations
r problems associatedwith military
specifrcations/standards
The last two apply specificallyto suppliers of
military equipment but, becauseof the importance of the government market, affect
the entire industry.
The effect of the size and cycles of the
market receivedconsiderableCongressional
attention throughout the 1970s.
However, very little has changed as a
result of extensivetestimony and somedocumentation by the Navy. Many subcontractors
and component suppliers simply frnd it uneconomicalto upgradeor evenretain capacity
for a market that is both small and sporadic.
Additionally, the small subcontractorssee a
much more cyclical market than is apparent
from observing the aggregate.When business
is slack, shipbuilders tend to make components in their own facilities-components
that would be subcontracted in times of
greater demand. The result is an accentua-

25

tion of the feast-or-faminenature of the market for the lower-tier suppliers.

The cost of government regulations, including pollution abatement,becameparticularly significant during the 1970s.In most
casesthesecostsapply acrossthe board to all
industries. Theoreticallythey should not adversely affect any one industry. However,
complianceassurancesare included in government contractrequirements.This may result in suppliers of material to industries for
which the government is the primary customer bearing a disproportionate share of
the costs.
Military specificationVstandardsare a perennial problem for suppliers of Navy shipbuilding and repair programs. Manufacturers
state that military specifrcations,in general,
do not keep pace with the technologicalprogress made by the industry.[llJ Specific
problems are:
r interpretation of requirements
r outdated specifications
. equally acceptablecommercialcomponents not qualifred under military
specifrcations
. unreasonable software/documentation
requirements
. high cost of quali$'ing a product not commensurate with the expectedreturn
3.5. IndustryOrganization
and Performance
From an industry point of view, U.S. shipbuilding and repair can be characterized as
a relatively small industry operating in a
highly cyclicalmarket controlled by the U.S.
government.Technologically,the industry is
somewhat backward compared to the aerospaceindustry, which is similar in terms of
market structure. The economicvalue of the
commercial side of the industry is minimal,
and with few exceptions,ships can be purchasedabroad more cheaply.Even the Navy
has examined the feasibility of purchasing

26

SHIPPRODUCTION

ships outside the United States. The strategic value of the industry is well recognized,
however, and maintenance of a peacetime
shipbuilding industrial base is a practical
necessity. The industrial organization and
investment policies of the industry have
evolved in responseto signals from the government, and the structure of the industry is
essentially a product of national maritime
policy, or the lack ofsuch a policy.
The industry is anything but homogeneous in terms of ownership,companysize,and
primary markets. However,somerough generalizations can be made with respect to
profrtability and capital budgeting policies.
Beginning in 1959, the major shipyards
changed from mostly independent firms to
divisions of large corporations.With the exception of Todd, this transition was accomplished through conglomeratemerger or acquisition by aerospaceor other large companies.
The reasonsfor conglomeratemerger are
complex. In general, however, this form of
acquisition is used because it provides a
means of increasing the price/earningsratio
ofthe parent company,and offersscaleeconomies in the capital market without running
afoul of the antitrust laws. Either vertical
integration or merger with a company that
manufactures the same basic product line is
often infeasible for large corporations becauseof the antitrust laws.
Whatever the particular corporate reasons for acquiring shipyards, it is not clear
that such acquisitionswere successful,based
on usual economicstandards. The business
environment has been characterizedas "one
of high risk and low returns."[12] Profitability
has in general been low. It has been shown
that profrtability was two-tiered: mediumsized yards that retained their traditional
old line shipbuilding identity were profrtable,
while those that were primarily aerospace
oriented sustained consistent losses.It has
also been suggestedthat cash flow rather

than immediate profits played an important

part in corporate decisionsto acquire shipyards.[13JThroughout the period of conglomerate takeover,the Navy made progresspayments weekly,while most ofthe shipbuilders'
bills were paid monthly. Such "free loans"
would be an important sourceof frnancing for
the parent company.Due to the labor-intensive nature of shipbuilding and the fact that
little capital investment was required for
the shipyards when they were purchasedby
conglomerates,the opportunity existed for a
high ratio ofprofrt to invested capital. This,
coupledwith relatively low purchaseprices,
made entry into the shipbuilding industry
attractive to conglomerates.
In retrospect,the capital budgeting policiesofthe major shipyard companiesindicate
a fairly consistentpattern of miscalculation.
However, in terms of the signals emanating
from the federal government and the world
economyimmediately prior to and after 1970,
these policies appear very logical. Prior to
1970,investment in shipyards was very conservative. This reflected an industry for
which the market prospectswere minimal.
Most investment was aimed at maintenance
of the ability to remain in the business of
building ships rather than expansion. Foreseeabledemand and potential for becoming
competitivein the world market certainly did
not warrant an expansionarypolicy. Control
over the domestic market was virtually nil,
and government policy did not indicate any
major changesin demand.
Commencingin the mid to late 1960s,the
Navy and the Maritime Administration signaled major changesin government policy.
The Naly embarkedon an efforbto implement
Total Package Procurement (TPP), whereby
the prime contractor is responsiblefor all of
the componentsof a weaponssystem rather
than just the platform. Seriesproduction of
a class of ships was intrinsic to the concept,
as opposedto allocating a handful of ships to

INTRODUCTION 27

eachofseveral yards. The Navy hopedto shift

management responsibility for the total
weapons system to the contractor, and to
gain effrciency through series production.
From the viewpoint of the shipbuilder, TPP
required more sophisticated management,
the necessity of integrating several diverse
technologies and product lines, and large financial resources. These were considered to
be primary strengths of aerospaceand conglomerate firms, which were consequently
attracted to the industry.
Concunent with introduction of Total Package Procurement,the Congressenacteda Maritime Administration program calling for the
constructionof 300 commercialvesselsin the
period 1970-1980.This program was predicated on a projected expanded world market,
particularly for very large crude carriers and
high technolory ships, such as liquefied natural gas carriers. The MarAd 70 program, as
it was called, also hinged on productivity
gains through standardization and series production.
In apparent response to the Navy and
MarAd incentives, the industry invested an
averageof $200 million (4.5 percent of sales)
annually between 1970 and 1978.[14]This
was a period during which industry profrts
were nearly nonexistent. It is estimated that
in the period 1969-1976,the industry borrowed externally, or from the parent companies, approximately$800million, while earning profrts of less than $50 million. To put
this in perspective,this means that the industry took on a long-term debt equal to 2.5
times its net worth. The MarAd 70 program
was not successful,resulting during the 1970s
in construction of only 80 new ships and the
conversionor reconstruction of 56 ships. With
the subsequent collapse of the commercial
market, and failure to achieve the higher
profits commensurate with the added risk
assumedunder TPP, the financial position of
much of the industry was shaky at best.

Shipbuilding is two tiered: it is a defense

industry, but there is also an important commercial side to the market. Defense industries are characterizedby a lack ofcontrol over
markets and a consequentadversarial relationship between producers and customer. Even
much of the commercial side ofthe shipbuilding market is controlled by the government.
Profitability and capital investment are therefore highly responsive to government policy.
The decline of military construction will
require a fundamental change in the industry's organizational structure, evenmore dramatic than that which occurred with the termination ofthe subsidysupportprogram in 1981.
It appears that by the turn ofthe century the
U.S. shipbuilding industry will have reduced
its employment by at least 40 percent from
what it was in 1990.
The survival of any specific shipyard will
be perhaps attributable to two primary factors. The first will be the ability of the yard
to maintain its military market. This factor
will undoubtedly impact Newport News, Electric Boat, Bath Iron Works, and Ingalls.
Newport News, whose parent corporationis
the largest of the shipbuilding conglomerates, also enjoys a position of some market
power, compared to most of the industry.
Newport News is one of the three nuclear
yards and is the only builder of aircraft carriers. Electric Boat is dedicatedto building
nuclear submarines,and wiII probablymaintain that assignment. Bath Iron Works is the
lead yard on the Burke-classdestroyer.And
Ingalls is the lead yard on the Aegis cruisers.
The second factor is the ability of the
shipyard to develop and maintain a commercial ship market. In this instance,the factors
favor the smaller, more agile shipyards, or
shipyards that can more easily leverage from
their military ship construction program into
commercialship construction.Yards that will
survive as a result of this secondfactor might
include Avondale and NASSCO. who cur-

28

SHIPPRODUCTION

rently dominate in the construction of naval

auxiliary ships, or McDermott and Trinity,
who have had extensive experience in the
construction of ships and platforms for the
offshore industry.
An observable response to government
market control is conservatism,both technologic and economic.This is evidencedin the
defenseindustries by slow growth rates, low
profitability, and apparent inability to market high-technology products in the more profitable civilian markets.[lS] The stmcture
and performance of the shipbuilding industry
has been greatly influenced by government
policies that have been implemented on a regular basisduring the last 50 years.It remains
to be seen what is in store for the industry in
the future. Government policy will certainly
play an important role. However,many shipbuilders have embarkedon a courseofinvesting in innovations in management and processesgearedto operatingefliciently in diversifredmarkets. In the final analysis, the performance of the industry will be determined
by the successofthese innovations.
3.6.TechnologyDevelopments
and Productivity
Efficient production in any industry depends
on the nature of demand for the product and
the relative prices of resources.Much has
been written about labor intensivenessand
capital formation in the shipbuilding industry comparedto other industries. Similarly,
shipbuilding technical processesand labor
intensivenessin the United Stateshave been
measured against foreign shipbuilders. The
conclusions have generally been that U.S.
shipbuilding is more labor intensive than
other U.S. industries and foreign shipbuilders. Suchcomparisonsare meaningless,however, by themselves. The economic performance implications of labor intensivenesscan
be evaluated only in the context of resource
and product markets.

Although labor rates have shaped national policy, they do not explain the labor
intensivenessofthe industry nor the low productivity compared to foreign shipbuilders
and other construction and batch manufacturing industries. The U.S. government has
shapeddemand,and demand has guided technolory and capital formation. The demand for
ships in the United States has historically
been uneven and, more importantly, uncertain. Demand is based on the underlying
world demand for shipping and national defense requirements, both highly cyclical, as
modifred by the political and economicclimate. Additionally, in an attempt to broaden
the shipbuilding base, Na'vy acquisition has
traditionally been spread among several
builders. Programs comprising 20 and 30
ships have been allocated to as many as
sevenshipyards. The fragmentation of orders
and ensuing uncertainty did not offer incentive either to modernize facilities or reorganize pmduction. The result was a high-cost
labor-intensive industry.
In the late 1960s and 1970s, the Navy
initiated acquisition policies, such as multiship contracts,intended to encourageinvestment in shipbuilding technology.In addition
to Navy efforts to provide incentives for modernization, the Maritime Administration developed standard ship designs to promote
multi-ship construction programs. Through
the National Shipbuilding ResearchProgram,
the Maritime Administration also conducted
an active education and technology transfer
program in conjunctionwith the shipbuilding
industry. The industry responded to these
policies and has made major organizational
and technologicalchangesover the past ten
years. Despite significant capital expenditures
in the 1970s,an overall decreasein productivity during approximately the same period
was reported to be between 15 and 35 percent.[16]The United States is recosnizedas

INTRODUCTION

a leader in the construction of nuclear and

conventionally powerednaval combatant ships.
But in terms of man-hours and dollar costs,
productivity has also continued to declinein
this sector.
In late 1979a team ofsix individuals experienced in shipbuilding visited six Japanese
shipyards.[17]The visit was intendedto identify low-investment, high-return shipbuilding
technology. The purpose of the study, and
subsequentreports publishedby membersof
the team, was to "encourageU.S. shipbuilders
to adopt the observedadvancedtechniquesfor
the purpose of improving productivity." Primary reasonscited for high productivity were:
o the utilization and application of the
logic and principles of zoneplanning
and construction
. the developmentand use of a very
effective material classifrcation scheme
for definition, procurement, and control
of material
r the extensiveuse and continued development of high-quality shipbuilding
standards and modules
r the rationalized developmentand use
of effective cosVman-hour-reducing
computer aids
A key element of the abovelist of reasonsfor
high productivity is that they do not depend
on large orders and series production. The
list does,however,highlight the principles of
group technology through block construction
basedon zone,problem area, and stage.
In 1994,a projectwas completedthat had
conducteda comparative survey of major U.S.
and selectedforeign shipyards.[18]The survey indicated that U.S. shipyards, on average, employed lower levels of technology
than that found in five foreign yards in all
eight major functional areasthat were examined. Importantly, the gap in technological
differenceshad increased from a level that

29

was found in a similar survey conducted in

1978.[19]
The functional areas that were evaluated were: outfitproduction and stores,desigrr/
drafting/production engineeringAofting, ship
construction, steelwork production, organization and operating systems,layout and environment, and other pre-erectionactivities.
The largest differences were noted in the
first two (outfit production and stores, and
design/drafting/pnrduction engineering/lofting).
There was a medium gap in the next three
(ship construction, steelwork production, and
organization and operating systems). The
smallest gap existed in the last two (layout
and environment and other pre-erection activities). The study concluded that even
though U.S. yards havelost ground compared
to their foreign competitors in the sixteen
yearsbetween 1978and 1994,there were some
reasonsfor optimism. Labor costsand average
hours worked for U.S. yards are world competitive. Additionally, technology improvements neededare generally of the sofbor management technologytype, rather than facility
or hardware type. Thus, major capital improvementsare not required to producemajor productivity improvements.
References
1. D'ArcangeIo,E.,Ship Designand Construc/lon, New York: Society of Naval Architects
and Marine Engineers,1969.
2. Leanse,J.I., and L.L. Jaekel,'"TheStatusof
Skilled Tlades Tlaining in U.S. Shipyards,"
National Shipbuilding ResearchProgram,
Maritime Administration, in cooperationwith
the University of Michigan, December 1983.
3. Fisher,J.R.,and P.J. Coady,"U.S.Shipbuilding: The Seventiesin RetrospecV
The Prospectsfor the Future," Washington:
National DefenseUniversity, June 1981.
4. Peters,Hans Jurgen, "The International
OceanTransport Industry in Crisis," Working Paper #WPS 1128,Washington:World
Bank, April 1993.

30

SHIPPRoDUcTIoN

5 . I b i d ,p . 4 6 .
6. U.S. Industrial Outlook, 1993--Shipbuilding
and Repair,pp.2I-22.
7. Whitehurst, C.H., The U.S.Shipbuilding
Industry, Annapolis, Md: Naval Institute
Press, 1986.
8. Op. cit., U.S.Industrial Outlook,pp.21-23.
9. Martin, J.C., "The Labor Market of the
United States Shipbuilding Industry,"
Ph.D. dissertation,Washington: George
Washington University, 1978.
10. Martin, J.C., N.B. Davis, M.H. Kahn, and
W.J. Shafer, "Accomplishing Shipyard
Work for the United States Navy: Institutions, Systemsand Operations,"Institute
for DefenseAnalysis, August 1975.
11. Coordinatorof Shipbuilding, Conversion,and
Repair, Department of Defense,"Annual
Report on the Status ofthe Shipbuilding
and Ship Repair Industry of the United
States,"1982.
72. Kattz, E.M., and Associates,'"TheCapital
Budgeting Policiesof the U.S. Shipbuilding
Industry: An Analysis of Defense Industry
Behavior,"Office of Naval Research,
October 1979.

13. Kavanagh,G.L., "The United States

Shipbuilding Industry and Influences of
Conglomerates,"Mastels Ttresis,Cambridge:
Massachusetts Institute of Technology,
May 1977.
t4. Op. cit., Kaitz.
15.Op. cit., Kaitz.
16.Marine Equipment Leasing Inc., "Technolory Survey of Major U.S. Shipyards 1978,"
National Shipbuilding ResearchProgram,
Maritime Administration, 1979.
t 7 . Vander Scaaf,J.R., "JapaneseTechnology
That Could Improve U.S. Shipbuilding Productivity," REAPS 7th Annual Symposium
Proceedings,1980.[Containedin NSRP
DocumentationCenter Library, #0007.1
1 8 . Storch,R.L., J. Clark, and T. Lamb, "TechnologySurvey of U.S. Shipyards-1994."
1995NSRP Ship ProductionSymposium,
Seattle.
1 9 . Lowry, R.L., "TechnologySurvey of U.S.
Shipyards,"Transactions,SNAME, Vol. 88,
1980,pg. 151-772.

CHAPTER II

SH I P B U I L D I N G M A N A G E M E N T T H E O R Y

1. lntroduction
generalprinciplesofgroup technology,which
are introduced and related to the traditional
economicmodel of shipbuilding in the second
sectionof this chapter.Two work breakdown
structures, one system-orientedand one product-oriented,are describedin the last section. The product-orientedwork breakdown
structure (PWBS) is the vehicle for applying
group technologyto ship construction.PWBS
will be exploredin depth in Chapter III.

The theoretical,economicmodel of shipbuilding introduced in this chapter is based on

theoretical and empirical analyses of shipbuilding programs carried out in U.S. shipyards in the 1970sand 1980s.[1]The model
will be used to highlight major cost drivers in
ship construction and to show how costscan
be controlled through improved organization
of the work.
Shipbuilding work can be organizedusing

2. Shipbuilding
Economic
Theory
productionrate varies over time accordingto
the stageof completion.Second,ship cost estimating is inexact and as construction proceeds,production rates have to be adjusted.
Consequently,this shipbuilding model reflects
the effects, on total cost, of production rate
changes over time. These effects, and the
underlying variables which determine production rate, will be examinedin detail as the
modelis developed.This modelhas beenused
to analyze and explain cost overruns in shipbuilding programs. It is used here to gain
insights into costdrivers and how costsmight
be controlled using better production managementtechniques.
Defining a production function for shipbuilding entails three basicbut essentialcon-

The shipbuilding model is based on the theory of production applied to a production run
for a predetermined,frxedtotal output. When
total output is frxed,the producer'sdecisions
include the productionrate and when to start
and stop production. These decisionsare, of
course, interdependent; when any two are
made, the third is determined. Total output
is fixed at the time of contract signing. The
times to start and complete production are
generally determined as part of the shipbuilder's bid, subjectto somemodifrcationas
a result of frnal negotiations.Consequently,
the average production rate is also determined. However, traditional shipbuilding,
comparedto most manufacturing industries,
is complicatedby two important factors. First,

31

32

S H I PP R O D U C T I O N

siderations:(1) defining a measureof output,

(2) the time dependenceof production costs
on resourceutilization, and (3) the effect on
ship production costs of changesin production rate.
Ships are often producedsingly or in batches
of a few over a period ofyears. Therefore,it
generally does not make sense to treat an
individual ship as a unit of output. Usually
output is measuredas percentagecompletion
of the ship. This is a very subjectivemeasure
and dependsheavily on assumptions about
the work content of all the componentswhich
make up the ship. Components,called work
packages,are estimated to require somenumber of man-hours,accordingto certain industry
or shipyard standards. The actual required
hours vary according to when and in what
order the work is done,and what other work
is being accomplishedin the samegeographic
area of the ship. Additionally, the required
man-hours for work packages are in fact
budgets. If a work package is not complete
when the budget is used up, there is a tendencyto "borrow" man-hours from other jobs
and credit them to the work packagewhich
has overmn its budget. In effect the shipbuilder starts with less than an exact estimate of the number of man-hoursrequired to
build the ship, and then attempts to measure
progressby counting inputs (man-hours).
Progress is also cross-checkedusing actual physical completion,such as feet of pipe
installed, tons of steel in place,or number of
compartmentsclosedout. However,theseare
not homogeneousunits. A ton of steel is not
necessarily equivalent, in terms of output,
throughout the sameship. Such a measureis
even less meaningful when used acrossdifferent ship designs or $rpes. "Equivalent tons,"
used by many authors, is basedon the work
content of a ton of steel in some theoretical
ship type (see glossary). This assumeslike
capital and work processesacrossshipyards,
another shaky assumption.

The second and third major considerations, time dependenceand production rate
effectson cost, are particularly important to
this discussion.A ship consistsof many intermediate products, called work packages.A
work package consists of some number of
required man-hours.An increasein required
output implies an increasein number of work
packages or an increase in required manhours per work package.The optimum erection sequenceimplies the availability of each
completedwork packageat a specificpoint
in time. Completedinterim products(work
in process)represent significant inventory
costs.If interim products are completedtoo
early, capital is tied up in the form ofvalue
added, and storage costs may be increased.
Interim products generally require more expensive storage facilities, corrosion control,
and security than raw materials. Late completion ofinterim products results in bottleneck delays and interference among work
packageswhich were originally scheduledfor
different time periods.Similar problems can
be anticipated if material or equipment that
is supplied by vendors outside the shipyard
arrives too early or too late.
The shipbuilding progress curve is Sshaped,as shown in Figure 2-I(a). This implies a bell-shaped production rate. If production proceedsaccordingto plan, the production rate curve correspondsto the rate of
resourceapplication, as shown in Figure 21(b). In the early stages of ship production,
steel is being cut and formed. At this time,
outfitting is limited; the work area is limited
by shop and platen space,and these determine the optimum production rate. As erection proceeds,more resourcescan be applied
in the form of simultaneous prefabrication,
fabrication,erection,and outfitting. This corresponds to the central portion of the curve
shown in Figure 2-I. Later in the cycle, the
work area is again severelyrestricted.A limited number of workers can be in the same

SHIPBUILDING
MANAGEMENT
THEORY

compartment, and many operations,such as

painting and welding, are completelyincompatible. The rate ofresource utilization (production rate) therefore decreasesduring the
last half of the construction cycle until it
again reacheszero.
Optimum construction rate is determined by severalfactors which include fixed costs,
work space,inventory costs,and changesin the
construction rate. The latter factor will be
discussedin detail below. In addition to the
cost of capital and the usual overheaditems,
fixed costsinclude security,fire watches,corrosion control, and some support crafts. A
ship sitting in the yard incurs these costs
whether work is being accomplishedor not.
Consequently,these costs lead to a shortening of the optimum construction time. This
effect is countered by costs associatedwith
increased construction rate. As manning
and construction rate are increased, so is
interference among the crafts. This effect
tends to lengthen the optimum construction
time.
Figure 2-1(b)can be interpreted as either
production rate (output per unit time) or required labor hours (manning) per unit time
as long as actual production correspondsto
planned (optimum) production. If interim products are not available as early as planned or
if changesto plans or specifrcationsadd required resourcesat any point in time, then
the time path of construction is no longer
optimum. In other words, a truly optimum
plan leavesno room for shifting resourcesto
a different time period. Such a shift would by
defrnition result in local interference among
crafts. To compensatefor "unavoidable"schedule slippage, the actual time path of construction usually allows some slack. In
creating flexibility, the schedule departs
from the theoretically optimum time path.
However, it will be shown in the remainder
of this chapter and in Chapter III that flexibility can also be gained through better or-

33

u1
>U
FF
<O

>Od

TIME

U
F
cc

zA
oz
FZ
oz
l<
o>
o
tr
L

TIME
Fig. 2-1.Productionplan for traditionalship construction.
ganization ofthe work and tighter control of
resourcesand work processes.
The dynamic shipbuilding model presented in the next subsectionexplains the
combinedeffect of resourcemarket phenomena presentedin Chapter I, and changesto
the originally planned production schedule.
The optimum constmction rate over time dependson frxed capital, the quantity and quality of workers, required changesto the original plan, and changesin the number of workers. During the construction of a ship, or a
seriesof ships,only changesin the number of
workers are generally controllableby the shipbuilder. The model is used to show how factors not under the control ofthe shipbuilder
may disrupt the constmction processand reduceproductivity, and how this disruption is
traded off against delay. It will be shown in
Section 4 that many of the adverse effects of
these factors can be reduced or eliminated
using group technology.
2.1.The Shipbuilding
Model
Once a ship, or series of ships, has been
contracted for, the shipbuilder's incentives

34

SHIPPRODUCTION

are to producethe ship(s)in the specifredtime

period at minimum cost.Wage rates are generally not controllable by the firm, different
crafts are not usually substitutable,and most
material is contractedfor in advanceor supplied by the owner. Consequently,cost minimization is essentially equivalent to minimization of total labor costs.
Shipbuilding is assumed to consist of
many (n) discrete operations or tasks. The
output of most of these tasks is an interim
product, which is an input for a subsequent
product. Each interim product is called a
work package.Output is defined in terms of
the number of work packagesrequired. The
total required output (Q) is the sum of all
required work packages(n).
Progresson the ship is measured as the
ratio of output at time t, Q1,to required output. This is just the number of work packages
completed at time t, (WPt) divided by the
number required. The units of output can
also be thought of as "standard man-hours"
(SMH). For example, suppose a particular
task (say, the installation of 1,000 feet of
pipe) is expectedto require 200 man-hours.
One thousand feet of pipe actually in place is
then equivalent to Q=200 standard manhours of physical output. Note that the number of actual man-hours used to install the
1,000feet of pipe in the examplemay or may
not be 200. Progressat time t is simply:

tual time is t'. At this time, planned progress

is about 40 percent.Actual progressis about
35 percent, while actual man-hour expenditures are nearly 60 percent of the total manhour budget.The variance indicates that some
work packages are requiring more manhours than were estimated and budgeted.
2.1.1. ProductionRelationship.A stable production technology across operations is assumed, and work packages are defined to
consistof the samenumber of required standard man-hours.It is assumedthat the rate of
production at time t, q1,dependson the level
of manning and capital facilities.
Rate of productionin shipbuilding is also
dependenton changesto the productionplan,
either becauseof customer-initiatedchanges
to plans or specifications,or becauseofestimation or scheduling errors by the builder.
Changes may require additions to the required number of man-hours per work package or to the number of work packages.They
may also require rescheduling of work to a
different time frame than originally planned.
The rescheduling of work and the physical
interface of changed work with unchanged
work may causedisruption of the production
PLAN (SMH)
A C T U A L P R O G R E S S( S M H )
A C T U A LM A N - H O U R S( M H )

number of work packages completed at time t, in SMH

_
n r =
H

In shipbuilding terminology,nl is percent

physical completion or just percent completion. Note that work packagesare assumed
to be of uniform work content. This is a simplifying assumption which is generally not
true in shipbuilding. Work packagesof uniform work content are a goal in group technolory shipbuilding for reasonswhich will be
explained later.
Figure 2-2 is an exampleof a ship production progresscurve. In this example,the ac-

a
:

G3.
-l

2
=
U\l
OF
rrq

tz

dg

Fig.2-2. Sampleship productionprogresscurve.

SHIPBUILDING
MANAGEMENT
THEORY

processthroughout the ship. It may also result in spillover effectson other ships or programs.
These assumptions can be stated as a
standard production function.
Qt

- q (Kt, Lt, Et)

where:
Kt = rate of use of capital at time t
Lt = rate of use of labor at time t
E6 captures the efficiency effects of changes
to required output. The componentsof E include the effects of incomplete or missing
plans, disruption ofworkers or support services, and worker morale.
Physical output rate, qt, is measured in
terms such as feet of pipe installed or tons of
steel erectedper day. However,a measureof
output rate is neededwhich will allow a comparison oftons ofsteel and feet ofpipe. The
units for production rate are "standard manhours" per unit time. A standard man-hour is
the amount of physical completion which
should be accomplishedby one worker working one hour if all of the other determinants
of productivity are as originally planned.
In the long run the shipyard's capital and
labor inputs can be adjusted to achieve the
optimal size for the yard. The shipbuilder's
choiceof capital stock will depend on many
things, including expectedlong-run demand
and the need for flexibility. The size of the
capital stock,in turn, will also determine the
optimal rate of production for the yard.
In the short run, capital is frxed and output is varied by adjusting the sizeof the work
force or the averagelength ofthe work day.
Short run doesnot mean that the shipbuilder
cannot change capital stock but rather that
such changesare not soughtbecausechanges
to required production rate are viewed as
temporary.
Labor, L1, is defined as quality-adjusted
labor. Labor is not homogeneous,and the pro-

35

ductivity per unit time of a worker depends

on many factors, such as:
o ship sequencenumber
. number of workers (manning level)
o time rate of change in the number of
workers
.length of the work day (overtime)
. skill level of the workers
Ship sequence number measures learning
which is expectedto occur when more than
one ship of the sametype is built in the same
yard.[2,31
The number of workers (manning level)
determines output directly. It also affects effrciency due to congestion and craft interference.If the manning level in a confinedarea
is greater than the planned optimum, output
per worker is decreased.
The time rate of change in the number of
workers affects productivity becausechanges
absorb resourcesthat could be used in producing output. This will be exploredin depth
in the adjustment cost sectionbelow.
Skill level is a straightforward index of
labor quality. It is often representedas the
number of skilled workers (journeymen) divided by the total number of workers (journeymen plus apprentices and helpers). The
greater this ratio, the more productive the
work forceis expectedto be. There is, ofcourse,
somepoint at which productivity might begin
to decrease,becausethere are too many experienced workers and not enough helpers to do
the more mundane tasks.
Experiencelevel is related to familiarity
with the specifictask at hand and crew integrity. Ship construction is accomplishedby
teams of varying size located at work stations. If the compositionof theseteams is not
constant,crew integrity is lost and productivity is decreased.Workers newly transferred
to a crew require orientation and on-the-job
training, which reduces the productivity of

36

SHIPPRODUCTION

the crew. Experiencelevel is commonly represented by total yard or craft turnover. Journeyrnan/total worker ratio and turnover are
both commonlyused predictors of productivity in shipbuilding.
2.1.2. ExperienceCurue Effects.The experiencecurve is the name applied to an observed
relationship between costs and accumulated
experience.It is related to quality adjusted
labor, as defined earlier, and is used for a
variety of operating management purposes,
especiallyin the areas of planning and cost
control. The conceptof the experiencecurve
can be expressed,mathematically, as follows:
V' = a./xb
where:
V* = the averagecost of"x" units
produced
x = rurnber of identical units
a = the cost ofthe frrst unit produced
b = &n exponent that varies with the
complexity of the ship
The experiencecurve indicates that the averagecostof all units produceddeclinesat some
constant rate as a function of accumulated
experience.The term is now defrned as the
percentimprovement that occursin the average cost of all units producedeach time the
accumulated unit production is doubled. In
the shipbuilding industry, the experience
curve benefit has generally been found to be
in the 5-10 percent range. It should be noted
that while unit costs (exclusiveof inflation)
tend to decrease,the reduction is not automatic. Such cost decreasescan, in fact, result
only from the actions of management.
The reasons underlying the experience
curve effect include:
o learning
. specialization
o investment
. scale effects

As describedearlier, peoplelearn to do tasks

better and more efficiently as they become
more experienced.Labor costs per unit decreasewith accumulatedexperience.This is
the well-known learning curve for labor
which has been documentedin management
literature. (It should be noted that the formula, shown above,would also apply in developing learning curve benefits, exceptthat
the costs would be expressedin labor units,
rather than in monetary units.)
When two or more peopleare involved in
the sametask, it is often possibleto break that
task into two or more parts. Then, for a given
output, each person doeshis respectivetask
more often. This specialization of task permits each worker to accumulate experience
twice as fast as before. From the learning
curve effectabove,costsshouldthen decrease
faster than otherwise would be expected.
Capital effectively invested to increase
productivity (as opposedto capital invested
to increase capacity) also contributes to the
cost reductions.This aspectofcapital investment is alsoincluded in the experiencecurve.
Scale effects can occur in many different
ways. For example,volume discountson material purchases,etc., might lower unit cost.
The overall effectofincreasedoutput on cost,
however,is complexand often uncertain.
The experiencecurye is a relatively simple means to predict cost reductions. (Likewise, the learning curve is a mechanism for
predicting labor reductions from learning a
process.)In application the principal problems encounteredare those of defrning cost
elements and defining the units of experience.In shipbuilding, recognition must also
be given to the fact that all costelementsmay
not have the same experiencebase. For example, the experiencecurve effect may be
different in the production of structural assemblies than that encounteredin the production of outfrt units.
The concept of the experiencecurve is
particularly useful in strategic planning. For

I
I
l

SHIPBUILDING
MANAGEMENT
THEORY

example,considerthe implications of the experience curre in terms of market share. According to the experience curve, the market
leader would also have the lowest unit costs.
As a result the market leader would also be
in the strongest position to be the pricing and
technologicalleader in the market.
2.1.3. AdjustmentCosts. Adjustments to labor
cause output to be less than is theoretically
possible, assuming a particular production
relationship and level of resources.This occurs because the adjustment absorbs resources which could be used in producing
output. For example,more capital and labor
are devoted to training when manning is increased.The resulting change in the span of
control, particularly in middle level management, also results in decreasedproductivity.
The resulting decreasedproductivity is generally referred to as frictional or internal
labor adjustment costs. These adjustment
costs are termed internal costs in contrast to
external adjustment costs which are associated with market conditions.
Examples of external labor adjustment
costs are severancepay, recruiting costs,or
short-run increasesin wagesfor a firm hiring
from a relatively non-mobile labor force.
Shipyard workers, in general, are geographically less mobile than other construction
workers.[4] Thus, external adjustment costs
can be expectedto be important in shipbuilding. However,sinceman-hour costsare being
considered,external adjustment costsshould
show up in the skill and experience variables.
The rate of changein the number ofworkers is of particular theoretical and practical
interest.[5] Adjustment costshave the following characteristics: (1) both positive and
negative adjustments to the work force may
result in positive adjustment costs, and (2)
adjustment costs are assumedto be convex.
The frrst characteristic is straightforward for
positive adjustments. A positive adjustment

37

would be expectedto result in increasedcosts.

In practice, negative adjustments may also
increase costs.This phenomenonis partly the
result of labor hoarding. Labor hoarding is
the retention of workers who are not required
for a particular output rate. Highly skilled
workers are often retained during periods of
slack demand because of the high cost of
hiring equally qualifred labor when demand
picks up. Positiveadjustment costsalsooccur
becausecrew integrity may be lost when the
work forceis reduced.In the caseofa general
reduction in yard manning, it has also been
observedthat labor costsrise, partly because
workers slow down in order to preserve their
jobs.This is oneofseveralreasonsfor the cost
tail-up observedat the end of a program.[6]
The second property, convexity, means
that adjustment costsincrease at an increasing rate (increasing marginal adjustment
costs). Convexity refers to the graph of the
adjustment costs (c) versus the time rate of
change in the labor force (dN/dt). This property is necessary in order for adjustment to
take place slowly. For other than increasing
marginal adjustment costs, the shipbuilder
would attempt to adjust employmentinstantaneously. Such a practice does not agree
with observedindustry behavior.
2.2.GraphicalAnalysisof the Shipbuilding
Model
The variables of the model presented above
can be examined graphically in terms of simple one-periodeconomicaverage cost minimization models. Total cost effects of labor
adjustment are then shown in terms of varying the time path of production. Note that
costis merely the other side of the productivity coin. It is more convenient and perhaps
more natural to address costs rather than
productivity.
To ensure that a unique optimum average cost exists, it is su{Iicient that a producer
have a U-shaped average cost curve. In a
short-run model such as this, the U shape

38

SHIPPRODUCTION

results from the interaction ofshort-run frxed

and variable costs. There are some costs
whether or not work is being doneon the ship.
When the productionrate goesup, thesefixed
costsare averagedover more and more units,
so averagecosts(costper unit output) fall. At
the same time as the production rate increases,management costs rise and worker
productivity falls, becausemore workers are
sharing the frxed capital stock.Thus average
costsfall as the output rate is increasedto the
optimum, and rise thereafter.
This simple model illustrates two important points. First, in the absenceof labor
adjustment costs, the optimum manning
level should be attained instantaneouslyand
maintained at a constant level throughout
the constructionperiod.Second,the optimum
output rate determines the minimum averagecostflow and total averagecost.Oncethe
optimum rate is chosen,the optimum construction period is determined. Thus, this
model demonstratesthat constructionschedule decisionscannot in generalbe made independent of decisionsto changerequired total
output.[11
2.2.1. Changesin RequiredOutput. Required
output can changefor severalreasons:initial
estimatesmay have been too low; or the customer may require changesto plans or specifications after construction is underway.
Requiredoutput may alsoincreasebecauseof
unanticipated rework. This may be attributable to the production process itself or to
reduced labor quality. Labor quality will be
treated directly in Section2.2.2.
Generally,initial estimatesare lower than
frnal ship costs.Customer changesalso usually increaserequired output. Thesetwo factors are particularly prevalent in Navy shipbuilding programs.Contract changesmay be
initiated by the customerin order to incorporate newer technology,enhance safety, correct design errors and omissions,or comply

with higher-level directives such as environmental standards.Changesmay also be proposedby the builder either to correct design
deficienciesor to accomplisha given task at
lower cost.A changemay also be a "constructive change,"resulting from someact or omission of the customer, such as customer-furnished material or documentsthat are late,
defective, or otherwise different than originally specifred.Additionally, required output may changebecauseofnatural disasters,
such as hurricanes. Underestimating total
required output may occur becauseof a tendency to "buy in," or bid low on an initial
contract with the expectation of recouping
any lossesthrough follow-on contracts. Underbidding has also been shown to be a natural result of some contract forms becauseof
risk-aversebehavior of the bidders.[7]
One effect ofchanges to required output
is often describedby the term "disruption." In
addition to the identifrable increase in requirements, changesin required output may
have a compounding effect on effrciency over
a number of ship systems, cost centers, or
programs.Planning breaksdown becausethe
changedor added work must be done out of
the usual sequence,and this causesscheduling problems in other parts of the ship or
shipyard.Ifthe addedwork requiresa change
in production method, it may have the same
effect on learning as breaking the production run into a larger number of smaller
production runs. The result could then be
additional setup costs and greater total and
averagecosts.
It is also possiblethat addedwork resulting from customer-directedchanges can be
done concurrently with the basic ship work
and with little increase in capital requirements. This is equivalent to an increase in
the business base for a yard with unused
capacity. The result could be no change or
even an increasein productivity. Curve A in
Figure 2-3 showsthe averagecostfor a hypo-

SHIPBUILDING
MANAGEMENT
THEORY

Fig. 2-3. Shift in the averagecost curve when

requiredoutputchanges.
thetical shipbuilding program as originally
planned. The optimum rate of output is q*,
and the averagecost equals (c/q)r.
Assume that required output is increased.
The disruption causedby changesshifts the
average cost curve up and to the left (curve
B). The optimal output rate falls to q*x, and
the lowest average cost rises to (c/q)2.The
amount of the upward shift in the cost curve
and the shape of the new cost curve B will
dependon many factors,suchas the construction phase in which the change is implemented, the lead time available to plan and
schedule the added work, and the kind of
work entailed.
Many kinds of shifts are possible.A shift
could even increase the optimal rate of production. For example, a change could add
work that could be done independently of
other tasks such that it would pay to add
workers to do the changework concurrently.
Figure 2-3 also shows four choicesthe
shipbuilder could make with respect to the
rate of output. The first is to maintain the
initial rate of output, qx. This requires
lengthening the schedule.Becauseof the disruption causedby the accompanyingschedule change, workers must be added just to
maintain the previous output. A secondoption is to slow the program down to the new
optimum production rate. This requires slipping the scheduleevenmore. Productionrate

39

q', a third option, requires no increase in

manning or overtime. The differencein total
cost between this choice and the original
plan is the sum of the identifiable added
work (called "hardcore cost" in the case of
engineering changes) and what is called
direct disruption. The fourth option is to
maintain the schedule (production rate q').
This requires additional manning and/or
overtime abovethat neededto maintain the
original output. The added total cost incurred
with this option over the previous choiceis
called indirect disruption.Indirect disruption
cost,then, equalsthe additional costs(on top
of hard-core and direct disruption costs)
that result when manning or overtime is
adjusted to acceleratethe program. Note
that the cost and schedule effects depicted
in Figure 2-3 are in essential agreement
with documentedtheoretical and empirical
frndings.[1,8,9,10]
2.2.2. Labor Quality. Shipyards sometimes
run into problems obtaining the number of
skilled and experiencedworkers planned for
a shipbuilding program. Three types of quality variables are considered:(1) worker skill,
(2) worker tenure or experience,and (3) the
amount of overtime hours worked. In terms
of measurablevariables,skill level is usually
representedby the journeyman/apprenticeor
journeyman/total worker ratio. Worker experience is represented by labor turnover.
Turnover is closely related to labor adjustment but is not perfectly correlated with it.
Theoretical and empirical analyses have
shown that the two variables can be treated
as distinct. Overtime hours per worker or the
average number of hours worked per week
are straightforward measuresof overtime.
The solid average cost curve shown in
Figure 2-4 represents costs when output is
varied and labor quality remains constant.In
this example optimum output is q*. The
dashed curves illustrate two hvnothetical

40

SHIPPRODUCTION
B

^g

(c/q) 3
-

9
F
a
u

1 . 2

\aq2

\----!-

I
I
'| /
I
I

G
U
(c/q) 1

RATE OF OUTPUT(q)

Fig. 2-4. Shifts in the averagecost curve with

changesin laborquality.
shifts of the short-run average cost curve
resulting from a declinein labor quality. Both
shifts imply increasingcosts.
Supposethat the base cost curve in Figure 2-4 (solid line) was planned for a given
shipbuilding program, and that, becauselabor quality is lower than planned, curve A or
B is the actual costcurve.IfA were the actual
curve, the shipbuilder's average cost would
rise to (c/q)2,but the optimal schedulewould
remain unchanged. Curve A is particularly
plausible if overtime or a secondshift is used
to maintain output at q*. Overtime hours are
generally consideredless efficient becauseof
fatigue (reducedlabor quality) and they are
more expensive.However, by reducing congestion and the demands on capital equipment, increased overtime may be used to
maintain the optimum output level when labor quality is reduced.In other words, overtime in this caseis efficient.
If B were the actual curve, however,the
optimal output rate would fall to q*x. Minimum cost would rise to (clq)2,but, were the
program kept on schedule, average cost
would be (c/q)s.The cost could therefore be
reduced by slowing down the construction
schedule.This is perhaps more usual, particularly late in the constructioncycle,when
congestionis most evident. In order to make
up for loweredlabor quality, workers must be

added to maintain the same output. This

increases congestion and further increases
cost per unit output. As a result the average
cost curve shifts up and to the left.
This example shows that the optimal
schedule can changebecauseofchanges in labor quality. Thus, the scheduleis an important
variable in controlling the effect of labor
quality on the cost of a shipbuilding program.
2.2.3. Bottlenecks.Late delivery of material,
plans, specifrcations,or subassembliescan
cause"bottlenecks"in the productionprocess.
If any of these necessaryinputs are lacking
when a job is scheduled,the shipbuilder has
to delay the job and other related work, or
alter the schedule.
These bottlenecks are analyzed in terms
ofhow they shift the averagecost curve. As
illustrated in Figure 2-5, the shift takes a
form different from that shownin Figures 2-3
or 2-4. Curve A representsthe basecase,and
qx the optimal output rate. One responseto
a bottleneck is to stop work on the affected
part of the ship, and lay off or shift workers.
This eliminates the costs of reschedulingor
workaround, but it slows down the overall
rate of output. The output rate that can be
sustained without rescheduling or workaround costsis labeled q'in Figure 2-5. Bottlenecks will shift the cost curve up to the

ll
U)

o
o
!!

. .

(c/q) 3
(c/q) z

ff (.rq)r

"

'

RArE OF OUTPUT(q)

Fig. 2-5.Bottlenecksincreasecostsand reducethe

optimal rate of production.

MANAGEMENT
THEORY
SHIPBUILDING

right ofq', but costsare not affectedat output

rates below q'.
Two hypothetical shifts of the cost curve
are illustrated in Figure 2-5. Curve B illustrates a case where the effrciency costs of
working around the bottlenecks are small.
Although costs are higher than in the base
case,it is still less costly to maintain output
at q" rather than cut back to q'. The optimal
costin this caseis (c/q)2.Curve C showsa case
where the costsofworkaround are very large,
and q' itself is the shipbuilder's least-cost
alternative. Ifcurve C rises very sharply, the
shipbuilder has no practical choice but to
reducethe rate ofoutput to q'.
2.2.4. Adjustments to the Production Plan:
Adjustment Costs. In Sections 2.2.1. through
2.2.3.,average costs for a given single time
increment were examined.In this subsection
man-hour costs over the remaining production path, causedby production rate adjustments, are analyzed.
The shipbuilder's short-run production
and manning paths are shown graphically in
Figure 2-6. In this examplethe costeffectsof
the state of completion of the ship when a
changeoccurswill be further developed.The
effect of allowing the completiondate to be a
decisionvariable will also be demonstrated.
The purpose of this exposition is to further
developan intuitive feel for the shipbuilding
processand the alternatives available to the
shipbuilder.
and 2-6(a.2)show
Figures2-6(a),2-6(a.1),
cumulative output, production rate, and
manning level for the situation where the
delivery date after the scheduleis adjusted,
t', or is frxed at the original date,T. In Figures
2-6(b),2-6(b.1),and 2-6(b.2),the shipbuilder
is free to determine the final delivery date.
The original time paths of output, production
rate, and manning are shown as solid lines.
It is assumedthat these were chosenas optimal paths for the original output, Q. At time
t0 required total output is changed(or recog-

41

nized) by an amount, A. The builder must

determine a new optimal manning level over
time. This is doneby determining the optimal
path for the rate of changein manning and,
in the secondexample,an optimal completion
date. In the example, overtime will be held
constant in order to highlight the independent effect of labor adjustments.
The "S curve" shape, typical of systemoriented shipbuilding, is assumed for this
analysis.In the early stages,steelis being cut
and formed. Production rate is limited by
available shop facilities. In the middle portion, production rate frrst increasesas more
intermediate products become available to
the work force. As erection proceeds,work
space becomeslimited and production rate
decreasesat an increasing rate. Toward the
end ofconstruction, the production rate goes
smoothly to zero as the last work packageis
completed.Note that productionrate may be
in one-to-onecorrespondencewith manning
throughout construction. Total man-hours
are equal to the area under the manning
curve times the length of the work day, which
is assumed to be constant for the original
optimum plan.
. Completiondete the sameas originally
planned.
Under the production plan describedby
the dot-dash curve in Figure 2-66), production rate is increasedabovethe original plan
at time to. The combinedeffectsof decreasing
productivity-resulting from greater than
optimal manning and adjustment costs becauseof the more rapid increaseand decrease
in manning-results in a bulge in the manning curve. This bulge is displacedfrom the
correspondingbulge in the production rate
curve. The manning curve is displacedto the
left and upward in the rising portion because
of adjustment costs. Production lags manning in this region.At the apex,the manning
curve is higher than the comparableproduc-

42

SHIPPRODUCTION

Q+A

> z

r l l
l r l

{=
>l

(b)

| | l
I l
ill

F
(I

z
F
l

(r
o-

t 0

t6

t i = l

T|ME(0

Tl[rtr /r\

z_
z
z
z

1\

to

tr=T

T r M E( t )

Fig. 2-6. Production plans beforeand after total output adjustment

tion rate curve because of decreasing marginal productivity of labor.
As production rate is decreased, the manning curve is displaced more and more because of initial adjustment costs. As the manning adjustment decreases nearly lo zeto
just before reaching time tr the manning and
production paths approach the original planned optimums. Again, manning lags the production rate path.

The policy describedby the dashedcurve

maintains production rate for a time at the
originally optimal level. Because the work
space is becoming increasingly restricted,
manning continues to increase even though
production rate is constant. This representation implies decreasingreturns to a fixed
factor where the fixed factor is work space.
Here the rapid decreasein manning at the
end of construction results in a horizontal

SHIPBUILDING
MANAGEMENT
THEORY

displacement of the manning curve due to

increased adjustment costs. This displacement is greater than the correspondingbulge
in the production rate curve.
. Completiondate determinedby the
shipbuilder.
When delivery date is a control variable,
the builder has more discretion in choosing
the optimal production path. Figure 2-6(b)
shows how the shipbuilder might balance
the costs of manpower adjustments and
the daily labor costswhich accruewhether
progressis being made on the ship or not.
Corrosion control, frre guards, and overhead
labor such as quality control or ship management contribute to these fixed man-hour
costs. Because of such costs, the builder
might frnd it economicallyefficient to man
abovethe technically efficient level. The dotdash path illustrates a case where fixed
costs are relatively high. The builder chooses
to incur greater adjustment costsby initially
increasing production rate to a level above
that originally planned.In this way, the costs
of having the ship remain in the yard are
reduced.
The dashed curves show the casewhere
daily costs are lower relative to adjustment
coststhan in the previousexample.This plan
maintains production rate, for a time, at the
original rate and then extends the period of
construction in order to reduce the rate at
which manning is reduced.The costs due to
the decreasedproductivity of higher manning levels late in the constructionperiod are
greater, but adjustment costs are reduced
from the dot-dash plan. Note that the plans
shown in Figure 2-6(b) result in fewer manhours overall than those depicted in Figure
2-66). This overall reduction in cost when
construction time is extended illustrates the
expected result of maintaining production
rate more nearly the same as the original
"optimum" rate.

43

2.3.Summaryof the Shipbuilding

Model
The shipbuilding economicmodel shows the
highly detrimental effects of changing production rate once the program has been
planned.It also showsthe impact of unanticipated shortagesin material or interim products, reduced worker skill level, increased
turnover, or changesto the rate of manning
on production rate. Changes in production
rate affect cost. These effects are well documented and are the result of the breakdown
of the planned production schedule,often to
the point ofloss ofcontrol over the production scheduleby the shipbuilder. Schedule
changesresult in the need for higher manning levels, overtime, or both, later in the
constructionsequence,when interferencebetween trades is more likely to occur.
The following summarizesthe model:
(1) For a given yard capital and for each
time interval, there is a least-cost
(optimum) rate of output, which
yields minimum averagecosts.
(2) Given capital, minimum total cost is
realized by producing at the optimal
production rate in every time interval.
The optimum productionperiodis determined by the optimal time path of
production rate and the required total production at each point in time.
( 3 ) Reduced labor quality (skill level,
worker tenure or experience, the
number of overtime hours worked)
will increaseminimum averagecosts
and may change the optimum production rate. Generally the increases
in manning necessitatedby lower labor quality require a reductionin production rate because of increased
congestion.Overtime, which may be
used to compensate for decreased
skill level or experience,is generally
less efficient becauseof fatigue and
higher hourly wagerates. However,

44

sHrPPRoDUcloN
overtime worked during off-hours (at
night or on weekends)may be an effi.cient way to keep the ship on schedule
becauseit decreasescongestionand
allows more intensive use of capital
facilities.
(4) Bottlenecks(becauseof late or missing plans or equipment) reduceoptimum production rate and increase
minimum average cost. Some bottlenecks are too costly to work around,
and the delays they causemay delay
completionof the ship.
(5) Changesin required output (changes
ordered by the owner or correction of
builders' mistakes) increase minimum average cost and may also reduce optimum production rate. This
occurs partly becausecongestionreduces productivity when manning is
increasedin responseto the increase
in required output.
(6) Adjusting labor in responseto changes
in the optimal production rate results
in internal adjustment costs. These
costs occur becauselabor and management serviceswhich could be producing output are diverted to training or schedulingtasks. This means
that in addition to the added cost
caused by congestion after labor is
adjusted upward, there is a cost of
getting to the new level. This cost
may be incurred when manning is
reduced as well. This occursprimarily becauseoflabor hoarding.

2.4.TheShipbuilding
Model:An Example
Figure 2-7 demonstrates the effect on ship
cost of the variables discussedindividually
earlier in this chapter. This example is representative of man-hour profiles for one ship
in a multiship construction progrzrm.Average
daily man-hours(plannedand actual),average
change in required man-hours, and average

overtime hours are plotted as functions of

time after the start of fabrication for one
ship. This chart shows some of the key elements of ship cost associatedwith traditional
shipbuilding in an environment of changing
requirements.
The graph shows signifrcant changes in
required production beginning shortly after
the start of fabrication (A). Such changes to
requirements are characteristic of "design instability." Design instability can result from
customer-originatedchangesto plans or specffications, the discoveryof errors or major producibility problems in the plans as drawn, or
late or defectivematerials, components,or documents. Given a required increase in output,
managementcan increaseproduction rate (acceleration),or the length ofthe production cycle. Assuming capital is fixed, production rate
can be increasedby a combinationof increased
manning and working overtime. For the ship
construction project depicted in Figure 2-2,
management choseto acceleratethe program
in order to minimize delay. Acceleration combined with critical periods of performancefor
some changed work resulted in schedule inflexibility. This in turn meant that it was nearly impossible to maintain a reasonably constant level ofresource usage.
Although signifrcant increases to total
output had been accumulatingsincepoint A,
the original production rate was maintained
until point B. This meant that a great deal of
work was being deferred. At point B, the
builder acceleratedthe program in order to
get back on schedule.The remainder of the
constructionperiodwas characterizedby "compression" (g1eater than optimal production
rates) as well as some designinstability. Although it may not be readily apparent to the
eye,the actual man-hour increaseaboveplan
is greater than the total change in required
output. This means that productivity was
lower than planned.A heavy costwas associated with the high manning levels used to
attain the increase in production rate dur-

SHIPBUILDING
MANAGEMENT
THEORY

.-.
- O-O

=
=
=
=

45

AVEBAGE ACTUAL MAN,HOURS

AVERAGE PLANNEDMAN HOURS
AVERAGE CHANGE TO TOTAL REOUIREDOUTPUT
AVERAGE OVERTIMEHOURS
.
/\

A
l\

!\/\t\./\,ir

, y r , . JY
. t Vi i , i
\. / i\

a
E
l

iI

\
\

o
u

E
u

TIMEAFTERSTARTFABRICATION

Fig.2-7 . Manpower profiles in an environment of changing requirements

ing the last halfofthe constructionperiod.

This cost is caused primarily by congestion.
In addition to the costs associatedwith
very high manning levels,considerableextra
man-hours were expendedin order to accomplish out-of-sequence work and in moving
individuals and crewson and offthe job. Note
that throughout the period when required output was being changed, manning was fluctuating sharply. These fluctuations are typical
of a program where design instability is significant. The rapid changes in resource level
result in signifrcant adjustment costs.
A secondpart of adjustment costswhich
is not as obvious from the graph is waiting
time. The graph shows rapid shifts in manning. It does not show the instances when
workers were waiting for accessor for parts
because of the breakdown in planning and

control associated with design instability.

Nevertheless,the hours lost to waiting are a
significant part of adjustment costs.
Overtime was concentrated in the latter
stages of construction. This is fairly typical.
This period is crucial in the shipbuilding constructioncycle.Machinery and other systems
are being interfaced, trials conducted, and
final quality assurance milestones completed. The builder's choices become even
more limited, becausethere is only so much
physical spaceavailableaboardthe ship. It is
generally more effrcient to reduce congestion
and accommodateschedulechangesby working overtime rather than increasing manning. This has been confirmed by empirical
analyses of major programs, which showed
that the partial effect of increased overtime
is a reduction in total man-hours.

3. GroupTechnology
The economic model of shipbuilding developed in the previous section describesimportant cost drivers and the mechanics of cost
measurement in shipbuilding. Knowing the
major sourcesof costs and how to measure
them, however, is not necessarily the same

as knowing how to control them. In this section the basis for controlling costs through
improved management and production organization is presented.The meansof organizing the work is the application of group
technology.

44

SHIPPRODUCTION

overtime worked during off-hours (at

night or on weekends)may be an efficient way to keep the ship on schedule
becauseit decreasescongestionand
allows more intensive use of capital
facilities.
(4) Bottlenecks (becauseof late or missing plans or equipment) reduceoptimum production rate and increase
minimum averagecost. Somebottlenecks are too costly to work around,
and the delays they causemay delay
completionof the ship.
(5) Changesin required output (changes
ordered by the owner or correction of
builders' mistakes) increase minimum average cost and may also reduce optimum production rate. This
occurs partly because congestion reduces productivity when manning is
increasedin responseto the increase
in required output.
(6) Adjusting labor in responseto changes
in the optimal production rate results
in internal adjustment costs. These
costs occur becauselabor and management serviceswhich could be producing output are diverted to training or schedulingtasks. This means
that in addition to the added cost
caused by congestion after labor is
adjusted upward, there is a cost of
getting to the new level. This cost
may be incurred when manning is
reduced as well. This occursprimarily becauseoflabor hoarding.
2.4. The Shipbuilding
Model:An Example
Figure 2-7 demonstrates the effect on ship
cost of the variables discussedindividually
earlier in this chapter. This example is representative of man-hour profiles for one ship
in a multiship construction program. Average
daily man-hours(plannedand actual),average
change in required man-hours, and average

overtime hours are plotted as functions of

time after the start of fabrication for one
ship. This chart shows some of the key elements of ship cost associatedwith traditional
shipbuilding in an environment of changing
requirements.
The graph shows significant changes in
required production beginning shortly after
the start of fabrication (A). Such changes to
requirements are characteristic of "design instability." Design instability can result from
customer-originatedchangesto plans or specifications, the discoveryof errors or major producibility problems in the plans as drawn, or
Iate or defectivematerials, components,or documents. Given a required increase in output,
managementcan increaseproduction rate (acceleration),or the length ofthe production cycle. Assuming capital is fixed, production rate
can be increasedby a combinationof increased
manning and working overtime. For the ship
construction project depicted in Figure 2-2,
management choseto acceleratethe program
in order to minimize delay. Acceleration combined with critical periods of performancefor
some changed work resulted in schedule inflexibility. This in turn meant that it was nearIy impossible to maintain a reasonably constant level ofresource usage.
Although significant increases to total
output had been accumulating since point A,
the original production rate was maintained
until point B. This meant that a great deal of
work was being deferred. At point B, the
builder acceleratedthe program in order to
get back on schedule.The remainder of the
constructionperiodwas characterizedby "compression" (greater than optimal production
rates) as well as somedesign instability. Although it may not be readily apparent to the
eye,the actual man-hour increaseaboveplan
is greater than the total change in required
output. This means that productivity was
lower than planned.A heary costwas associated with the high manning levels used to
attain the increase in production rate dur-

SHIPBUILDING
MANAGEMENT
THEORY

= AVERAGE ACTUAL MAN.HOURS

= AVERAGE PLANNED MAN,HOURS
= AVERAGE CHANGE TO TOTAL REOUIREDOUTPUT
= AVERAGE OVERTIMEHOURS
.

/\

; t
\,/\/\

I\ . .
. t\ /\ /\

/ / ' y r ' . J ' l Y V\. /\i \i

a
E
l

\
I
\
\I

2
!
o
U

E
u

TIMEAFTERSTAFT FABRICATION

Fig. 2-7. Manpower profiles in an environment of changing requirements.

ing the last halfofthe constructionperiod.

This cost is caused primarily by congestion.
In addition to the costs associatedwith
very high manning levels, considerableextra
man-hours were expendedin order to accomplish out-of-sequencework and in moving
individuals and crewson and offthe job. Note
that throughout the period when required output was being changed, manning was fluctuating sharply. These fluctuations are typical
of a program where design instability is significant. The rapid changes in resource level
result in significant adjustment costs.
A secondpart of adjustment costswhich
is not as obvious from the graph is waiting
time. The graph shows rapid shifts in manning. It does not show the instances when
workers were waiting for accessor for parts
because of the breakdown in planning and

control associated with design instability.

Nevertheless, the hours lost to waiting are a
significant part of adjustment costs.
Overtime was concentrated in the latter
stages of construction. This is fairly typical.
This period is crucial in the shipbuilding construction cycle. Machinery and other systems
are being interfaced, trials conducted, and
final quality assurance milestones completed. The builder's choices become even
more limited, becausethere is only so much
physical spaceavailable aboard the ship. It is
generally more effrcient to reduce congestion
and accommodateschedulechangesby working overtime rather than increasing manning. This has been confirmed by empirical
analyses of major programs, which showed
that the partial effect of increased overtime
is a reduction in total man-hours.

3. GroupTechnology
The economicmodel of shipbuilding developed in the previous section describesimportant cost drivers and the mechanics of cost
measurement in shipbuilding. Knowing the
major sourcesof costs and how to measure
them, however, is not necessarily the same

as knowing how to control them. In this section the basis for controlling costs through
improved management and production organization is presented.The meansof organizing the work is the application of group
technology.

46

SHIPPRODUCTION

3.1.GroupTechnology,an Overview
Group technology (GT), also called family
manufacture (FM), began as an outgrowth of
an attempt to developa more efficient system
of classification and coding for use in the
managementof industrial processes.As with
any scientific endeavor,a classifrcationsystem is essentialto the organizationofdata in
order to facilitate analysis and synthesis,the
formulation of hypotheses,experimentation,
deduction, and finally generalization to a
practical application. However, the classification systemis only a techniqueor tool ofthe
scientist. Likewise, group technology is an
innovation in the broader freld of management of manufacturing processes,not just a
technique for keeping track of material,
parts, subassemblies,etc.
Group technology is also called cellular
manufacture.The word "cell" conveysimportant information essential to understanding
what group technologyis and how it can be
applied to shipbuilding. In the machining
industry, where GT has been most extensively applied, a cell consistsof somenumber
of grouped machines and the peoplewho operate them. Generally the operators are
cross-trainedto operateall the machinesin a
given cell. Instead ofpiece parts being scheduled sequentially, the cell is scheduledand
loadedwith parts which are classifiedaccording to shape,material, size, etc., into a "family." The cell is then effectivelyoperatedas a
single machine.A part is essentially worked
continuously from the time it is loaded into
the cell until it emergesas a completedinterim product. As a consequence,the time
spent in processand the inventory level of
work in processcan be a fraction of what is
normal for a traditional manufacturing layout and production control system.
In establishing families, an effort is also
made to reducethe number of different individual parts. This may involve marketing as
well as design personnel,sincethe salesman
would generally like to be able to frll every

order no matter how low the level of demand

or how unprofrtable a given product line
might be. Particularly in the caseof custommade products,this may involve "selling" the
customer on the overall cost savings of production-kindly designseventhough they may
involve somedegradationin operationalsuitability.
In addition to minimizing the number of
individual parts and the number of operations, batch sizes are also reduced to the
minimum. A major objective of GT is to reducethe inventory of work in processto only
what is needed.This may sound like setup
and teardown times will be excessiveand
that the risk of running out of some assemblies will be high. On the contrary, setup
times within a cell are minimized, because
cells are designedso that machinesneedonly
be adjustedrather than completelyreset,and
much tighter control of stock and work in
processis maintained. The savings in stock
and interim product inventory, often the single greatest componentofvariable cost, can
more than compensatefor added machine
adjustment time.
An additional benefit is that job satisfaction tends to be high. Workers are crosstrained in all aspects of the cell, including
inspection.When a cell is loaded,the workers
actually see the finished products emerge
within a short period of time. Additionally,
schedule control can be maintained very
closelyso that there is very little if any unanticipated idle time. The result is higher morale and higher productivity.
Figure 2-8 is a comparisonof a machine
shoplaid out conventionallyand in cells.The
conventionallayout is basedon grouping like
machines,while the group technologylayout
is organizedinto cells that are structured to
produce similar products. The completemachining of an individual part in the conventional layout requires considerablehandling
ofthe part, possibly back and forth between
lathe, mill, grinder, and other machine type

SHIPBUILDING
MANAGEMENT
THEORY

areas. Thus the benefit of repeating many

similar operations is balanced by the additional material handling, storage,inventory
control, and work in processcosts.For example, assuming a batch size of 100 and 4 minutes for each process,the conventional six
processsteps shown in Figure 2-8(a)require
2,400 minutes for each batch (4 x 6 x 100 =
2,400 minutes). Using the group technology process,Figure 2-4(b), the total batch inprocesstime is 420 minutes. The first part
requires 4 + 4 + 4 + 4 + 4 + 4 =24 minutes.
The remaining parts require 4 x 99 = 396
minutes, for a total of 420 minutes.
The group technology layout results in
the more or less continuous manufacture of
an individual part, within a single cell, by
the same group of workers. Not only are
throughput times greatly reduced for the
layout shown in Figure 2-8(b), but the machine grouping enhancesthe cohesivenessof
the workers in the cell. Expanding this approach to an entire manufacturing operation

47

producesthe shipbuilding system to be described in this text.

3.2. GroupTechnologyDefined
The purposeof addressingGT in this book is
to help the reader better understand shipbuilding and how productivity can be improved in the shipbuilding industry. While
the treatment will be rigorous with regard to
the description of group technology,the emphasis will be on those aspectsof GT which
are applicable to and interpretable in the
context of shipbuilding.
Group technologymay be defined as:
The logical arrangement and sequenceof
all facets of company operation in order
to bring the benefits of mass production
to high variety, mixed quantity production.[11]
This generaldefinition emphasizesa systems
approach to management, as opposedto a

A) CONVENTTONAL
PROCESS
METHOD

B) GROUPPROCESS
METHOD
KEY: L=LATHE
M = MILLING
MACHINE
C = C U T T I N GM A C H I N E

D=DRILLPRESS
G =GRINDER

Fig. 2-8. Comparisonofconventional and group processbatch flow.

48

SHIPPRODUCTION

techniquefor organizing a limited aspectof a

manufacturing process independent of the
total system. As such, it supports a crucial
precept put forth by Mitrofanov in his pioneering work on the subject.[12J
Mitrofanov stated that the group technological processis a variant of ". . . the systematization and generalization of the experience
of the entire machine building industry . . ."
This book deals with the total management
of a shipbuilding or repair activity. The group
is the basicproductionunit of zone-or product-oriented ship constr-uction.However, a
crucially important characteristicof productoriented ship manufacture and construction,
which sets it apart from traditional shipbuilding, is the total integration ofall departments in the company.As cited by Ranson,
"There
are only two peoplein a manufacturing organization,the man who sells and the
man who makes; everyone else runs their
messages."[11]The group is the basic production element.Everyoneand everything in
the yard should be organizedto support the
group.
A seconddefinition of group technology
provides further insight.[ 13J
GT is a technique for manufacturing
small to medium lot sizebatchesof parts
of similar process,of somewhat dissimilar materials, geometry and size, which
are producedin a committed small cell of
machines which have been grouped together physically,specifrcallytooled,and
scheduledas a unit.
It is useful to analyze the essential elements of this definition:
. Small to medium lot size batchesMany of the interim productsof shipbuilding are one of a kind or only a
very few like parts. Group technology
is not applicableto lot sizeswhich can
be e{ficiently producedon an assembly

line. Group technologyis a means of

realizing certain benefits of mass production (i.e., relative permanencyof
location and function, moving work to
the worker, balancedproduct flow, etc.)
for essentially small batch interim
products.It is not massproduction.
Perhapsone of the major errors made
by innovative shipbuilders in the 1960s
and 1970swas attempting to adapt mass
production assemblyline techniquesto
what is a small-batchprocess.The result
was yards which dependedon series
production with large, unrealistic
throughputs in order to have any
chancefor efficient productivity. In
essence,these yards dependedon
government manipulation of the
market for productivity rather than
adapting the production processto
the existing and anticipated market.
When the world shipbuilding market
collapsedin the 1970s,many of the
most modern assemblyline type yards
were the first to go bankrupt or be
nationalized.
. Similar process-This implies categorizing interim productsby "problem
area."A "problem area" is a specific
type of work, involving the use of
similar prodrictiontechniques,tools,
and worker skills. For example,manufacturing curved pipe piecesand
straight pipe pieces are two different
problem areas.So too are flat panel
and curved panel assembly.This will
be discussedin greater detail in the
next chapter.
o Somewhatdissimilar materials,
geometry,and size--That the same
problem area doesnot imply identical
shape,material, size, etc., is a crucial
concept.In a GT product-orientedoperation, the installation of curved pipe
and curved ventilation ducts may be
the same problem area and be accom-

S H I P B U I L D I NMGA N A G E M E NTTH E O R Y

plished by the same crew. However,

in traditional shipbuilding, these two
operations,even if physically adjacent,
would be accomplishedat different
times by different work crews.
. Processedin a committed small cell
of machineswhich haue beengrouped
togetherp hysically -In machining
industries, where GT has beenprimarily
applied, this is self-explanatory.In
shipbuilding, the cell often consistsof
a crew of workers whosemost sophisticated pieceof equipment is a spanner
wrench or a simple arc welder. The
essential conceptimplied by this phrase
is parallelism. A cell in a machining
industry consistsof a group of machines
which completeall processesnecessary
to completepieceparts in a particular
family, regardlessof sequenceor
machine utilization. Similarly, in shipbuilding, a cell or group is responsible
for completing all aspectsof a given
block or unit, including piping,
ducting, painting, etc., regardless
of overlapping functional systems.
Consequently,subassembliescan be
completedsimultaneously,rather than
systemsbeing completedsequentially.
.Specifi.callytooled-In machining, this
implies the use of equipment such as
turret lathes, where tools need only be
adjusted, never removed.Shipbuilding
has the added advantagethat the
operator may be thought of as an
essentialpart of the machine in some
cases.Adjustment of tools may imply
only moving the operator'sanatomy.
. Scheduledas o group-This means
that the cell or unit is scheduledas a
single machine. In shipbuilding this
is comparableto commencingwork on
a work packageassignedto a given
group (i.e., a subassemblyor unit)
only when all resourcesare on hand.
This has important implications con-

49

cerning management,engineering,
and material control. It means that
the former must be responsiveto production control in a way not normally
expectedin conventionalshipbuilding.
3.3. Classification
and Coding
Group technology is not synonymous with
classification and coding. However, classification of the elementsof production is perhaps
the frrst step in the successfulimplementation of GT. The defrnition of group technology
presentedby Ransonis valuable becauseofits
generality and applicability to all aspects of
companyoperation.Sotoo must a classification
system be based on the assumption that all
elements of the company are subject to classification and coding (seeFigure 2-9).
3.3.1. C/assification.The Webster defrnition
of classificationis "Systematic arrangement
in groups or categoriesaccording to established criteria."[14] This definition is straightforward and suitable for the purpose of this
discussion. A key word in this defrnition
which perhaps requires some elaboration is
"criteria." The hierarchical
classifrcationsystem used in botany or biology is familiar to
every scientist. This system ofclassification
is called the Linnean hierarchical taxonomy,
after the Swedish botanist Karl von Linn6.
Organisms are classed into kingdom, phylum, subphylum, class,order, family, genus,
species,and variety according to mutually
exclusiue and permanent characteristics.
Edward Brisch, a mechanical engineer
and designer,adoptedthe decisiontree type
hierarchical classificationsystem derived by
von Linn6, but addedtwo principles or criteria. In addition to mutual exclusivity and
permanent characteristics,he required that
the systembe all-embraclngand basedon the
user'spoint of uiew.llS) Brisch's taxonomy
has becomeone of the standards for industrial application.

50

SHIPPRODUCTION

Fig. 2-9.Companyfunctionsaffectedby grouptechnology.

One classification system, the Productoriented Work Breakdown Structure, is introduced in the last section of this chapter
and will be treated in detail in Chapter III. In
this subsection, general application of the
abovefour principles is addressed.
. User'suiewpoint- A classification
system must be responsiveto the
objectiveofthe next higher system of
which it is a part. A primary application of a classificationsystem is to
define families of parts in order to
facilitate engineeringor manufacturing
decisions.Shapemight therefore be
an important attribute. However,
shapeis important only as it relates
to problem area or work process.For
example,two nearly identically shaped
parts might present two completely
different problem areas becauseof
differencesin chemistry. Extrusion or

casting are two very different methods

of producingidentically shapedparts
of different material. A familiar example is the classificationofplaying cards.
A bridge player might categorizethe population in samplesof 13 cards arranged
first by suit and secondby value with
the ace at the high value end. This
classificationsystem would be of little
value to a poker player. For poker the
population would be broken into samples offive or sevencards ranked
according to certain combinationsof
cards, such as flush, straight, pair, etc.
Within these classesranking is determined by card value, with the acehigh.
. Scopeof the classification - A classification system must be able to accommodate the product, the means of
production,and the controls over production. In line with the secondprinciple, that the systembe all-embracing,

SHIPBUILDING
MANAGEMENT
THEORY

the scopemust be defined basedon

the population as well as the objective
of the classificationsystem.This
means that numbers of categories
must be sufficient to accommodate
all of the characteristicswhich are of
interest to the particular business.It
also means that information should
be stored and be retrievable according
to the specificuser. For example,a
production engineer,a salesman,and
a purchasing agent are interested in
different characteristics, even though
those characteristicsmight identify
the same item. The supplemental
characteristicsrequired by different
users must therefore be stored in the
data base so that each user receivesor
inputs only the required information.
. Mutually exclusiue- Any given set of
identical characteristicsmust describe
a unique object.This may seemtoo obvious to receivemore than passing interest. In a companydealing with
thousands of parts, however,it may
be of more than academicinterest. In
the absenceofa high level ofdiscipline it is not uncommonto frnd
the same part describedby more than
one set ofcode characters.
o Permanentcharacteristics- Permanenceis. ofcourse.relative to the
user's point of view. In general, a permanent characteristicis one that describeswhat an objectis, not how or
where it is used. Where a pipe pieceis
used is important to an assembler,
but there are ways of identifying
where it is used which are distinct
from the classifrcationsystem.Material, size, and degreeofbend, on the
other hand, are permanent, easily
identified attributes which affect design and manufacture decisions.Figure 2-10 presentsan example of such

51

a classificationscheme,as applied by
the BoeingAirplane Companyto the
design and manufacture of commercial aircraft.[15]
3.3.2. Coding. Classifrcationand coding are
often used as if they were one word. They are
not. The codeis the vehicleby which a classification systemis made operational.A classifrcation systembasedon ke5'words,suchas is
used for somelibrary searches,is feasible.In
general, however, a code consisting of numbers, letter characters, or symbols is much
more effective.A sequential numbering system should not be confusedwith coding. A
code must not only identify an object, but
must be basedon permanent,mutually exclusive attributes accordingto someuser objective. For purposesofretrieval and easeofuse,
it is desirablethat the codereflect yes or no
questions.An objecteither has a certain characteristic or it doesn't. This does not mean
that codesmust be binary. A particular hierarchical category such as size may be described by the digits zero through nine to
describeten different sizes.
Codes may also be mnemonic. Letters
correspondingto the first letter ofkeywords
may be used to represent certain attributes.
For example,oneshipyardusesIL10 to identify a ten-footinclined ladder. Further examples will be given in Chapters III and VII,
where codingwill be discussedin more detail.
3.4.GroupTechnology
and the Shipbuilding
Model
It was shown in Section1 that a major source
of low productivity (high costs)in shipbuilding is unanticipated changes in production
rate. This takes many forms (increasedmanning, overtime, rapid fluctuations in manning, idleness,etc.)and can be traced to more
than one direct cause(poor cost estimation,
designinstability, bottleneck delaysbecause
of missing material or plans, high turnover,

52

SHIPPRODUCTION

etc.).The effectofsuch cost drivers is exacerbated by accounting and production control

procedures which are oriented to functional
ship systems.Functional system orientation
contributes to a highly sequence-dependent
production operation. For example, a frre
main system spans nearly the entire ship.
The labor and material required to fabricate
and install a fire main system would not be
diflicult to estimate if the system were laid
out in an open field. However, the system
must interface both in time and space with
the remainder of the ship. This creates a
massive accounting and control problem. It
also creates a practical problem having to do
with human nature.
If a work order is issued for the entire
system,as is commonpracticein manyyards,
it must remain open for nearly the entire
construction cycle.This work order then becomes a prime candidate for "creative progressreporting" by various shops.Sometimes
the man-hours and material budgeted for
another work order are used up before the
work is completed. A work order having a
large budget and spanning a longtime frame,

such as the fire main work order, then becomes a logical source of borrowed budget.
Shop foremen simply charge resources expendedfor onejob to the job with the remaining budget. It is something like a pyramid
club. The final accounting can be deferred as
long as somework orders are still open.The
shop foremen,of course,hope to bring budgets into line through various effrcienciesbefore the final accounting.Even if this is done,
it is impossible to properly account for expendedcostsof somesectionsof the ship. As
a result, estimating future jobs or even
ships in the same seriesis very inexact. Additionally, areas where productivity might
be improved may be disguised.Management
doesn't know that such areas are contributing to costs in excessof what was planned.
Consequently, no effort may be made to correct the situation.
Another sourceof low productivity is idleness.A major sourceof idleness is a breakdown in resource scheduling and control.
Workers report to a job and find someone
from another trade in their way becauseof a
lack of schedulecoordination. The workers
sz!

/!'rccs-s
-^sls eNo

{tu'
*at"r,";

--{%c)

1:onlasLE
ME|{r
eo\.IP

{ PARTg

't)-3600

"'

5t

eE^

Spa^
--AlD
-

ds

AlRPt l{EOEStGr
ANOMAXUFrcTUR}{O

tsit

Fig. 2-10.The BoeingCompanygroup technologyclassificationsystem,basedon the E. G. Brisch concept.

(Note: BUCCS refers to Boeing unifred classificationand codingsystem.)

MANAGEMENT
THEORY
SHIPBUILDING

wait. Or workers need someparl to complete

a task. One goes to find the part. The rest
wait. Drawings are not available as needed
or a change to the drawings is incomplete.
The workers wait. A critical previous task is
not completed, or some owner-provided
equipment doesnot arrive on time. The workers wait. The list goeson and on.
So, too, the list ofapparent causesoflow
productivity goeson and on. What is important is that these are inherent in the management system, not in the performance of
the individual workers. Borrowing man-hours
and idlenessare not an indictment of the work
force as sneaky and lazy people. Rather, the
indictment is against the management system. Group technoloryis onepossibleapproach
to improving the managementsystem.
If the ship is subdividedinto smaller geographical areas (zones), then the master
schedulenever has to be concernedwith the
many interfaces between the fire main system and all the other systems on the ship.
Thesezonesrepresentthe output ofa GT cell
or group. The cell is then loaded with all of
the resourcesnecessaryfor the completionof
somemanageablework package.Sucha work
package may include a portion of the frre
main system, along with ventilation ducts,
electricconduit,etc.If problemswith the pian
and scheduledevelop,such as resourcesnot
becomingavailable as planned,management
shouldbe able to recognizeand respondto the
situation without adverselyaffectingthe rest
of the project.For example,the delayedwork
packagemay be deferred and another packageloaded.Most of the idlenessproblemthen
disappears.

53

An important effect of zone construction

is to straighten out the S curve described in
Section 2. Hull construction and outfrtting
are accomplishedconcurrently.This allows a
steeper progress curve in the beginning and
frnal portions of the building cycle, and a
reduction in congestionand idleness.The result is a smoother,more constant production
rate throughout, and a minimization of adjustment costs.
The group, which may consistof between
two and a dozenindividuals under oneleader,
takes care of interface problems.Sequencing
problems are greatly simplifred and are resolved at the lowest level. This greatly reducesfluctuations in production rate and manning. Additionally, if the budgetedresources
are not suffrcientto completethe work package, it is immediately apparent, and corrective action can be taken.
There are, ofcourse,costsassociatedwith
increased productivity through the application of group technology.However, the cost
savings should greatly outweigh the implementation costs. To realize the benefits, it is
necessaryto increase planning and control.
All of the resourcesrequired for a given work
packagemust be deliveredto the work site at
the right time. This entails more extensive
and detailed engineering and thus a larger
engineering force. A detailed classifrcation
and codingsystemis alsorequired in order to
realize the accuracyand cost savings made
possibleby the application of computers to
inventory control, work scheduling and control, design,and manufacture. The next section will address some of the mechanics of
adapting group technologyto shipbuilding.

4. WorkBreakdown
Structures
Any management approach must specify
what is to be done, where it is to be done,
when it is to be done,and what resourcesare
to be applied. This specification generally
takes the form ofdivision ofthe total process

into componentparts. The system by which

these componentsare subdividedin order to
control the processis called a work breakdown structure (WBS). A work breakdown
structure is a classification svstem. Work

54

SHIPPRoDUcTIoN

breakdown structures commonly used in

shipbuilding are either systems-or productoriented. Both will be initially describedin
this section.The product-orientedwork breakdown structure will be addressedin detail in
the remaining chapters.
4.1. Systems-Oriented
Work Breakdown
Structure
Systems-oriented work breakdown structures are useful for initial estimates and the
early design stage.They are not appropriate
for planning, scheduling, and executing a
zone-oriented manufacturing process. By
their very nature they institutionalize the
problems addressedin the previous section
with work packages which are too large for
effectivecontrol of material, man-hours, and
schedules.
The systems-orientedstructure used by
the U.S. Navy will be describedas an example
of such a system. The Navy Ship Work Breakdown Structure (SWBS) is used throughout
". . . the
entire ship life cyclefrom early design
and cost studiesthrough production and subsequent layup, including cost, weight, specifrcations, system function and effectiveness,
design, production, and maintenance."[16]All
classification groups are defined by a threedigit numeric code according to functional
system. There are ten major groups,the last
two of which are used primarily for cost estimating and progressreporting. The ten major groups are:
000. General Guidanceand Administration
100. Hull Structure
200. Propulsion Plant
300. Electric Plant
400. Command and Surveillance
500. Auxiliary Systems
600. Outfrt and Furnishings
700. Armament
800. Integration/Engineering
900. Ship Assembly and Support Services

Each major group (000, 100, 200, 800,

etc.) is broken down into hierarchical subdivisions called subgroupsand elements.Subgroups are those three-digit numbers ending
in a singlezero.All otherthree-digit numbers
are calledelements.An exampleofthis structure is illustrated in Figure 2-11:
Briefly, the structure may be interpreted
as follows:
o Groups 100 through 700 equal hardware cost and weight condition A
(light ship without margin).
. Groups 100 through 200 plus 800 and
900 equal ship constructioncost.
. Groups 000 plus 100 through 200 plus
800 and 900 equal total ship cost for
conditionA.
4.2. Product-Oriented
Work Breakdown
Structure(PWBS)
The U.S. Navy has used a classificationsystem, part of which is incorporated in the
SWBS described above, for over 50 years.
This classification system was tailored to
past design practice, where drawings and
interim products were identifred by system.
Such a schemeis appropriate for estimating
and the early designstage.However,the way
that ships are actually producedis by procuring or fabricating parts and joining them to
createsubassemblies.In turn, these are combined through several manufacturing levels
to produce increasingly larger subassemblies. Thus, the ideal way to subdivide ship
constructionwork is to focuson neededparts
and subassemblies(interim products).
Aclassification schemeto subdividework
in accordancewith an interim product view
is a product-oriented work breakdown structure (PWBS).[17] Parts and subassemblies
are grouped by common permanent characteristics and classifredby both design and
manufacturing attributes. The classification
system typically specifiesparameters, such
as form, dimensions,tolerances,material. and

SHIPBUILDING
MANAGEMENT
THEORY

types and complexity of machinery operations. The codesused to processdata according to this classifrcation scheme must be
applicable to previously manufactured as
well as current parts for the purpose of retrieving processstandards. This sectionand
Chapter III will concentrateon the classifrcation system (PWBS) and its relationship to
group technology.Coding systems and specific codesin use in the United States will be
addressedin Chapter VII.
Classifrcationby product aspectsrelates
a part or subassemblyto a systemor zone of
a ship design and also to work processesby
problem area and.by work sfage.Thus, product families are determined both by design
and manufacturing attributes. This concept,
combined with a greater degreeof interaction
between design and production engineers,
has proven to be a powerful means for improving productivity.
4.2.1. WorkPackageClassification PWBS first
divides the shipbuilding processinto three
basic types of work: hull construction,outfitting, and.painting,becauseeachimposesmanufacturing problems that are inherently different from the others (see Figure 2-12).
Zone-orientedproduction,i.e., the Hull Block
Construction Method (HBCM), is already being applied for hull construction by most
shipyards. The same logic is not employed
everywhere for outfrtting by zones,which is
more complex and diffrcult to undertake.

55

Thesethree types of work are further subdivided into fabrication and,assemblyclassifrcations which are normally associatedonly
with hull construction and outfitting. Within
the painting classification, fabrication applies to the manufacture or preparation of
paint, and.assemblymeans its application.
These assembly subdivisions are naturally
linked to zones and are the basis for zone
dominancein the managementcycle.
Secondly,PWBS classifiesinterim products in accordancewith their needs for resources,i.e., material, manpower, facilities,
and expenses.For example,resourcesare classified and allocated in accordancewith common parameters to different structural panels, regardlessof their location in the ship.
Different outfit units are treated the same
way. Definitions of theproduct resourcesarel
. Material, to be used for production,
either direct or indirect, e.g.,steel
plate, machinery, cable,oil, etc.
. Manpower, to be chargedto production,
either direct or indirect, e.g.,welder,
gas cutter, fitter, finisher, rigger,
material arranger, transporter, etc.
. Facilities, to be applied to production,
either direct or indirect, e.g.,buildings,
docks,machinery, equipment, tools, etc.
In order to optimize productivity, a ship must
be constructedin accordancewith a carefully
establishedplan that provides for processes

(Group)
100- HullStructure
- Structural
(Element)
101- GeneralArrangement
Drawings
(Subgroup)
110 ShellandSupporting
Structure
(Element)
111- ShellPlating,
Surface
ShipandSubmarine
Pressure
Hull
(Element)
112- ShellPlating,
Submarine
Non-Pressure
Hull
(Subgroup)
120- HullStructural
Bulkheads
(Element)
121- Longitudinal
Structural
Bulkheads
122 - Transverse
Bulkheads
Structural
Fig.2-11.Example0f U.S.Navyhierarchicalsubdivisions.

56

SHIPPRODUCTION

for manufacturing parts and subassemblies

leading to outfrt units and structural blocks
within time frames that can be coordinated,
and simultaneoususe of each processfor the
requirements of different systems, even in
different ships.
The third classification,by the fourproduct aspects,addressesthese needsbecause
it contains essentialsneededfor control of
production processes.Two product aspects,
system and zone, are means for dividing a
ship design into planned manageable parcels.Each, for example,can apply to a number of parts or to one specificassembly.Each
of the latter is usually addressedby a separate work package. The other two product
aspects,problem q,rea and.stage, are means
for dividing the work process from material
procurement to complete ship delivery. The
product aspectsare:
o System -A structural function or an
operational function of a product, e.g.,
longitudinal bulkhead, transverse
bulkhead, fire main system,mooring
system,fuel oil servicesystem,lighting system, etc.

. Zone-An objective of production

which is any geographicaldivision of
a product, e.g.,cargohold, superstructure, engine room, etc., and their subdivisions or combinations,e.g.,a structural block or outfit unit, a subassembly of either, and ultimately a part or
component.
. Problem area-Adivision of the production processinto similar types of
work problemswhich can be:
-by featur, .g.,curved vs. flat
blocks,steel vs. aluminum stmcture,
small-diameter vs. large-diameter
pipe,pipe material, etc.
-by quantity, e.g.,job-by-jobvs.
flow lane, volume of onblockoutfitting for machinery spacevs. volume
ofon-block outfitting for other than
machinery space,etc.
-by quality,e.g., grade of workers
required, grade of facilities
required, etc.
-by kind of work, e.g.,marking,
cutting, bending,welding, blasting,
bolting, painting, testing, cleaning,
etc. and

PRODUCT
WORK
BREAKDOWN
STRUCTURE
(PWBS)

PIPE
PIECE
FAMILY
MANUFACTURING
(PPFM)

Fig. 2-12. Product work breakdown structure components.

SHIPBUILDING
MANAGEMENT
THEORY

-by anything else that createsa

manifestly different work problem.
. Stage-A division of the production
processby sequences,.9.,substepsof
fabrication, subassembly,assembly,
erection,outfitting on-unit, outfitting
on-block,and outfitting on-board.
The classifrcationsystem and categoriesdescribedin the foregoingsubsectionare illustrated in Figure 2-13.
4.2.2. Work PackageProductivityValueAnalysis.
When an interim product is identifred by
product aspects,it is necessaryto evaluateits
efficiency as a work package which can be
expressedby the formula:
PV = f (T,N,Q)
where:
PV - productivity value, i.e., the productive efficiency of a work package
=
T
time allowed for its accomplishment, i.e., working time

57

= number of units of resources,

particularly componentsin the
material list and man-hours
allocated
= quality of work circumstance,e.g.,
downhand vs. overhead,high vs.
low, etc., and also quality specified
for the interim product.

T, N, and Q are interdependentand as shown

in Figure 2-14,they impact differently on PV.
As they cannot be evaluated separately,it is
useful to symbolizePV as a triangle having
sides that represent T, N, and Q. Optimum
PV is then represented by an equilateral triangle (seeFigure 2-15).In other words, PV is
optimized when the influencesof T, N, and Q
are balanced.
The function f (T, N, Q) must be determined empirically by each shipyard and separately for each classification of the production processby problem area. In addition,
each such determination must consider the
immediate preceding and following work
stages. For example, Q includes consideraPRODUCT
ASPECTSAXIS

o
x
PRODUCIRESOURCES
AXIS
(x1)
MATERTAL
MANPowER(x2)

@
U
E
l
@
U
c

(x3)
FACtLtTtES

expEHses(xa)

o
E
o

Fig. 2-13. Elements of the three-dimensionalPWBS matrix.

SYSTEM
(r1)

ZONE
lY2\

AREA
(y3)

(v4)

58

SHIPPRODUCTION

,,1
ln
\| \
r
l/
T

1 , ,
*1,/
o
Fig.2-l4.Impacts of time, units of resources,and
quality of work circumstance on productivity
value.

z.
z
z.

I
(L

DETAIL
DESIGN
SPECIFICATION
OFINTERIM
PRODUCTS
BY
ZONE/AREA/STAGE

Fig. 2-15. Optimum PV requires balancedinfluencesof T, N, and Q.

Fig.2-16.Iterativedevelopment
ofworkpackages.

tion of the quality specified for an interim

product. If its contribution to PV is not
enough,the quality of the interim product is
not good enoughfor a larger assembly.
Further, productivity values cannot be
precisely determined. Therefore, they are
guidance to serve a judgmental processfor
evaluating work packages.Their use at first
involves trial and error and thereafter experience. For example, a geographicaldivision
of a product into seeminglyideal zonescould
yield unacceptablework packageswhen the
neededwork processesare analyzedby problem areas. Zone boundaries would then be
adjusted until there is an ideal compromise

of zone and problem area considerations.

Each proposed work package should be so
evaluated regardlessofwhether it has been
employedin the past. It is probablethat some
circumstance,especiallyregarding resources
and time available,will have changed.
The iterative development and evaluation of work packagesthrough the planning
processis illustrated in Figure 2-16. Work
packages classifredby product aspects are
systematically analyzedin order to determine
their productivity values. The analysesmay
be iterative through several planning levels.
The work packagesare immediately updated
based upon restudy following production.

SHIPBUILDING
MANAGEMENT
THEoRY

Thus, work packagesreflect an accumulation

ofexperience.They are adaptableto ships of
different sizes and types. Regardlessofdifferencesin functional systems,zonelproblemarea/
stage classifications of comparable work packages for different size ships of the same type
change very little. Even for different types of
ships, such classifications remain essentially
the same for work related to bows, sterns,
engine rooms, and superstructures.Therefore,
initial costs may be amortized over subsequent
ship construction projects.
References
1. Hammon,C.P.,'Analysisof ProgramChanges
and Internal Labor Adjustment Costsin
Shipbuilding," Ph.D. dissertation,University
ofRhode Island, 1980.
2. Pegels,C.C.,"Start Up or Learning CurvesSomeNew Approaches,"DecisionSciences,
Vol. 7, No. 4, October 1976.
3. Cochran,E.B.,'T.{ewConceptsof the Learning
Cuwe," Journal of Industrial Engineering,
Vol. 11,No.4, July/August1960.
4. Martin, J.C.. "The Labor Market of the
United States Shipbuilding Industry,"
Unpublished Ph.D. dissertation,George
Washington University, 1978.
5. Brechling, F ., Inuestmentand Employment
Decision,ManchesterUniversity Press,
Manchester,England, 1975.
6. Cochran.E.B..'"The Pattern of Direct Labor
Cost During Phaseout,"Manufacturing,
Engineering Transactions,May 1975.
7. Baron, David P., "Incentive Contracts and
Competitive Bidding," Am erican Economic
Reuiew,Vol. 62, No. 3, June 1972(384-394).
8. Cochran,E.B., "Measuring and Predicting

59

Production Disruption Costs Due to Desigrr

Uncertainty and Delivery UrgenaSr,"Proceedin6s of Department of DefenseSeuenthAnnttal
Acquisition ResearchSymposium,Hershey,
Pennsylvania,June 1978.
9. Cochran,E.B. and A. J. Rowe,"The Sources
of Disruption to Project Cost and Delivery
Performance,"Proceedingsof the Department of DefenseSixth Annual Procurement
ResearchSymposium,West Point, New York;
June 1977.
10. Hammon, C., and D. Graham, "Disruption
Costs in Navy Shipbuilding Programs,"
Center for Naval Analyses, Alexandria,
Virginia, Study Report CNS 1149,April 1980.
11. Ranson,G.N., Grozp Techrnlogt: A Foundatinn
for BettcrTotal ComparryOpemtion,McGrawHill.Inndon. 1972.
12. Mitrofanov, S.P.,ScientificPrinciples of
Group Technology,English translation by
E. Harris, National Lending Library for
Scienceand Technology,Boston Spa,
England, 1966.
13. Hyde, W.F.,Improuing Productiuity by Clasafication, Coding, and Data Base Standardization, Mercel Dekker, Inc., New York, 1981.
14. Webster'sSeuenthNew CollegiateDistionary,
G & C Merriam Co., Springfield, Mass., 1963.
15. Thompson,A.R., and W.D. Beeby,"A
Broader View of Group Technology,"
Computers and Industrial Engineering,
4th quarter, 1979.
16. Naval Sea SystemsCommand, Ship Work
Breakdown Structure, Washington,D.C.,
1 April 1981.
17. Okayama,Y., and L.D. Chirillo, Product
Worh Breakdown Strucuture, National
Shipbuilding ResearchProgram, Maritime
Administration in cooperation with Todd
PacifrcShipyards Corporation,Revised
December1982.

CHAPTER III

P R O D U CT - O R I E N T E D
W O R K B R E A K D O W N S T R U CT U R E

1. lntroduction
Major shipbuilding cost drivers, the principles ofgroup technologyby which costs can
be controlled,and two classificationsystems
(work breakdownstructures)were presented
in Chapter II. In this chapter the logic and
proceduresby which group technology(GT)
is applied to shipbuilding will be introduced
and the product-oriented work breakdown
structure (PWBS) described in detail. The
theoretical model of shipbuilding partially
explainedwhy overlap and parallelism of planning, design,material definition, material procurement, and construction of interim products are key elements in high-productivity
shipbuilding. This overlap,alongwith highly
organizedplanning and controlling, contributes to maintaining a level or near-levelwork
load and to reducingidlenessand congestion.
Overlap is also necessaryto minimize the
inventory costof work in process,and to maximize the utilization of capital equipment.
The overlap of design, material definition,
material proflrrement, and production requires
that information developedin each phasebe
formatted according to the requirements of
the other. This commonalityis achievedusing
the product-oriented work breakdown structure
(PWBS) describedin this chapter.
Up to this point the generic term group
technologyhas been used extensively.A par-

60

ticular applicationofgroup technologyto shipbuilding, involving the integration of hull

construction, outfitting, and painting, will
be addressedin the remainder of this book.
[1] The componentsof this integrated approach are:
. the hull block constructionmethod
(HBCM), wherein hull parts, subassemblies,and blocks are manufac_
tured in accordancewith the principles
of group technology(family manufac_
turing) in organizedproduction lines
(also referred to as processlanes or
work flows)
o the zoneoutfrtting method (ZOFM),
which makes possibleefficiencygains,
through concurrent hull construction
and outfitting, by providing precise
zoneby stagecontrol for which there
are three basic stages:on-unit, on-block,
and on-boardoutfrtting, and a substage
for downhand outfitting on overheads
when blocks are upside down
e the zonepainting method (ZpTM), in
which sur{acepreparation and coating
are treated as an integrated aspectof
the overall constructionprocess
o family manufacturing, such as in pipe
piece family manufacturinC (ppFM)

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

. a classification system, the productoriented work breakdown structure

(PWBS), which facilitates the integration of the foregoing inherently different types ofwork by defining and
classifring interim products (parts,
subassemblies,outfit units, and blocks),
which permits coordinatedwork flows

61

Integrated hull construction, outfitting,

and painting affectsevery aspectofship construction. It requires collaboration among all
shipyard departments. Integrated planning
is achievedby discussion,trade-offs,and ultimately mutual consent.The overriding goal
is an increase in productivity for an entire
shipbuilding system.

2. Planning
for Production
Planning starts with preplanning, alsocalled
the marketing stage, and continues through
design and material defrnition. Preplanning
includes initial regulatory body review, preliminary definition of contract specifications,
proportions, lines, other technical data and
procurement specifications,and preliminary
block defrnition. Preplanning will be covered
in detail in Chapter VII. In order to successfully include production considerationsin preplanning, a building strategy is developed.The
building strategy reflects the capabilities and
preferences of the shipyard, modified to frt
the specifrcsof the vessel to be built. It is
developedby consideringblock breakdowns,
processlanes, and pallet lists and material
with which the shipyard has experience,and
which representits current best practice.The
building strategy helps to defrne and prioritize decisionsabout the shipbuilding project
at its earliest stages.The building strategy is
discussedin more detail in Chapters VI and
VII. An overview of designand material definition, the importance of overlap of these
stageswith production, and their impact on
PWBS will be treated in this section.This will
be expandedin a more detailed treatment in
Chapter VI.
Signifrcant overlap of design, material
procurement,and production is essentialfor
reducing the overall constructionperiod, but
overlap reducesthe time neededto organize
information developedby designers. Thus,
from the outset, desis-ninformation must be

formatted to more fully anticipate needs relating to material and production.

In addition to overlap in time, there is an
overlap betweenfunctional systemsand product aspects.The work breakdown structure
must allow for dual grouping. With PWBS,
work packagesare grouped by product aspects and cost classificationsby product resources.Thus, manpower expenditures can be
collectedby zondproblem area,/stage,while material costscan be collectedby system. Figure
3-1 showsthe basicmanagementcycleof any
large industrial project. The cycle includes
project estimating, planning (including design
and material defrnition),schedulingof work,
executing or actually performing the work,
and frnally evaluating (a review of how effective eachofthe phasesand the overall project
has been).In applying PWBS, the unique nature of each of these phasesin the management cycle is addressedin terms of system
versus zone orientation. Since a key to shipbuilding employingPWBS is the construction
ofinterim products,the executionstageis organizedby zone.That is, the ship will be built
by zone,to bejoined in the final stagesofconstruction. Consequently,efforts are made to organize support functions, such as planning, desigrr,material definition, scheduling,and testing
to match the zoneorientation of construction.
Figure 3-2 indicates the primary focus,
system or zone,of each of the phasesin the
managementcycle.Note that the processbegins with a systems orientation. This is a

62

SHIPPRODUCTION

ESTIMATING

PLANNING

SCHEDULING

EXECUTION

EVALUATION

Fig. 3-1.Industrialprojectmanagement
cycle.
view ofthe total final product, the ship as a
whole, broken down by systems (structural
and functional). During planning (including
design), a key transformation from systems
orientation to zone orientation occurs.This
zone orientation is then maintained through
execution and some testing (a part of evaluation) to mirror the manner in which the
work is performed.Finally, a transformation
back to systems orientation takes place to
permit overall evaluation of the product and
systems testing. These system-to-zoneand
zone-to-systemtransformations are a key to
group technologyshipbuilding (PWBS).
2.1. Designand MaterialDefinition
Design is divided into:
. basic design
. functional design

Fig. 3-2. Systemand zoneorientationsin the

managementcycle.
r transition design
. work instruction design (detail design)
During basic design, specifications which
establish performancerequirements are determined. Systems diagramrnatics are developed from the basic design as a part offunctional design.Simultaneously,material lists
divided into material-ordering zonesare developedfor eachsystemdiagrammatic.Other
key drawings,suchas generalmachinery and
block arrangements,are prepared as a part
of functional design.
During these frrst stages,drawings and
material lists are developedby system. During basic design, products are classified by
total system. Individual systemsare treated
in functional design. A system orientation is
appropriate in the early planning stages in
order to facilitate estimating and establish
spatial relationships.
During transition designthe work breakdown structure must accommodatea transition from system to zone groupings. Transformation to zone orientation begins with a
block plan for hull constructionand composite drawings for outfrtting. Theseincorporate

WORKBREAKDOWN
STRUCTURE
PRODUCT-ORIENTED

all gystemsand show zoneboundaries.They

are further processedto add problem areaL/
stageconsiderationsduring the work instruction stage. For hull construction these are
designatedon assembly,subassembly,and cutting plans. For outfitting, they are designated
on working drawings (work instruction drawings), each of which is developedtogether
with its own material list of fittings for onunit, on-block, or on-board outfitting. The
hierarchical subdivision continues by zone/
problem area./stagewith the preparation of
detail designdrawings forpipe piecesand components other than pipe and their respective
material lists. As shown in Figure 3-3, the
design processcontinuesuntil each zoneis
broken down to components that are to be
purchasedand to material requirements for
parts that are to be fabricated. This is the
lowest hierarchical level of classification.
Note that the compositesshownin Figure 3-3
are used for quickly conveying arrangements
and system/zonerelationships to detail designers. The latter refine arrangements and
designate stages during preparation of work
instruction and material detail design drawings. The entire design processis discussed
in more detail in Chapter VI.
Each design stage more clearly defines
material requirements.Each successorstage
honors commitments made and constraints
imposed previously, while producing more
detailed information. Most imporbant, each
successorstage transforms the developing
designinto a format better suited to the subsequentuser's essentialneeds.
Design as well as production groups are
organizedaccordingto classesof problemsin a
manner which complements planned zoning.
These zonesalso correspondto material procurement zones.In commercialship construction, there are typically three groups.Theseare:
. Hull
. Machinery
o Superstructure

63

Hull, machinery, and superstructure designate contiguousthree-dimensionalzones.In

naval ship construction, additional groups,
such as combat systemsand nuclear propulsion systems,may be included.
Each ofthe abovedesigngroupsprepares
key drawings, working drawings, and material lists in a manner to suit initial zone
designationssuch as those shown in Figure
3-4. Within such groups there is improved
"horizontal" communication,such as that between piping and vent duct designers assigned to machinery outfrtting. They become
more expert about their particular class,are
led away from insignifrcant fine tuning of systems, and instead focus on compositedrawings (marked to show how a ship is to be
assembled)and on structured material lists.
Systemar:rangementdrawings are eliminated,
and interference-freeand simplified composites (drawings or scale models) are developed
directly from diagrammatics. Thus design as
well as production is segmentedby classesof
problems in accordancewith the principles of
group technology.
2.2. Production
Figure 3-5 showswork processlanes, organized by classesor problem areas, and how
their end products must integrate for zoneoriented production. Fabrication shops and
assemblysectionsare groupedalongthe various processlanes.
Hull construction has traditionally been
assignedto a single production division associated with a single trade union. Therefore,
the general adoption of hull block construction in processlanes similar to those illustrated in Figure 3-5 precededzone-oriented
outfitting. The outfrt and integrated (hull
constructionand outfit) processlanes shown
are quite different from those of shipbuilders
who use system-oriented work packages for
outfrt fabrication and assembly.
For example, in a system-orientedproduction organization,workers are assignedto

64

MLS
MLF
MLP
MLC

SHIPPRODUCTION

- MateEt
- Mateiat
- Maleriai
- Material

List by System
List tor Ffiinggzane/Atedstage
List for ptE pece
List lorComponent (othet than ppe)

HULL

SUPERSTRUCTUBE

DIAGBAMMANCS

COMPOSITES

WORK INSTFUCNON

& MATEFIAL DETATL DESIGN ORAWINGS

SUPERSTRUCTUFE

MACHINEFY

Fig. 3-3. Desigrrprocess.

H - HULLOUTFIT
M - MACHINERY
OUTFIT
S - SUPERSTRUCTURE
OUTFIT
Fig. 3-4. Design outfrt specialtygroups.

PRODUCT.ORIENTED
WORKBREAKDOWN
STRUCTURE

rJ
*

ffi

B
ltr
t?

HE

66

SHIPPRODUCTION

a pipe shop which fabricates and assembies

pieces required for all pipe systems. In a
zone-orientedorganization,suchworkers are
assignedeither to a fabrication shop or to a
team specialized for a specific category of
assembly problems. This has given rise to
trades which cut acrosstraditional jurisdictions. Workers called "assemblers"or "fitters,"
in addition to putting together all but highpressurepipe, assembleeverybhingfor which
a manual "stick" welder or spanner wrench
suffices. This includes pipe supports, walkways, handrails, electric cable trays, etc.
Other trades represented on a team are as
neededfor special or extensivewelding, rigging, insulating, joinery, etc. In shipyards
that have changed completely to zone-oriented construction,many related specialists
have been combined into two: a fitter who
does some welding, and a welder for special
or large amounts of welding. As zone outfitting becomesmore universally adopted, a
logical developmentwould be an "outfitter"
trade. While such simplified trade demarca-

tions would facilitate group technology shipbuilding, traditional trade organization would
not precludethe adoptionofthese principles.
2.3.Controlling
Zone-orientedschedulingis necessaryto control the flows of work on various processlanes
so that the creation of interim products anticipates only immediate needs.Such scheduling coordinateshull construction,out- fitting, and painting, and allows periods after
work stages for the collection and distribution of interim products to other work stations. The goal is to minimize buffer storage.
Thus, integrated schedules,as shown in Figure 3-6, are essential for fabrication through
final outfitting. The schedulesshould address
all fabrication and assemblywork, including
lofting and painting.
A primary end product of schedulesis
flexibility, the ability to quickly identify good
options basedupon constant feedbackabout
material procurement and work progress.
Adjustments are neededto counter potential

INTEGRATED
HULL
CONSTRUCTION
OUTFITTING
&
PAINTING
MASTEF
SCHEDULE

ON-BOARDOUTFITTING
MASTERSCHEDULE

Fig. 3-6. Integrated schedulesfor hull construction,outfitting, and painting.

PRODUCT-ORIENTED
WORK BREAKDOWNSTRUCTURE

delays a4d early completions.Adjustments

could include transfer of workers between
processlanes, the use of overtime, or shortterm schedulechanges.The objectivesare to
maintain uniform work flow within each process lane and coordinated outputs from all
processlanes.
Feedback implies accurate progress reporting of manpower and material costs by
zonelproblem area./stage.Accurate progress
reporting is facilitated by relatively small,
uniform work packages.Small, uniform work
packages lend objectivity to progress reporting; a work packageis either completedor it
isn't. The problems cited in Chapter II of
borrowing man-hour budgets, and of losing
track of man-hours becauseof the long time
that a work packageis open,are eliminated.
This also brings work packages into congruence,time-wise, with the most detailed (weekly)
schedules.
2.4. Costing
Zone orientation introduced the concept of
control linked to many relatively small
amounts of material groupedby zonelproblem
area./stage.Progress reporting and cost collections are zone-oriented,so that managers
have tangible means of corroborating work
completedin order to forecastwork remaining
and resourcesrequired for completion.In order
to serveestimators, manpower costs by system have to be rationalized. Certain indices,
describedin Chapter VII, are neededfor distribution of spent man-hours to systems.
The indirect collectionof costsby system
may seemto be a degradationof feedbackto
estimators. However, while collecting costs

by a zone-oriented method and applying

them to systems in accordancewith estimated distributions is less precise, it producesmore accurate data due to inherently
better control. Becauseof the multiple character of a PWBS, material usage is easily
collected both by system and zone. Where
functional designers are required to identify all materials for each system diagrammatic, there is a quick corroboration of the
material estimate. If a catastrophicerror is
disclosed,there is time for remedial measures before the major procurement effort
begins.
Additionally, when functional designers
are required to divide each material list by
system into lists of material required for
various material ordering zones,it is possible
to quickly corroborateestimated manpower
requirements. This is feasible when the systemlzone transformation indices are based
upon material, suchas man-hours-per-hundredweight of frttings, man-hours-per-footof electric cable,etc. Where thesetechniques are applied, the rapid feedback to estimatorsis of
sufficient accuracyfor immediate use in preparing another estimate.
All material requirements are listed by
system for purchasing and subsequentlyon
structured (zone-oriented)material lists for
issue purposes.Therefore,the interrelationships maintained by designerspermit material progressingby zone to be accurately
convertedto material progressingby system
if a customer so desires. Similarly, the systemJzonetransformation indices could serve
a customer'srequirement to progress manpower by system.

3. ZoneConstruction
Method
Becauseinherently different types of work
are required, a product-oriented breakdown
of ship construction work should accommodate the following zone-orientedmethods:

rr
,'

67

. hull block construction method

(HBCM)
. zoneoutfitting method (ZOFM)
. zonepainting method (ZFTM)

I
68

SHIPPRODUCTION

Also, because large quantities and varieties

of pipe pieces,ventilation ducting, structurals, etc., are needed,the work breakdown
should accommodateproblem area-oriented
family manufacturing (FM).
Work packagesare ideally sized for the
three zone-orientedmethods when, for each
processlane, their required working times
for all manufacturing levels are the same,
and, within each manufacturing level, their
work contentsare the same.Compliancewith
these conditionspermits eachprocesslane to
be operatedlike an assemblyline where work
starts, flows, and stops in unison. In order to
balance work accordingly, special manufacturing levels outside the main flow are neededto
adjust work amounts and to provide for interim product features that would otherwise
be disruptive.
3.1.HullBlockConstruction
Method(HBCM)
Optimum blocks(zones)are key objectivesas
the basis for control in HBCM. But blocks
also impact on zone outfitting and painting.
Therefore,the defrnition of blocks,compared
to other interim products,has the greatestinfluence on shipbuilding productivity. Blocks
should be designedso that:
. for block assemblypurposes,they are
assignableto one of a minimum number
of work packagegroups, considering similarities in problem area and the need
to minimize variations in working times
. for block erection purposes,they will
be stable confrgurations requiring no
temporary support or reinforcement
and otherwise shapedto achieveminimum working times
. for on-block outfitting and painting,
they are sized for maximum accessibility (maximum area and./orvolume)
Also, there should be similarities in volume,
weight, shape, etc., even at the expenseof
designconvenience,to distribu te work evenly

throughout the fabrication and assembly

levels which precedeblock assembly.Thus,
planners have to keep in mind that breaking down the work leading to block assembly requires shifting welding from difficult to
downhand positions to reduce the working
times needed, and distributing much work
traditionally performed during block assembly amongearlier levelsto equalizetheir working times. Chapter VI providesadditional discussionof the formulation of a block breakdown for a new ship construction project.
For large ships,blocksplanned in accordance with the foregoing should also be of the
largest size permitted by facilities. The same
planning applied to a smaller ship of the
same basic type quickly achievesnearly the
samework balancewith the samegameplan.
This is an important competitive advantage.
However, pertinent work packagecontents,
working times, and interim product sizes
becomesmaller. Thus, there is sometimes
need for an additional manufacturing level
for joining blocks into grand blocks.
With regard to this objective,it is practical to plan hull construction in seven levels
as shown in Figure 3-7. Starting with the
block level, work is subdivided down to the
parts fabrication level to optimize work flow.
In contrast, work assigned to the grand
block level serves to minimize the duration
required for erection in a building dock. The
main work flow path, or processlane, follows
the arrows on Figure 3-7. For maximum productivity, this main work flow must be level.
Within each level other than the grand
block and hull erection levels, the resulting
proposedinterim products are examined for
similarities in their product aspects. Then
they are groupedby similarities in order to:
. further modularize the production
processes
ojustify expensive but highly efficient
facilities

. achieve
manpower
savings

STRUCTURE
WORKBREAKDOWN
PRODUCT-ORIENTED

Fig. 3-7. Hull block constmction method (HBCM) manufacturing levels.

Typical groupings by product aspects are

presentedin Figure 3-8. The horizontal combinations characterize the various types of
work packages that are requisite and suffrcient for the work to be performed for each
level. Vertical combinations of the various
types of work packages denote the process
lanes for hull construction work flow which
correspondto those simply illustrated in Figure 3-5.

are allocated'
Whenpr0ductresources
each work packageis optimally sized,based

on determination of its productivity value

(PV). Some reiteration can be expected, becausegrouping by problem area at each level
is dependent upon the productivity values
achievable. Maximum productivity is obtained when work is evenly allocated to work
packages grouped by their product aspects,
and there are quick responses to potential
work imbalance, such as shifting workers
between manufacturing levels and/or flow
lanes, authorizing overtimer or even making
astute short-term schedulechanges.

70

SHIPPRODUCTION

3.1.1. Parts Fabrication.As shown in Figure

3-8, parts fabrication is the first manufacturing level. It producescomponentsor zonesfor
hull construction which cannot be further
subdivided. Work packagesare grouped by
zone,problem area, and stage.Distinct problem areas based on differencesin raw materials, finished parts, fabrication processes,
and relevant facilities are:
. parallel parts from plate
. nonparallel parts from plate
o internal parts from plate
o parts from rolled shapes
o other parts, such as parts from pipe
Stage is determined by grouping similarities
in part types and sizes,as follows:
. plate joining or nil
. marking and cutting
. bending or nil
"Nil" indicates
that no product aspectexists;
thereforeit is left blank in categorizationand
coding and is skipped in processflow.
For large quantities of parts to be bent,
problem area can be subdivided by the resourcesavailable, such as:
. universal press (single-axisshallow
curvature)
. presswith die (small parts, such as
bracket flange)
. mechanizedline heating (double-axis
shallow curvature)
. manual line heating (double-axisdeep
curvature and correctionof any part)
A faceplate, for example, is marked and
nested on a plate with other such parts that
can be cut in onepassby a multiflame planer.
Those which require different curvatures are
then grouped together, provided they can be
processedby a press without the need to
changedies.Faceplates,including thosethat
are to remain straight, are then grouped by

block and distributed to succeedingworkpackages. Typical groupings of work packages

for parts fabrication are illustrated in Figure 3-9. Each part shown correspondsto a
hull construction zone which cannot be subdivided.
3.1.2.PartAssembly.The secondmanufacturing level is specialand outsidethe main work
flow. Its typical work packagesare grouped
by problem area as:
r built-up part (such as tee- or el-section
longitudinals of large or unusual
sectionsnot rolled by mills)
. sub-blockpart (such as a part which is
a weldment, typically consistingof a
bracket fitted with a faceplate or flat
bar, as shownin Figure 3-10)
The sub-block "part" concept is a planning
technique for shifting work from the subblock assemblylevel, where excessivework
volume is otherwise probable, to an earlier
level outside the main work flow. Undertaken with simple facilities as comparedto
those required for sub-blockassembly,such
as mechanized conveyors,manufacturing
sub-block"parts" in the part assemblylevel
is a means of balancing work and conserving
resources.Further, as such "parts" are only
used in sub-blocks,zone identification employs the same code as for sub-blocks(see
Figure 3-8). Stageis divided into:
. assembly
. bending or nil
3.1.3. Sub-blockAssembly.Sub-blockassembly appearsin the third manufacturing level
of Figures 3-7 and 3-8. A zone is generally a
weldment, consisting of a number of fabricatedand/or assembledparts, which will eventually be fitted on a panel during block assembly. Typical work packagesare grouped oy
problem area for:

PRODUCT.ORIENTED
WORKBREAKDOWN
STRUCTURE

CODES

ASPECTS
PRODUCT

PLAN'G MFG
LEVEL LEVEL

AREA

ZONE

ZONE AREA STAGE

STAGE
ul

=
-

SHIP

=
o
o

o
J

o
I

o
o
cc

U
(E

o
L

ul
z.

TEST

J
J
l
T

z.

trJ

ut

ERECTION

PRE-ERECTION

q,
UJ
L

U
(E

L
J

o
o

g
o
U

z.

UJ
t

o(t,

e
f

JOINING

NIL

BACKASSEMBLY

NIL

l!

U'

ASSEMBLY

(r
o
E
t!
of
U'

FRAMING

NIL

PI.-ATE
JOINING

NIL
NIL

Y
()

U
o
o
o

J U

40
AO
=<)
CE

('

!(rrr

6o

do

o
q t6

()

NIL

NIL

rlJ
U'

o
Oul

:<

Qut
d)X

=
6
U
U'

U'

()
U

o
U
o

OUJ

d8 d8

d(J
:l

U>

ASSEMBLY

6()

U
o

o
U

f
U'

BENDING

NIL

Qu.t

4F

ASSEMBLY
l

rr
=t
FdgF
i d fr.6

BENDING

IJJ

E
{

i4sE
l'oIIJ

=7
F
l!;

>n .t;

(l
U

Fig. 3-8. HBCM classifrcationby product aspects.

o
F
U)

EE

,B.BLOCI BUILT.UP
PAKT
PART

)z
()

5
x
q6

Qtu
-o
=o

ASSEMBLY

:<
o

U'

(t

o
()

C)
:<
o

SIMILAR SIMILAR PLATEJOINING

SIZEINA SIZEINA
I.ARGE SMALL
QUANTITYQUAMNry
BACKASSEMELY

oO
z.o

1!

o
F

()

o
dH

BACKASSEMBLY

NIL

)v,

NIL

FLAT lt tnrtrl
PANEL PANEL

=
(t,

BACK
PRE.ERECTION

=
F

o
2

t
F
U'
t
ut
l
U'

NIL

MARKING
& CUTTING

PI.ATEJOINING

U
o
rlu

NIL

Ful

FE Fg

o
U

(,
F
o

PAFI FABRICAIION

P L A T EJ O I N I N GO R N I LS T A G E

LEVEL

MAFKINGAND CUTTINGSTAGE

BENOING
O B N I LS T A G E

AREA:PARALLEL
PARTSFROM PLATE
+

<t--:1'^

AREA:NON.PARALLEL
PARTSFHOM PLATE

<-.---->

\-------

AREA:INTERNAL
PARTSFROM PLATE

.,.....'_

ffi
\

t
| t r
\_:i--:-i----:

s-

'

, \ '
ji-:i---t:J

-------'

\ r ' . . , . . \ r
\--i--i--,--------

\
. _ f l
=YJ

s13

AREA:PARTSFROM
ROLLEDSHAPES

Fig. 3-9. Parts fabrication problem area and stage classifications.

r similar size in large quantities, such

as large transverse frames, girders,
floors, etc.
r similar size in small quantities
Subassembliesfalling within the first problem area regardless of their design differencescan be mass produced size by size on
process lanes with appropriate facilities,
such as conveyors.Thosein the secondcategory require ajob-shop approachbecauseof
insufficient numbers for any one size, and
different working times required for the different sizes that are normally encountered.
Stage classificationsare:
. assembly
. back assemblyor nil

PARTASSEMBLYLEVEL
ASSEMBLYSTAGE

BLOCKPABT

Fig. 3-10. Part assembly level classification.

During back assembly,parts and./orassembled parts are frtted on the oppositeside ofa
marked surface of a main part (back assem-

STRUCTURE
WORKBREAKDOWN
PRODUCT-ORIENTED

bly is additional frtting after overturning).

Examples are shown in Figure 3-11.
3.1.4.Semi-blockand BlockAssemblyand Grandblock Joining. A block is the key zone for hull
construction as indicated in Figures 3-7 and
3-8. It may, dependingon circumstances,be
planned in three assemblylevels:
. semi-blockassemblv
. block assembly
. grand-blockjoining
Only block assemblyis in the main work flow.
The other levels provide useful planning alternatives. All are planned in accordance
with the conceptof grouping work packages
by problem area and stage.
A semi-blockservesthe needto assemble
a partial zone separate from a key zone
(block) whenever a block would otherwise
disrupt work flow. When a semi-blockis employed, the block assemblylevel is where it
joins its "mother" block,which was processed
in the main work flow.
Grand-block joining (combining a number ofblocks to create a larger block at a site
near a building dock) reduces the working
time neededfor erection in a building dock,
producesa shapethat is more stablefor erection purposes,and providesgreater area and
volume, which facilitates further on-blockoutfrtting and painting. The grand-block level,

73

which is outside the main flow, is neededin

order to quickly achieve a nearly uniform
work balance when zone divisions from a
large ship are applied to a small ship. The
ensuing smaller size blocks are joined into
grand-blocksto minimize the working time
neededin a building dock for erection. The
zone of semi-blockand block assembly,and
grand-blockjoining rangesfrom block to ship,
as shown in Figure 3-8.
The semi-blockassemblylevel is divided
by problem area in the samemanner as for the
sub-blocklevel Most semi-blocksare rather
small in size and two-dimensional, so that
they can be producedin a sub-blockassembly
facility. In planning work, this should be the
point of divergencefor separatingsemi-block
assemblyfrom block assembly.The grouping
for stage for semi-blocksis also the same as
for sub-blocks,as also shown in Figure 3-8.
The block assembly level is divided by
problem area using distinguishing features
of the panel needed as a base for attaching
parts, assembledparts, and./orsub-blocks,and
uniformity of required working times. These
characteristicsdeterminewhether platens or
pin jigs are required, or blocks are to be
assembledin a flow where work starts and
completesin unison. Becauseof their uniqueness,superstructureblocks are addressedseparately. Pertinent problem area divisions and
necessarydefinitions are:

SUB.BLOCK
LEVEL
ASSEMBLY
ASSEMALY
STAGE

@@ =K
*il9

$VM,
AREA:SIMILARWOFKCONTENTIN LARGEOUANTITY

Fig. 3-11. Sub-blockassemblylevel examples.

f
VP
lN SMALLOUANTITY
AREA:SIMILARWORKCONTENT

74

SHIPPRODUCTION

. flat (working time is uniform and

there are no projectionsfrom panel
undersideswhich require specialjigs
or which would interfere with platens
equippedwith conveyors)
. specialflat (working time is non-uniform and,/orunique jigs or supports
are needed)
. curved (working time is uniform)
. specialcurved (working time is nonuniform and,/orunique jigs or supports
are needed)
r superstructure
Becauseof variations in working times and/
orjigs, specialflat and specialcurved blocks
are not assembledin facilities designedfor
work flow where starts and completionsare in
unison.Thus, they require ajob-shopapproach.
Ifthe quantity ofblocks to be producedis
small, less than five problem area classifrcations should be considered.Typical classifrcations by problem area are illustrated in
Figure 3-12.
As shown in Figure 3-8, the block assembly level is phasedby stage as follows:
.
.
.
.

plate joining or nil

framing or nil
assembly
back assemblyor nil

The assemblystageat the block level is for combining a panel with parts, assembled parts,
and,/orsub-blocks,and sometimesa semi-block.
When many blocksare required it couldbe useful to add further classificationsby problem
area, basedupon internal framing, such as:
'
r
.
r

eg9 box
longitudinals attached beforewebs
longitudinals attached after webs
other

At the grand-blockjoining level, only three

classifrcationsby problem area are normally
required:

. flat panel
. curved panel
. superstructure
Stage at this level is subdividedinto:
r joining or nil
o pre-erectionor nil
. back pre-erectionor nil
For very small ships, the pre-erectionstage
provides forjoining grand blocks in order to
creategrand-grandblocks.Back pre-erection
providesfor further assemblywork after turnover, such as attaching bulwarks, chain
pipes,etc. Figures 3-18 through B-28 show
relationships between semi-blocks,blocks,
and grand blocks that were actually employed for construction of a 22,000 dwt general cargo carrier. It was purposely selected
as the basis for illustration becauseit is one
of a kind, rather than a ship of a standard
series.
3.1.5.Hull Erection.Erection is the frnal level
of hull constmction where the entire hull is
the zone.Problem areas at this level are:
. fbre hull
o cargohold
. engine room
. aft hull
. superstructure
Stageis simply divided into:
o erection
r test
Tests at this level, such as tank tests, are
independent of erection and are distrnguishedby the size of their work packagesas
comparedwith the tests and inspections of
other levels.The latter tests and inspections
are includedin the packagesofeach level and
respectivelyimplemented at the time when
eachinterim product is being finished.

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

Fig.3-12(a)

Fis.3-12(b)

Fig.3-12(c)

Fig.3-12(d)

Fig.3-12(e)

F i g .3 - 1 2 ( f )

75

Fig. 3-12. Block assembly problem area and stage classifications. (a) Problem area: flat; stage: egg box
framing. (b) Problem area: flat; stage: assembly. (c) Problem area: special flat (greater work content);
stage: assembly off flow. (d) Problem area: curved; stage: assembly. (e) Problem area: special curved
(gteater work content); stage: assembly off flow. (f) Problem area: curved and special curved; stage:
assembly on and offflow.

76

SHIPPRODUCTION

3.2. Zone OutfittingMethod

The zoneoutfrtting method (ZOFM) is a natural consequenceofthe hull block construction
method (HBCM), becauseboth employ the
same logic. Shipyards which employ ZOFM
assemble most outfit componentsindependent ofor on hull blocks.
Just as for hull construction, zone divisions from a similar previously built ship are
tailored to fit a new contract design. The
coded significance in work package numbers
remains essentiallyunchanged.Thus, everyone involved in design, material definition,
procurement, parts fabrication, and assembly has knowledge of how outfrtting is to
progTess.
3.2.1.Zone by ProblemArea by Stage.HBCM
planners define interim products starting
with a hull as a zone,then subdivide it into
block zones which in turn are divided into
sub-blockzonesand soon.The processis completed when zones correspondto parts and

cannot be further subdivided. The nature of

any of these zonesassociatesitwith a specifrc
manufacturing level. This breakdown is natural for hull construction but not for outfitting.
ZOFM planners have to consider block
zones for hull construction. Elsewhere, they
shouldbe free to devisezoneswhich best suit
the work at hand. Their outfrt zones at one
manufacturing level can be independent of
zones in previous or succeedinglevels. For
example,in hull construction, zonesizes increaseas manufacturing progresses.On the
other hand, for control purposes,outfitting
zones at earlier stages could be larger and
have no common boundaries with zonesdefined for subsequent stages. Overlapping
zonesare ofno consequenceprovided they are
designated for different stages. Thus, while
there is greater freedom in defining outfrt
zones, specifying zone by problem area by
stage affords absolute control ofwork even in
a confined region that contains portions of
many systems.Outfitting work packagescor-

BL@K ASEMALY LEVEL

tuTEJorNrNGsTAo

FMMTNGSTAGI

ASSEMALYSTAGE

B^CKASEMALYSTAG

Fig. 3-13. Semi-blockand block assembly,showingthe bottom center block of a cargohold

STRUCTURE
WORKBREAKDOWN
PRODUCT.ORIENTED

77

B L @ K A S S E M B L YL E V E L
P $ T E J o I N I N G S T A G E I F F A M I N G S I A G E I A S S E M B L Y S T A G E

BL@KASEMBLY LEVEL
ASSMBIY SIAGT

PANEL + PARE + SUBAL@KS =

590TONS
INCLUOING
OUTFIT

Fig. 3-14. Block assembly and grand-block joining, showing the top of a wing tank.

SMI8L@X

ASSEMELY LEVEL

ASSEMALY STACE

gL@K A$EMALY LEVE!

ASSEMBLY SIAOE

Fig. 3-15. Semi-block and block assembly, showing a bottom wing tank side shell with hopper and the
transverse hopper in a cargo hold.

S H I PP R O D U C T I O N

B L O C KA S S E M E L YL E V E L
A S S E M B L Vs r A c E

ercx rsseuarv

s'nct

GMNDA!@X

J O I N I N GL V E L

J O I N I N CS I A G E

z2
))

75E TONS
INCLUOING
outa'T

Fig. 3-16. Block assembly and grand-blockjoining,

center between cargo holds.

P U T E J O I N I N GS - A G E

showing a corrugated transverse bulkhead and a deck

A S S E M S L YS I A G E

GFANO BLOCK JOINING LEVL

A S S E M B L YS I A G E

PANEL + SUBAL@KS =

Fig.3-17. Block assemblyand grand-blockjoining, showing a cant block.

WORKBREAKDOWN
STRUCTURE
PRODUCT.ORIENTED

B L O C KA S S E M A L YL E V L
A S S M 8 L YS T A G E

Fig. 3-18. Block assembly, showing an upper deck and an engine room flat.

S E M I . B L @ KA S S E M E L YL E V E L
assEMaLy srAGE

B A C K A S S E M B L Ys r A G E

/'v

M I

trg
u_-,'
\
ft t ,\ \ \
('\l-uH

\ \-/v ' ,

Fig. 3-19. Semi-block and block assembly, showing a bulbous bow.

LEVEL
STAG

79

SHIPPRODUCTION

80

LEVEL
8L@KASSEMBLY
putE JorNrNG
srAGE

AssEvB.tsracE

)
re*

"'^,,eleo

Fig. 3-20. Semi-block and block assembly, showing a forecastle and the upper deck in a fore body.
GMNOBL@(JOINING
JOI!ING STAGE

PRE.EBECTIONSTAGE

LEVEL
BACK PAE.RECTIONSTAGE

Fig. 3-21. Grand-blockjoining, showing a forecastleand the upper deck in a fore body

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

gL@K ASSEMSLY LEVEL

pur

JorNtNo sracE

A S S E M B L Ys r A G E

G M N O - 8 L @ K J O I N I N GL E V E L
J O I N I N GS T A G E

Fig. 3-22. Block assembly and grand-block joining, showing the bottom of an engine room.

BLOCK ASSEMBLY LEVEL

ASSEMBLY STAGE

PANEL + PARTS +
SU&SLOCKS =

Fig.3-23. Block assembly. showing the side shell of an engine room.

81

82

SHIPPRODUCTION

responding Lo a zonelproblem area,/stageare

called pallets. Pallets sequencedin their order for executioncomposethe outfrtting plan.
Adapting a pallet list from a previously constructed ship avoids much duplication of effort. It is a singular means for managers to
control the application of experience,as opposed to depending on experiencevested in
individuals.
3.2.2. On-unit,On-block,and On-boardOutfifting.
On-unit refers to a zone which defines an
arrangement of fittings to be assembledinhouse,independentof hull structure. Assembly of suchfrttings is calledoutfitting on-unit.
It enhancessafety and reducesboth required
man-hoursand durations which would otherwise be allocated to outfrtting on-block and
on-board.
On-blockfor outfrtting purposesrefers to
a rather flexible relationship between block
and zone.The assemblyof fittings on any structural subassembly(e.g., semi-blocks,blocks,
and grand blocks),is referred to as outfitting
on-block.The zone applies to that region being outfrtted. The fitting arrangement on the
ceiling of a block set upside down is a zone.
Following block turnover, the fitting arrangement on deckis anotherzone.
On-boardis a division or zonefor packaging work for the assemblyof fittings during
hull erection and after launching. An ideal
zone for outfrtting on-board avoids the need
to disperse and/or continuously relocate resources,particularly workers.In general,compartments defined by shell, bulkhead, deck,
or other partitions are suitable. Even entire
cargoholds,tanks, enginerooms,superstructure decks, or weather decks can be useful
zonesfor frnal outfrtting on-boardstages.
ZOFM planners,addressingthe need to
break down outfit work into packages,consider outfit componentsfor all systems in an
on- board zoneand try to maximize the amount
frtted into on-blockzones.Similarlv. thev con-

sider outfrt componentsfor all systemsin an

on-block zone and try to maximize the
amount fitted into on-unit zones.Their objective is to minimize outfit work during and
after hull erection.
As in HBCM, maximum productivity is
achievedwhen:
. work is equally apportionedto work
packagesgroupedby product aspects
at the most efficient manufacturing
levels
. uniform and coordinatedwork flows
are maintained by shifting workers,
overtime, and,/orshort-term schedule
adjustments
Work packages are optimally sized when
their work contents are nearly uniform. The
balancing of work among packagesrequires
considerationofgroups ofcomponentsby the
product aspects zone, problem area, and
stage.This balancingof work strongly affects
other factors, such as the allocation of manpower and scheduling.Other important objectives of ZOFM planners include:
. shifting fitting work, especiallywelding,
from diffrcult positionsto easier downhand positions,thus reducing both the
man-hours neededand the durations
required
. selectingand designingcomponentsso
as to organizegroups offittings that can
be assembledon-unit, thus simplifying
planning and schedulingby keeping the
different types ofwork separateat the
earliest manufacturing levels
. transferring work Ilom enclosed,narrow,
high, or otherwise unsafe locationsto
open,spacious,and low places,thus
maximizing safety and accessfor
material handling
o planning simultaneousexecutionof
many work packages,thus decreasing
the overall fittins duration

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

83

OPERATION
A N DT E S T

ON.BOARD
OUTFITTING

ON-BLOCK
OUTFITTING

COMPONENT
PROCUREMENT

Fig. 3-24.Zone outfrtting method (ZOFM) manufacturing levels

Consideringthese requirements,it is practical to plan outfitting in six manufacturing

levels as shown in Figure 3-24. The component, unit, and grand-unit levelsare executed
independent of the hull structural zones in
which they will eventually be fitted. The onblock and on-board levels are, ofcourse, entirely dependenton structural entities.
In order to minimize the impact of these
dependencies,fitting componentsshould be
assembledinto units and grand units as much
as possible,provided that they are truly independent (i.e., rigid and stable without extraordinary temporary reinforcements or supports). This approach is the primary means
for shortening the durations required for onblock and on-boardoutfitting.

Within each level, other than that for

grand units, the resulting proposedinterim
products are examined for similarities in
their product aspects.Then, as in the HBCM,
they are grouped by similarities to further
modularize the production process,justify
expensivebut highly efficient facilities, and
achievemanpowersavings.
Typical groupingsby product aspectsare
shown in Figure 3-25. Horizontal combinations characterize the various types of work
packagesthat are requisite and sufficient for
the work to be performed for each manufacturing levei. Vertical combinations of the various work packagetypes denote the process
lanes for outfitting work flow which correspondto thosesimply illustrated in Figure 3-5.

84

SHIPPRODUCTION

As the implementation of ZOFM progresses,the need becomesgreater for balanced planning and scheduling,and cooperation between hull construction, outfitting,
and painting planners.
3.2.3. ComponentProcurement.As shown in
Figures 3-24 and 3-25, component procurement is the initial manufacturing level. It
producesinterim products or zones for outfitting for which no further subdivision is
neededby the shipyard. Typical work packages and material requisitions are grouped
by zone and by problem area to addressthe
separateprocurementproblems:
. in-house manufacturing
. outside manufacturing
e purchasing
These problem areas are further classified
by requirements for manufacturing drawings,
purchaseorder specifications,and raw materials, as shown in Figure 3-26. When preparations for outside manufacturing are the
same as for in-house,a shipyard retains better control, avoidsvendor drawing approvals,
and makes eligible many small firms that do
not have design or purchasing departments.
After having performed groupings by
zone,problem area, and similarities in component types and sizes, further grouping is
made by stage as follows:
. design and material preparation or nil
. manufacturing or nil
o palletizing
The palletized componentsare assigned to
their respectivework packagesat subsequent
manufacturing levels.
3.2.4. Unit Assembly and Grand-unit Joining.
Just as a block is a key zonefor hull construction, a unit is a key zone for outfrtting which,
as illustrated in Figures 3-24 and 3-25, may
require only a single manufacturing level.

Productivity is enhancedwhen units are similar in required man-hours for assembly,numbers of components,volume, weight, design
standards,etc. Grouping by such similarities
facilitates organizing and uniformly loading
processflow lanes.
As indicated in Figure 3-27 (Plate I), unit
sizesvary significantly. Therefore,two problem areas are designatedat the unit assembly level:
r large size
. small size
The distinction is by required lift capacity.
Large units weigh more and small units
weigh less than one ton. If many small units
are planned for assemblyof larger units, another manufacturing level may be included
for sub-unit assembly. Problem areas at the
unit level could be further subdividedinto:
. machinery unit (machinery combined
with aII adjacentcomponents,including
foundation,pipe pieces,valves,supports,
walkways, ladders, etc.)
o pipe unit (no machinery,just pipe
piecescombinedwith valves, supports,
walkways, etc.)
. other (hatch coverswith coaming,
masts,etc.)
Stagefor unit assemblyis divided as:
. assembly
r welding or nil
The welding stage applies for extensive or
specialwelding requirements,as welding incident to routine unit assemblyis performed
by fitters during the precedingassemblystage.
Someshipyards have developedmachinery units into standard arrangements which
can be adaptedfor various types and sizesof
ships. As required design and material definition is already available,much planning for
a standard machinery unit can progressjust

PLATE I

Fig.3-27(a)

Fig.3-27(b)

Fig.3-27(c)

Fig.3-28(a)

Fig. 3-29

Fig.3-28(b)

Fig. 3-27. Outfit units. (a) An engine room tank top

unit. (b) Pumps, valves, pipe pieces, and foundations
incorporated as a unit. (c) Pneumatic tubing and supports form a small unit. Fig. 3-28 (.a & b). Typical
machinery outfit units (the same problem area). Fig.
3-29. Machinery outfit unit assembled on a common
foundation (courtesy Avondale Shipyard).

PLATE II

Fig.3-35

Fig.3-36

Fig.3-38

Fig. 3-39

Fig. 3-40

Fig. 3-35. Engine room tank top components

fitted on-biock. Fig. 3-36. Outfitted engine room
blocks erected over previously landed outfit
units (courtesv Avondale Shipyardt. Fig. 3-37.
Downhand outfitting on-block, in an upside-down

block. Fig. 3-38. On-block outfi tting and painting,

on the ceiling and on the deck. Fig. 3-39. Two
complex blocks, outfitted together, upside down,
to ease erection. Fig. 3-40. On-board outfittirre
performed priorto landing the main engine.

PLATE III

Fig.347

Fig.3-48

Fig. 3-49

Fig.3-50

F i g .3 - 5 1

Fig. 3-52

Fig. 3-47. On-block outfitting of a forecastle deck

upside down. Fig. 3-48. On-block outfitting of'a
forecastle deck right side up. Fig. 3-49. On-block
outfitting of a side shell and engine room flat
upside down. Fig. 3-50. On-block outfitting of'a

side shell and engine room flat right side up. Fig.
3-51. On-block outfitting of the block shown in
figures 3-49 and 3-50 being completed right side
up next to the building dock. Fig. 3-52. Erection of
the block shown in figures 3-49 through 3-51.

PI"ATE IV

Fig.3-54

Fig. 3-53. On-block outfitting of one_halfof a

superstructuredeck. Fig. 3_54.On-blockoutfit_
ting ofanother superstructuredeckblock.Zone,
problem area, and stage are the same as rn
f i g u r e 3 - 5 3 . F i g . 3 - b 5 . G r a n d - b l o c kj o i n i n g o f
the aft half of the superstructure.

Fig. 3-55

Fig

Fig.3-57

Fig. 3-56.Erection, keel laying plus 11 workdays.

Fig. 3-57.Erection,keel laying plus 18 workdays.

PI.ATE V

Fig.3-58

trin ?-trO

Fig.3-60

F i g .3 - 6 1

Fig. 3-62

Fig.3-63

Fig. 3-58.Erection, keel laying plus 15 workdays.

Fig. 3-59.Erection, keel laying plus 19 workdays.
Fig. 3-60.Erection, keel laying plus 22 workdays.

Fig. 3-61.Erection,keel laying plus 22 workdays.

Fig. 3-62.Erection,keel laying plus 24 workdays.
Fig. 3-63.Erection,keel laying plus 24 workdays.

PLATE VI

Fig. 3-64

Fig.3-65

Fig.3-66

Fig.3-67

Fig. 3-64.Erection,keel laying plus 27 workdays.

Fig. 3-65.Erection,keel laying plus 28 workdays.
Fig. 3-66.Erection,keel laying plus 29 workdays.
Fig. 3-67.Erection,keel laying plus 29 workdays.
Fig. 3-68. Operation and test (vessellaunched
43 workdays after keel laying, delivered seven
months after starting fabrication).

Fig.3-68

PRODUCT.ORIENTED
WORKBREAKDOWN
STRUCTURE

ASPECTS
PRODUCT

PLAN'G M'F'G
LEVEL LEVEL

ZONE

AREA

85

CODES
ZONE AREA STAGE

STAGE

U
E
F

=
(J
=

F
a
E

IJJ

o
z

OPERATION
AND
TEST

z.

cr

U)

E
uJ

o-

-(t

o
z.

z
o
E

-at)

clo
uo
oo

'ad

IIJ

U'

Pz

Eg
?>

=-= =>j

322

NIL

z^
o-

oo=
>
- =Yru ,

=eiEEfi
=vg
H ; F=E
>89

uJ
att

U)

()
o
J

NIL

NIL

ONPEN.SPACEFITTING

WELDING

EH=EE5

ua*

624

HRU
-6q

ID

NIL

NIL

NIL

Fo>
B O

NIL

6>

z.
o

ge

gts

H8
<)fl
5
E
Ou)

8x
=9

z
?2.
oo

e
6
t
(,

d.4

>E

5
Q
Y=

Vr=

ASSEMBLY

)ut

8e

AQ
6=
E6

UNIT

JF

dr

d=
do

sMAu"slzE

s8
sU

Eg
E=
5e

LARGE.
SIZE
UNIT

9H
-

AC

OUJ

Y
E
OUJ

=o

NIL

WELDING

=
8
=o

JOINING

UNIT

?uJ

WELDING

WELDING

()
=8=e
=zu,
o

NIL

ON.CEILINGFITTING

( t x
.n=o
o?<)

WELDING

RTNNG
ON.FLOOR

NIL

FITTING
ON.CLOSED.SPACE

}e

z.r-

WELDING

G.

S u r { u r

3Z
q5

z
o
g,

z
l

='

llJ

z
o

=
o

(9
z.
u rE>
g

(,

z
,"aC
=F

o()

+<
-z z3

(ro
5-<
o=

(,

PALLETIZING

6U
.-O

z
U'
o
E
f

o-

Fig. 3-25.ZOFM classifrcationby product aspects.

MANUFACTURING
DESIGN
AND
MATERIAL
PREPAMTION

NIL

NIL

z
a
s- u)

-o
2<)
l--

zjtJJz
z.tlt

oi

>cr
6l
do

lrJ
2.4

*E
o
o

o
>
o- tu
t
o-

z
I

U'

o
z

86

SHIPPRODUCTION
PROBLEM
AREASUBDIVISIONS
PROBLEM
AREA

DESIGN
TO FURNISH

MATERIALTO BE FURNISHED

IN.HOUSE
MANUFACTURING

MANUFACTURING
DRAWING

YES

OUTSIDE
MANUFACTURING

MANUFACTURING
DRAWING

YES/ NO

PURCHASING

PURCHASE
ORDER
SPECIFICATION

SELDOM
/ NO

Fig. 3-26. Problem area subdivisionsfor designand material requirementsfor componentprocurement.

as if it were a single component.Two typical

standard machinery units are shown in Figure 3-28 (Plate I). They represent interim
productsfrom the sameproblem area despite
their physical differences.Figure 3-29 (Plate
I) showsanother typical machinery unit, utilizing a common foundation, and with machinery protected from the weather. Pipe
units are generally unique becausethey reflect the pipe passagesand details peculiar to
eachtype and/or sizeship, evenamongstandard series ships for different owners.A variety of out{it units are shown in Figures 3-30
through 3-33.
The grand-unit joining level providesfor
combining two or more units in order to reduce the working times needed for frtting
on-block and on-board,and to producemore
stable entities for erection purposes.Classification by problem area is limited to large
size unit or nil. Phasing by stageis:
. joining
. welding or nil
The welding stage applies only if there are
specialor extensivewelding requirements.
3.2.5. On-block Outfitting.Outfitting components, units, and gr:andunits are sometimes
frtted in a block zone defined for hull construction. However, when they are to be fitted to ceilings, blocks should be inverted
because fitting downhand enhances safety

and efficiency.Therefore,the outfrt zone for

a block set upside down encompasseseverything frtted to the ceiling. Following block
turnover, the outfit zone encompassesthe
components,units, and./orgrand units fitted
to the floor. Turnover representsa change
in stage. Specifying azone by stage for each
side suffices for absolute control ofon-block
outfitting.
Similarly, outfrt items should be fitted in
the zone of a double bottom block before its
tank top panel is installed. Then at a later
stage, a different outfrt zone encompasses
everything to be frtted to the tank top. The
primary goals of this manufacturing level are
to outfit ceilings and double bottoms when
blocks can be manipulated to provide ideal
access.
Typically, the divisions by problem area
address problems which are inherently different, so that each work packagefor outfitting on-block can be assignedto the appropriate team of assemblyspecialistsfor hull,
machinery,or superstructure.Theseclassificationsare further subdividedby the quantities of items to be fitted, resulting in the
following six problem area divisions:
. hull: large quantity or small quantity
. machinery: large quantity or small
quantity
. superstructure:large quantity or
small ouantitv

STRUCTURE
WORKBREAKDOWN
PRODUCT-ORIENTED

Fig.3-30

Fig.3-31

Fig.3-32

Fig.3-33

87

Fig. 3-30. Engine room outfit unit (courtesyAvondale Shipyard). Fig. 3-31. A hatch cover and coaming
unit. Fig. 3-32.Very complexpiping units, for the deck ofa product tanker (courtesy Avondale Shipyard).
Fig. 3-33. Mast units.
When the items to be frtted compose a small
quantity per block, outfit work can be performed at the site where the block was assem-

bled. When a large quantity is planned, the

completedblock should be transferred to an
indoor or outdoor region designatedfor out-

88

SHIPPRODUCTION

frtting in accordancewith an on-flow concept.

On-flow meanswork packagesstart and complete in unison.
Separationby stageis in accordancewith
the following sequencewhich reflects block
turnover:
.
.
.
.

on-ceilingfrtting
on-ceilingwelding or nil
on-floor frtting
on-floor welding or nil

The welding stages apply only for special or

extensive welding requirements. On-ceiling
frtting and welding usually are optimum for
blocks. However, most on-floor fitting and welding take place after on-ceiling outfitting is
completed,blocksare turned over, and blocks
are joined to create grand blocks. In order to
simplifu the erection schedule and minimize
duration in the building dock, such on-floor
outfitting should include all grand units,
units, and componentsto the maximum extent possible.Figures 3-34through 3-39show
on-block outfitting (see Plate II for figures
3-35 to 3-39). Outfitted blocks can be
erected in combination with outfit units
(as shown in Figure 3-36).On-blockoutfitting
work is often greatly facilitated by block
turnover for downhand on-ceilingoutfitting
and painting (seeFigures 3-37 and 38). Complex blocks may be outfrtted together to ensure accuratefits at erection.

Fig. 3-34. On-block outfitting in a pipe tunnel

under a cargohold.

3.2.6. On-boardOutfitting.Outfitting on-board

may seem at first to be the same as conventional outfrtting. However,the work required
is susceptible to the same analyses as for
on-unit and on-block outfitting. As a consequence, zonelproblem area./stagecontrol is
applicable. Much on-board outfit work progressessimultaneouslywith hull erection,as
shown schematically in Figure 3-b. Ideally,
outfrtting on-boardshould be limited to:
r frtting components,units, and./or
grand units that are too large or too
healy to fit on-block(e.g.,main engines,
dieselgenerators,most units and grand
units for engine room tank top, etc.)
o frtting fragile and weather-vulnerable
componentsthat could be damagedif
installed beforecompartmentsare
enclosed(joinery, insulation, electronic
equipment, etc.)
. connectingbetweencomponents,units,
and grand units that are either fitted
on-blockor on-board
One useful method of classifyingwork packages by problem area simultaneously addressesthe teams of specialistsneeded,work
volume sizes, and skill requirements. This
classificationis:
r similar work in small volume
o similar work in high volume
. high-skill requirements
These categoriesare applicable to hull, machinery, and superstructure.
Variety work in small volume should be
encompassedin an on-boardzone for execution by a team having the neededvariety of
skills. Variety work in large volume should
be divided by similarities in componentsand
units or sets of components and./or units.
Zonesfor such problem areas should not be
too long, wide, scattered,or otherwise unfavorable for executionand supervision ofwork.
At the same time, planners must regard the

--_-T_-_

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

need for high-skill frtting work required in

many ship compartments. In such cases,
large zonesgroupedby specifrcproblem areas
could be most beneficial. Stage for on-board
outfitting could be divided into:
.
.
.
.

open-space(blue sky) fitting

open-space(blue sky) welding or nil
closed-space
fitting
welding or nil
closed-space

The welding stagesapply only if there is special or extensivewelding to be done. Openspace frtting and welding should be completed before closures imposed by the continuing erectionofblocks in order to take full
advantage of easy access.Therefore, such
work should be incorporated in the erection
schedule. Closed-spacefrtting and welding
activities should be minimized as much as
practicable as they require more working
hours, more transportation services, and
longer durations (seeFigure 3-40,Plate II).
The on-boardoutfittinglevel useson-board
divisions as zoneswhich are subdivisionsof
the ship as a zone as for the erection level in
hull constrrrction work.
3.2.7. Operationand Iest The operation and
test level applies to work required to assess
the performanceof eachship'sfunctional systems. At this level, zone is the entire ship.
Problemsare groupedto match teams of specialists to the hull, machinery, and superstructure areas.
Operation and test are regarded as a
single stage.Thus, at this level, work is packaged by one or more systemswithin each of
the problem areas defrned for the specialist
teams. It is the traditional method for planning operation and test work.
3.3.ZonePaintingMethod
The ZonePainting Method (ZPTM) is a natural extension of the logic employed in both
HBCM and ZOFM. It transfers much paint-

89

ing work, traditionally performedin a building dock or at an outfit pier, to preceding

manufacturing levels by integrating painting with hull construction and outfrtting processes.Painting is treated as another assembly processaccomplishedat each manufacturing level, as shown in Figure 3-41. There
are certain prerequisitesfor successfulapplication of ZPTM:
o the painting interval between one coat
and a next coat must be shorter than
the allowable exposureperiod for the
former
. each hull block should be virtually
frnished in order to minimize surface
preparation and painting rework caused
by further cutting, fitting, and welding
. the shop primers applied to plates and
shapesshould not impede effrcient
cutting and welding
The aboverequires that hull, outfitting, and
painting planners work together to control
the durations between the shop primer and
primer levels, and between the primer and
frnish undercoatlevels.Additionally, managers must ensure effective accuracy control to
limit the need for surface preparation and
painting rework resulting from inaccurately
producedinterim products.
The main planning objectivesfor shifting
paint-related work to the manufacturing levels prior to on-boardpainting are to:
. shift positionsfrom overheadto downhand or at the minimum to vertical,
from high to low places,and from
confinedto readily accessibleplaces
. facilitate the use of temperature- and
humidity-controlled buildings, especially
for sophisticatedcoatings
. provide safer environmentswithout
extraordinary devicesthat would
encumberworkers
. prevent in-processrust and associated
rework

90

SHIPPRODUCTION

. minimize scaffoldsneededonly for

surfacepreparation and Painting
.level load work throughout the entire
shipbuilding processin order to avoid
large work volumes in the final stages
that couldjeopardize scheduleddelivery
Typical classification of paint-related work
packagesby their product aspectsis shownin
Figure 3-42.Horizontal combinationscharacterize the various types of work packages
that are requisite and suffrcient for the work
to be performed for each level. Vertical combinations denotethe processlanes for painting work flow. Obviously, there is need for
balancedplanning and schedulingand cooperation betweenhull construction,outfrtting,
and painting planners. Examples of paint
systems applied in accordancewith ZPTM
are containedin Figure 3-43.

3.3.1.Shop PrimerPainting.This manufacturing level applies to surface preparation and

application of shop primer to raw materials
beforethey are processedto createstructural
parts or outfrt components.Items which are
to be pickled after their manufacture are
usually excluded.Thus, useful divisions by
problem area are:
o plate
. shapesand other
The applicablestagecategoriesare:
. shot blasting
r painting
3.3.2. Primer Painting.This level is for application of an anticorrosive, including epoxy
and inorganic zinc-silicate,which is the first
coat applied to a componentor an on-board

Fig. 3-41.Zone painting method (ZPTM) manufacturing levels.

PRODUCT.ORIENTEDWORKBREAKDOWNSTRUCTURE 91

CODES

ASPECTS
PRODUCT

PLAN'G M'FG
LEVEL LEVEL

ZONE AREA STAGE

STAGE

AREA

ZONE

PAMNNG

NIL
U'

TOUCH.UP

(t
at9

CI..EANING

=
8 Oo ou
*ur
= 6 zfi
fg

HEFEE

D
6
oo

4)N

z
tr

PREPAMTION
SURFACE
PAINTING
AFTER
OVERTURNING

TOUCH.UP
AFTER
OVERTURNING

NIL

U'
r

ko

D
5
oo

e=H
HEEEE
CLEANINGAFTER
OVERTURNING

NIL

NIL

PREPARATION
SURFACE
AFTEROVERTURNINGN I L
PAINTING

HAHo o
o(ro

=tr
6g
z

tr

rcUCH-UP

OL

EE8
=gE

NIL

OVFRTIIRNING

=
L

NIL

PAINTING

Bs

(t)
ts

D
5
oo

PAINTING
E,

t!

z.

tr$
t5
62

Fig. 3-42. ZPTM classificationby product aspects.

o
U

HH

at

=fr
fg
z

z.
g

utlll
rO
iiio

=o
:z
+N

=
E

SHOT.BLASTING

O
()ou

oO
cro

SURFACEPREPAMTION

o
j
N

CLEANING

uJ
r

o-

TVFRTIIRNING

SURFACE
AFTER
PREPAMTION
OVEFTURNING

lL

CLEANING
PREPAMTION
SURFACE
PAIMNNG
NIL
AFTER
CLEANING
AFTER

U'

IU

92

SHIPPRODUCTION

MANUFACTURING
LEVEL

PAINT SYSTEM A

PAINT SYSTEM B

FINISH

COLOR

COLOR

FINISHUNDERCOAT

COLOR

COLORED PRIMER

PRIMER

PRIMER
NUMBEROF COATS
PER SPECIFICATION

PRIMER

SHOP PRIMER

SHOP PRIMER

S H O PP R I M E R

Fig. 3-43. Tlpical paint systems applied in accordancewith ZPTM.

division (as defined in ZOFM), or a block (as

defrnedin HBCM). Theseconstitute the zone
categories.Problem areas are grouped by:
r paint type (i.e., conventional,epoxy,
inorganic zinc-silicate,etc.)
. number of coats
. tYPeof zone
The latter further classifies each component,
block, or on-board division by problem area,
to anticipate:
. burn or wear damageof painted surfaces
during HBCM and ZOFM succeeding
manufacturing levels
. difficulty if there is a change in painting
conditions (e.g.,downhand to overhead,
low to high, spaciousto confined,etc.)
. need to maintain appearance
These considerations again demonstrate
that ZPTM, ZOFM, and HBCM planning
must be coordinated.Paint planners have to
consider the foregoing for each zone at all
ZOFM and HBCM manufacturing levels.
Stage at this level is separatedinto the following phases:
. surface preparation
. cleaning
r touch-up
o painting

. surfacepreparation after block turnover

or nil
. cleaning after block turnover or nil
. touch-up after block turnover or nil
o painting after block turnover or nil
The work at this manufacturing level is coordinated with ZOFM so that primer is applied
just before the on-ceiling fitting stage and,
following block turnover, just before the onfloor frtting stage. Nil applies to blocks that
are not turned over.
3.3.3. F/nish Undercoat Painting. This is the
semifrnal manufacturing level for paint application. Useful zoneclassificationsare:
. components(big in size or which
becomerelatively inaccessibleaft,er
fitting on-board,such as masts, cargo
booms,undersidesof hatch covers,etc.)
. units which are to be fitted on-board
r outfrtted blocks
o on-board divisions
. nil (applicableif epoxyis specifred)
Problem area divisions are:
e paint type
. number of coats
. type of zone(as describedin Part 3.3.2
for the primer painting level)
. scaffolding required only for paint

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

93

The classifrcationof work packagesby stage

is the same as for the primer level.

3.4. IntegratedHull Construction,

Outfitting,
and Painting

3.3.4. FinishPainting.Finish painting is the

final manufacturing level in ZPTM. Zone,
problem area, and stage classificationsare
the same as in the frnal undercoat level except that stages associatedwith block turnover are not applicable,and "nil" in the final
stagemeans a frnish coat will not be applied,
as in the caseofepoxy.

Integrated HBCM, ZOFM, and ZPTM for a

22,000dwt, multipurpose cargocarrier, which
is not one ofa standard series,is illustrated
in Figures 3-44through 3-68 (Figures 3-47 to
3-68, see Plates III-U). Typically, all work
shown was performed during a single eighthour shift per workday. Somenominal overtime was applied,usually for painting.

4. PipePieceFamilyManufacture
The integration of the HBCM, ZOFM, and
ZPTM representsthe application of the principles of group technology to shipbuilding.
Together they form a total shipbuilding
system.Group technology(GT) may also be
used to organizework within individual shops,
such as the pipe shop or machine shop. In
fact, these applicationsare closerto the traditional beginnings of GT. Pipe shop organization will be consideredas an example of the
application of GT to one particular shop in a
shipyard.[21
The engine room of a 22,000 dwt dieselpropelled ship contains about 3,600 pipe
pieces. The many differences among them
hide commonalitiesthat are useful for planning their manufacture. GT is used to systematically classifypipe piecesinto groupsor
families having design and manufacturing
attributes which are suffrciently similar to
make batch manufacturing practical. The processis called Pipe PieceFamily Manufacturing (PPFM).
The collectionof seeminglydifferent pipe
piecesinto suchfamilies avoidslaboriousjobshop type planning, scheduling, and manu-

progressionof developingpipe
stage-by-stage
pieces within such work flow lanes greatly
enhancesproduction control. Further, the separation by stages permits the switching of
work flow from one processlane to another
without diminishing control.
In PPFM the key zone is that which defines a planned pipe piece.It is an optimum
division of a pipe line, usually consistingofi
. cut pipe (including branch when
applicable)
. flanges
. elbows,sleeves,tees,etc.
A finished pipe pieceappearsin Figures 3-25
and 3-26 as a componentfrom the "in-house
manufacturing" problem area within the lowest ZOFM manufacturing level needed for
outfrtting on-unit, on-block,or on-board.Thus,
zone for PPFM is different from that for
HBCM, ZOFM, and ZPTM, in that it is not
derived from a hull block. Because of this
difference,PPFM is developedindependently
and is only problem area oriented, as distinguishedfrom the zoneorientationsof HBCM,
ZOFM, and ZPTM.

pieces
withina
facturing.
Instead,
different

A further
distinction
isthatPPFM
isa

family are designated for the same machines

and tooling setups, which are arranged in a
rationalized processlane. The manifestly clear

fabrication process,as compared to the assemblyprocesseswhich characterizeHBCM,

ZOFM, and ZPTM. Typically, PPFM is ap-

94

SHIPPRODUCTION

plied in the sevenmanufacturing levels presentedin Figure 3-69.The accompanyingproduct aspects in Figure 3-70 show that the
most important considerationsfor classifying
proposedpipe piecesby problem area must
take accountof similarities in:
e material specified(steel,copper,
polyvinylchloride,etc.)
e bore size
. shape(straight or bent)
. length
. other factors

Fig.3-44

{F

:*

Fig.3-45

Horizontal combinations of the product aspects characterize the various types ofwork
that are requisite and sufficient for the
work to be performed at each manufacturing level. Vertical combinations of the various work package types denote the process
lanes for pipe piece manufacturing. Typical
problem area subdivisionsfor only the pipe
fabrication, pipe piece assembly, and pipe
piecejoining levels are presentedin Figure
3-71.
The work packages,grouped by unique
similarities at all levels, facilitate modularization ofthe fabrication processesandjustification for expensive but highly efficient
facilities. Whether manual or automatic
fabrication methods are used, the number
and variety of pipe pieces needed for ships
justifres PPFM. Advantages are:
. less rearrangementof jigs and tools
r less variation in the work durations
and man-hours required among the
sametype work packages
o better accuracy
o signifrcantmanpower savings

Fig.3-46

4.1.WorkLots
Fig. 3-44. A center deck and center transverse
bulkhead grand block, upside down. Fig. 3-45. A
grand block, including the side shell, top side tank,
transverse hopper, and side transverse bulkhead
ofa cargo hold. Fig. 3-46. The same gtand block as
in frsure 3-45.

It is impractical to control PPFM with one

work packageper pipe piece,becauseeach is
a relatively small job and a wide variety and
large number of pipe pieces are needed per
ship. Controlby lot (i.e.,a group of work pack-

STRUCTURE
WORKBREAKDOWN
PRODUCT-ORIENTED

95

Fig. 3-69.Pipe piecefamily manufacturing method (PPFM) manufacturing levels.

ages) is far more effective, provided certain

principles are applied.
Within a processinglane the optimum
durations required for each work stage and
the times required for material handling between stages,when summed,yield the overall duration (Iead time) required for a processinglane. Thus, minimizing the time required for material handling between stages
also requires careful attention. Further, the
durations required to manufacture all pipe
piecesof a particular family should be equal
to each other. A lot should consist of a mix
from different families that will fully load a
facility during a designatedperiod, such as a
workday or a workweek. Thus, pipe pieces

which composea lot could be manufactured

in any random sequencewithout diminishing
control.One very effectivemanually operated
pipe shop posts a large calendar marked to
showa colorcodeassignedto eachwork week.
A dab of paint marks eachpipe piecein processwith a color assignedper work lot. Thus,
anyonecan readily determineifthe lot in processis on schedule.A lot should be sized by
evaluatingits productivity value (PV),particularly in considerationofits product resources.
4.2. MaterialReceiving
Material receivingis the preparation or frrst
manufacturing level, as shown in Figures
3-69 and 3-70.Problemarea is determinedby

sHtppRoDUcloN

96

PRODUCT
ASPECTS

PLAN'G M'F'G
LEVEL LEVEL

ZONE

AREA
o
1t
o<

PALLET

EIE
=9q

UJ

<vx

dEzi)

<fF

gtr

o-uJ

ZONE AREA STAGE

o==

Yul

STAGE

=O

CODES

3e$

PALLETIZING

o
o

o
z

il

(/)

I
U)

U
J

d<

COATING

NIL

PICKLING

NIL

COATING
PROCESS

()
o
=

k
o
o

o
z.

TEST
PROCESS

TESTING

NIL

U
o
r!
oU
I

PIPE
PIECE NIL

ut

FINISHING

PIPEMATERIAU
X-MYORNIU
BORSSTRAIGHT
ORBEMT/LENGTH

WELDING

BENDING
ON
PIPE
PIECE

PIPEMATERIAU
X-MYORNIU
MAINORBRANCII/
BORE/STMIGHT
ORBEI,IT/LENGTH

cuT

PIPE

NIL

PIPEMATERIAUPIPE
MAINORBRANCH
BORE

NIL

U
z.

z
E

z.

()
fl

c)
r
z

NIL

FINISHING

NIL

WELDING

NIL

ASSEMBLY

CUTPIPE
JOINING

NIL

MACHINING

NIL

BENDING
ON
CUTPIPE

U
z

5
('

JOINING
MARKING&
CUTTING

o
()

(,

d9
zf\
U=
Ud

ut
()
g
c
U
oo-

5
E
l
F

UJ

()
u
o
U
oo-

dro

E
5
o-z

NIL

MARKING
& CUTTING
U

MATERIAL

PIPE

PIPE
PIECE
PART

o
()

o
o
J

RECEIVING
uJ

Fig. 3-70. PPFM classificationby product aspects.

U
o

(r
ul

PRODUCT-ORIENTED
WORKBREAKDOWN
STRUCTURE

97

P I P EP I E C EJ O I N I N G
PIPE PIECEASSEMBLY
U

PIPE FABRICATION

MAIERIAL

X,RAY
O RN I L

MAIN OR
BRANCH

BORE

STRAIGHT
ON BENT

LENGTH

STRAIGHT
SMALL

BENT

STRAIGHT
MAIN

SHORT
LONG

MEDIUM

NIL

AENT

STEEL
STRAIGHT

SHORT
LONG

LARGE

U
E

EENT

o
BRANCH

X.RAY

NONFERROUS

PVC

OTHER

Fig. 3-71. Problem area subdivisionsfor pipe piecefabrication, assembly,andjoining levels.

type of material and shapein accordancewith

the following:
o pipe
. flanges,elbows,tees, sleeves,etc.
Stage is simply:
. material receiving
4.3. Pipe PieceFabrication
Pipe piece fabrication, the second manufacturing level, applies to processingofthe pipe
only. Problem area is grouped by:
r pipe material (steel,nonferrous,polyvinylchloride, etc.)
. main pipe or branch
. bore (small, medium, or large)

These problem area divisions and further

subdivisionsare incorporatedin Figure 3-72.
Stagesfor this level are:
o marking and cutting
. bending of cut pipe or nil
. machining or nil
Preferably, bending should be deferred to the
next manufacturing level becauseit is easier
to attach flanges or sleevesto straight pipe.
The bending stage at the pipe fabrication
level is only for thoseexceptionalcaseswhere
flanges would losetheir required orientations
during bending. The machining processapplies to the preparation of pipe ends for
welded or threaded joints. Typical PPFM
classifrcations and decision logic for determining pipe piecefamilies are shown in Fig-

98

sHrPPRoDUcroN

PPFM)
T Y P I C A LC L A S S I F I C A T I O NFSO R P I P E . P I E CFEA M I L YM A N U F A C T U R I N(G
PPFM
NO.

P I P EP I E C EC L A S S I F I C A T I O N S

01

Straight

S K E T C HO R R E M A R K S

50 mm

t-

04

20Omm

o7

250 mm

l1

Bent After Fabrication

50 mm
65 \

14

200 mm

ntc tested
z4

25

c lnso

HydrostalicTested

27

31

Plaslic

Bent By Heating

41

Bent Belore Fabrication

. 40 kgicm,

4 50 mm
65 !

44

51
I
I

Assembled

54

61
T

'---'--l

*--_l-

--__l

*-L
!

- - * I F - - - T - <

tr-_1

71

Threaded

77

Penetrations

F.q-

81

Heating Coils

;2

StainlessSteel

6l

NonJerrous

90

Urgent

91

Molded and adjusted

zilmm

- Lining
.Galvanizing

69

__J

4 50 mm

SpecialCoatjng

200 mm

4 200 mm

57
I

40 ko/cm2

---,,'-

J
-Tf-

J:L
.1
l-;

-ll-

-'--{

(M)'

92
93

Aluminum brass pipes

95

Unit Assembled

96

Long.term"

a7

Cast Steel

9!)

General

( M ) - E x c e p tf o r P P F MN o . 9 6

00

Molded and adjusted

(M)

(M)
r - r + t

( M )- E q u i v a l e ntto P P F MN o . 2 1 , 24,27 & 61

'(M) designatesmanually preparedpipepiece

drawings.
" Requireslongerthan averagetime.

Fig. 3-72. Typical classificationsfor PPFM.

l
r
f
L ' l - . - - t

WORKBREAKDOWN
STRUCTURE
PRODUCT-ORIENTED
M A N U F A C T U R I N GS Y S T E M

99

Altor-T,oatmont
Svstem

2
20@ & below

30

27

27

3t

29

29

33

35

35

26

a
Low Presuro
Pio6 in Gsnoral

30

34

g2

32

E E
30

30

23

23

2A

25

34

fi

31

4
4
n

29

27

27

a a

Attor-Treatment
Sysrem Code

Fig. 3-73.Typical decisionlogic for determining pipe piecefamilies.

Figures3-74
ures 3-72and 3-73,respectively.
through 3-76 show pictures oftypical PPFM
classifications.
4.4. Pipe PreceAssembly
At this manufacturing level, flanges,sleeves,
etc., are attached to cut pipe. The finished
assemblyis the zone for a pipe piece except
for main and branch subassemblieswhich

3.

are to be joined to create a branch pipe

piece during the next manufacturing level.
Problem area is subdividedas follows:
r X-ray test or nil
. short, straight or bent
o IonB straight
Stageis phasedin accordancewith the following sequence:

100

SHIPPRODUCTION

. cut pipe joining or nil

. assembling
o welding or nil
. finishing (grinding or machining) or nil
. bending or nil

Cut pipe joining is for producing a cut pipe

longer than a standard length (the break between short and long straight in the problem
area subdivision).The welding and finishing
stages for main and branch subassemblies

Fig.3-74(a)

Fis.3-74(b)

Fig.3-74(c)

Fig.3-74(d)
F i g . 3 - 7 4 . T y p i c a l P P F M c l a s s i f i c a t i o n s .( a )
Straight. (b) Straight. (c) Bent after fabrication. (d) Bent after fabrication. (e) Bent after
fabrication.

Fig.3-7a(e)