Type 45 Destroyer

TYPE 45
DESTROYER
Acquisition Options and Implications
John Birkler, John F. Schank, Mark V. Arena, Giles Smith, Gordon Lee
Prepared for the United Kingdoms Ministry of Defence

Approved for public release; distribution unlimited
R Europe
The research described in this report was prepared for the United
Kingdoms Ministry of Defence.
ISBN: 0-8330-3203-8
RAND is a nonprofit institution that helps improve policy and

decisionmaking through research and analysis. RAND is a
registered trademark. RANDs publications do not necessarily reflect
the opinions or policies of its research sponsors.
The cover art and Figures 1.1, 1.2, 4.4, and 5.1 are from the United Kingdoms
Ministry of Defence and are reprinted by permission. The photos in Appendix
C are by Tom Lamb. Figure C.4 is from Carlos Merino (IZAR Ferrol Shipyard),
F-100 and F-319: New Frigates for Europe, paper presented at RINA
Conference WARSHIP 2001Future Surface Warships, London, 2001.
Cover design by Maritta Tapanainen
Copyright 2002 RAND
All rights reserved. No part of this book may be reproduced in any

form by any electronic or mechanical means (including
photocopying, recording, or information storage and retrieval)
without permission in writing from RAND.
Published 2002 by RAND
1700 Main Street, P.O. Box 2138, Santa Monica, CA 90407-2138
1200 South Hayes Street, Arlington, VA 22202-5050
201 North Craig Street, Suite 202, Pittsburgh, PA 15213-1516
RAND URL: http://www.rand.org/
To order RAND documents or to obtain additional information,
contact Distribution Services: Telephone: (310) 451-7002;
Fax: (310) 451-6915; Email: order@rand.org
PREFACE
In April 2001, the United Kingdoms Ministry of Defence (MOD)

commissioned RAND to investigate procurement strategies that the
MOD could pursue as it acquires warships over the next 15 to 20
years. The research was intended both to inform MOD decisions
about its next-generation destroyer, the Type 45, and to help the
Ministry pursue its long-term warship acquisition programme. In
commissioning the research, the MOD asked RAND to evaluate nearterm and long-term strategies that would yield the highest value for
money, encourage innovation, allow for the efficient use of production capacity, and sustain the United Kingdoms (UKs) core warship
industrial base.
This book summarises RANDs analysis of acquisition options open
to the MOD during production of the Type 45 Destroyer. This destroyer, to be delivered from 2007, will become a main component in
the Royal Navys surface fleet, taking on roles as diverse as protecting
the fleet in littoral settings, participating in hostile engagements on
the open ocean, and conducting diplomatic and crisis-intervention
missions. The Type 45 also will constitute a large proportion of newship acquisitions that the MOD has planned for the next two
decades. As such, the acquisition and production techniques that
the MOD adopts for the Type 45 will be a bellwether for later acquisitions.
RAND quantitatively and qualitatively evaluated the effect on the
United Kingdoms shipbuilding labour force and infrastructure of
various options to acquire and build the Type 45. For their evaluations, RAND staff relied on data relating to the future demand for
iii
iv
The Royal Navys New-Generation Type 45 Destroyer
Royal Navy ships, to commercial work and Foreign Military Sales,

and to the existing capacities of the United Kingdoms shipbuilding
industrial base. The analysis results were a key input to the UKs decision on the Type 45 programme, as announced to the House of
Commons on 10 July 2001, by the Secretary of State for Defence, Mr.
Geoffrey Hoon.
This book should be of special interest not only to the Defence Procurement Agency and to other parts of the Ministry of Defence, but
also to service and defence agency managers and policymakers involved in weapon system acquisitions on both sides of the Atlantic
Ocean. It should also be of interest to shipbuilding industry executives in the United Kingdom. This research was undertaken for the
MODs Chief of Defence Procurement jointly by RAND Europe and
the International Security and Defense Policy Center of RANDs National Security Research Division (NSRD), which conducts research
for the U.S. Department of Defense, allied foreign governments, the
intelligence community, and foundations.
CONTENTS
Preface .........................................
iii
Figures .........................................
ix
Tables..........................................
xi
Summary .......................................
xiii
Acknowledgements ................................
xxi
Acronyms .......................................
xxiii
Chapter One
INTRODUCTION ..............................
Background and Study Objectives ..................
The Alliance Forms ...........................
The Alliance Dissolves .........................
The Domestic Industrial Context .................
RAND Is Asked to Step In .......................
Type 45 Destroyer ..............................
Sources of Evidence We Drew Upon to Meet
the Study Objectives ........................
Limitations on the Scope of the Study ...............
Organisation ..................................
Chapter Two
THE SHIPBUILDING INDUSTRIAL BASE AND THE MOD
SHIPBUILDING PROGRAMME ....................
Shipbuilding in the United Kingdom: 1900 to 1970 .....
Shipbuilding in the United Kingdom: 1970 to 2000 .....
1
2
2
3
3
4
5
7
8
9
11
12
13
vi
Royal Navy: Current Snapshot ....................

MOD Naval Procurement: Current and Future
Production Plans ...........................
Astute-Class Submarine ........................
Future Surface Combatant ......................
Future Aircraft Carrier (CVF) ....................
Landing Platform Dock (Replacement) [LPD(R)] ......
Future Offshore Patrol Vessels (FOPVs).............
Survey Vessels ...............................
Type 45 Destroyer ............................
RFA Procurements .............................
Alternative Landing Ship Logistics ................
Auxiliary Oilers (AO) ..........................
Roll-OnRoll-Off Vessels .......................
Planned Schedule of Future Ship Production ..........
Distribution of Ship Construction Across Shipyards .....
16
20
21
21
22
23
23
23
26
26
27
27
28
29
Chapter Three
IDENTIFYING AND ANALYSING MODS
ACQUISITION CHOICES .........................
Issues and Options .............................
Quantitative Effect of Using Two Shipyards ...........
Our Analytic Approach ..........................
Shipyard Labour Model ........................
Methodology ................................
Data ......................................
Specific Assumptions ..........................
Break-Even Approach .........................
31
31
32
36
37
39
41
42
43
Chapter Four
ANALYSIS OF PROCUREMENT ALTERNATIVES .......
Cases Examined in the Cost Analysis ................
Comparison of Labour Costs Across Strategies .........
The Reference Case: Sole-Source Production ........
Dual-Source, Competitive Procurement of Whole
Ships ....................................
Dual-Source, Competitive Procurement: Alliance
Proposal .................................
Comparison of Dual-Source, Competitive
Procurement with Sole-Source Procurement ......
Break-Even Analysis for Competition ..............
15
45
45
46
46
47
48
48
50
Contents
Dual-Source, Directed Buy of Whole Ships ..........

Dual-Source, Directed Buy of Ship Blocks ...........
Summary of Cost Analysis ......................
Other Factors Affecting the Type 45 Procurement
Decision .................................
Risks and Rewards to the Type 45 Programme .......
Risks and Rewards to Other Programmes ...........
Risks and Rewards to the UK Shipbuilding
Industrial Base ............................
Overall Integration of Analysis and Observations .......
Chapter Five
THE APPROVED TYPE 45 PROGRAMME .............
Chapter Six
ISSUES REQUIRING FURTHER STUDY ..............
What Is the Future Manpower Demand and Supply
Picture? ..................................
Is the Distribution of Ownership and Management
Responsibility Among the Shipyards, PCO,
and MOD Appropriate? ......................
Which Acquisition Strategies Are Most Viable for Future
Programmes?..............................
Number of Ships to Be Procured .................
Distribution of Other Business Across Available
Shipyards ................................
How Can Innovation Be Encouraged in the Absence of
Competition?..............................
How Can Core Industrial Base Capability Best Be
Sustained? ................................
What Are the Issues When Block Production Is
Performed Separately from Final Assembly? .......
vii
52
53
56
57
57
60
60
61
63
69
70
73
73
74
74
74
76
77
Appendix
A. SENSITIVITY ANALYSIS .........................
79
B. LONG-TERM IMPLICATIONS OF A LACK OF

COMPETITION ................................
85
C. BUILDING SHIPS IN BLOCKS AT MULTIPLE SITES:

IMPLICATIONS FOR THE TYPE 45 PROGRAMME ......
89
Bibliography .....................................
107
FIGURES
S.1. Summary Comparison of Cost Penalties of Six

Procurement Options ........................
1.1. Artists Rendering of the Type 45 Destroyer ........
1.2. Relative Size of the Type 45 Destroyer and the Type 23
Destroyer .................................
2.1. UK Shipyards Involved in Major MOD Shipbuilding
Projects2000 .............................
2.2. Royal Navy Warship Production Schedules by Ship
Class and Assigned Shipyard ...................
3.1. Increase in Total Man-Hours for Different
Symmetrical Learning Curves and Distributions of the
12 Type 45 Ships ............................
3.2. Frequency Distribution of Learning-Curve Slopes for
Competitive and Noncompetitive Programmes .....
4.1. Cost Penalties of Five Whole-Ship Procurement
Options...................................
4.2. Composition of Cost Penalties of Five Whole-Ship
4.3. Cost Changes Resulting from Competition in 31
Missile and Ship Programmes ..................
4.4. Type 45 Blocks .............................
4.5. Cost Penalties Associated with Four Procurement
Options...................................
4.6. Composition of Cost Penalties Associated with Four
4.7. Summary Comparison of Cost Penalties of Seven
ix
xix
6
7
17
29
34
36
49
50
51
54
55
55
56
4.8. Summary of Risks and Rewards of Alternative

Acquisition Strategies for the Type 45 ............
5.1. MODs Type 45 Block-Production Allocation .......
6.1. Estimated Future Demand for Direct Shipyard
Workers to Produce UK Military Ships ............
6.2. UK Commercial Shipbuilding Gross Registry Tonnage
Versus Year ................................
A.1. Cost Penalty for a Production Run of Nine Ships ....
A.2. Cost Penalty for the Alliance Strategy for Varying
Levels of Procurement........................
A.3. Competition Cost Penalty for the Alliance Strategy for
Varying Levels of Procurement .................
A.4. Cost Penalty with Different Learning Assumptions ...
A.5. Cost Penalty with Different CVF Workload
Assumptions at VT ..........................
A.6. Cost Penalty with Different Workforce-Productivity
Assumptions...............................
B.1. Relative Price Growth in Various Industries from 1989
to 2000 ...................................
B.2. Net Cost Resulting from a Lack of Competition .....
C.1. Piece-by-Piece Construction of a Small Craft .......
C.2. Assemblies Being Erected on Berth ..............
C.3. Typical Structural Block ......................
C.4. Spanish Frigate Structural-Block Breakdown .......
C.5. A Typical Grand Block, Comprising Seven Blocks ....
C.6. A Typical Ring Section ........................
C.7. Stern Grand Block at DANYARD in Denmark .......
C.8. Subcontracted Deckhouse Constructed in Poland for
Denmarks DANYARD ........................
62
64
71
72
80
81
81
82
83
84
86
87
91
92
93
94
95
96
101
102
TABLES
S.1. Data Used in RAND Shipyard Model .............

2.1. British New-Construction Shipbuilders
and Products...............................
2.2. Royal Navy Ships and Royal Fleet
Auxiliary Ships: 2002.........................
2.3. MOD Naval Procurement: Current and
Future Plans ...............................
3.1. Summary of Ship-Production Learning-Curve
Statistics ..................................
5.1. MOD Type 45 Work-Allocation Strategy
Announced 10 July 2001 ......................
6.1. MOD-Required PCO Actions for Type 45 Block
Construction...............................
xi
xvii
16
18
28
35
65
78
SUMMARY
The United Kingdom has several ship procurement programmes under way or planned for the future, including those for new attack
submarines, aircraft carriers, surface combatants, and auxiliary
ships. Among these programmes is the new-generation destroyer,
known as the Type 45. The United Kingdom intends to acquire up to
a dozen of these warships for the Royal Navy beginning in 2007.
These ships will be some of the largest surface combatants built for
the Royal Navy since World War II and will constitute a sizeable portion of the new surface warship acquisitions that the United Kingdom plans to make over the next two decades. As such, the acquisition and production techniques that the UK Ministry of Defence
(MOD) adopts for the Type 45 will be a bellwether for later acquisitions.
In recent years, the MOD has become interested in evaluating a variety of different strategies that it might employ to acquire warships. It
also has become increasingly aware of excess industrial capacity in
the UKs warship-building sector. As a result, the MOD in 2001 asked
RAND to help it (1) analyse the costs and benefits of alternative
acquisition paths and (2) evaluate near- and long-term strategies that
would yield the highest value, encourage innovation, use production
capacity efficiently, and sustain the UKs core warship industrial
base, given other current and future MOD ship programmes, with
particular reference to the Type 45.
Done quickly at the request of the Chief of Defence Procurement
(CDP), this analysis entailed a quantitative comparison of the advantages and disadvantages of having one or two shipbuilding compa-
xiii
xiv
nies produce the Type 45 over the next 15 years. Options examined
include employing either whole-ship or modular, so-called block,
production techniques in a single shipyard or multiple shipyards.
The analysis aimed to help MOD policymakers in two ways: first, to
gain an understanding of the costs and benefits of different Type 45
acquisition and production strategies; and second, to gauge the effect of those strategies both on the United Kingdoms shipbuilding
industrial base and on the costs of other current and future MOD
ship programmes.
HOW CAN MOD MOST EFFECTIVELY MAINTAIN A

COMPETITIVE PRODUCTION ENVIRONMENT?
The MODs stated policy is to make maximum use of competition
whenever possible, based on the theory that competition promotes
innovation, reduces costs, and sustains core industrial capabilities.
Therefore, MOD generally views acquisition and production strategies that allow for robust, sustained competition in a more favourable light than alternatives that rely on less competitive arrangements. However, in some situations (detailed in Birkler et al., 2001),
the use of multiple, competitive sources can incur some penalties,
especially in the near term. Careful, situation-specific analysis is
needed to identify the best course of action.
The Ministrys original conception for competitive acquisition of the
Type 45 called for the involvement of two companies, BAE SYSTEMS
Marine1 and Vosper Thornycroft (UK) Limited (VT),2 in manufacturing the destroyer. Both companies were to share in producing
the first three ships and compete to build subsequent batches of the
vessel.
______________
1 BAE SYSTEMS Marine is a subsidiary of BAE SYSTEMS, a UK-based company that
designs and manufactures civil and military aircraft, surface ships, submarines, space
systems, radar, avionics, communications, electronics, guided-weapon systems, and a
range of other defence products. Formed by the merger of British Aerospace and Marconi Electronic Systems in November 1999, BAE SYSTEMS employs more than 100,000
people worldwide and has annual sales of some 12 billion.
2 Vosper Thornycroft is a UK-based company that focuses on design, development,
and production of warships, and commercial and military marine electronic controls;
and provides management, technical, training, and educational support services to
military and civil markets. Together, these activities employ 7,500 people in a network
covering the UK, Europe, the United States, and the Middle East.
Summary
xv
This strategy required BAE SYSTEMS Marine and VT to cooperate in

designing and producing the first three ships. Their cooperation, referred to as the Alliance, succeeded initially, then foundered when
prices and terms acceptable to all parties could not be agreed on.
The strategy was further challenged in December 2000, when BAE
SYSTEMS Marine submitted an unsolicited proposal to the MOD to
build all 12 Type 45 Destroyers. BAE SYSTEMS Marine argued that
producing the destroyers in its own shipyards would enable it to
deliver the ships faster and at lower cost than it could through a
joint-production arrangement with VT.
The MODs original joint-production strategy and BAE SYSTEMS
Marines subsequent sole-source proposal came about in a domestic
industrial context characterised by excess commercial and naval
shipbuilding capacity. Over the past 20 years, shipyards in the
United Kingdom have seen demand drop as orders for commercial
ships have gone to lower-cost foreign competitors. The result: almost every significant UK shipyard has at one time or another seen
its capacity exceed demand.
RANDS THREE INTERWOVEN QUESTIONS

RAND researchers qualitatively and quantitatively addressed three
interwoven questions:
One producer or two producers? Should the MOD have the Type
45 built by one company or by two?
Competitive or directed allocation of work if two producers?

Should the MOD compete the 12 ships in the Type 45 class,
recognising that competition may result in a single producer, or
should it directly allocate work to specific shipbuilders to ensure
that both producers stay involved with the programme?
Whole-ship or block production? Should the company or companies producing the Type 45 construct the destroyer in its entirety in one shipyard or assemble it from segments or blocks
produced in several shipyards?
Answers to these questions have implications not just for the Type 45
but also for other MOD warship acquisition programmes, for a num-
xvi
ber of UK shipyards, and for the UKs pool of skilled shipyard workers. Depending on how the MOD decides to acquire, produce, maintain, and repair the Type 45, one or more UK shipbuilders could see
their revenues drop, causing the closure of one or more shipyards.
HOW RAND STUDIED THE PROBLEM

RAND researchers developed an analytic model of the UK shipbuilding industrial base specifically for this Type 45 study. The model
encompassed all current and future programmes at BAE SYSTEMS
Marine and VT shipyards and calculated workforce, overhead, and
investment costs.
Using as its baseline the option of having BAE SYSTEMS Marine
build all 12 Type 45s, the model displayed the relative-cost effects
of alternative procurement paths on the Type 45 and other programmes.
To assess the direct-cost consequences of spreading production between two firms, as would occur in a strategy involving continuing
competition (rather than concentrating it at one firm), we estimated
the labour costs, overhead rates, labour-force transition costs, and
learning improvements across several shipyards for a given shipbuilding strategy. We then separately estimated the likelihood that
competitive pressures would reduce costs enough to overcome the
cost penalties of distributing work among two producers. Thus, the
basic question is not how much money will be saved but, rather,
whether introducing an additional production source is a reasonable
strategy to pursue.
This model required RAND to collect extensive data of the type outlined in Table S.1.3 We collected this information from discussions
held during site visits with five UK shipbuilders: VT, BAE SYSTEMS
Marine, Swan Hunter, Appledore, and Harland and Wolff. Beyond
______________
3 Our study excluded costs of the weapon system and other Government-Furnished
Equipment, and the cost of common items and material purchased by the Prime Contract Office (PCO) and provided to the shipyards.
Summary xvii
Table S.1
Data Used in RAND Shipyard Model
Data
Description
Shipyard capacity
Steel throughput, docks, lifting capacity, outfittinga

berths, and the like
Workforce profile
Experience levels, age, productivity, costs for hiring and

training, termination costs, restrictions on hiring and
termination, current employment levels, and the use of
contract workers
Production
experience
Numbers and types of ships built over the past five years
(including commercial work)
Current and future

production
Current and anticipated production plans (by ship)
Workload projections
For each activity listed in the current and future

production plan, a listing of the labour profile by trade
and quarter. Further, these data included design and
development workload.
Wage rates
Wage rates for all the labour types
Burden rates
Overheads, General and Administrative, and profit rates

as a function of different site workloads
Investment levels
Fixed investments, such as facilities, necessary for a

particular programme or that investment required overall
aOutfitting includes installation of furnishings (e.g., desks and chairs) and

equipment (e.g., electrical systems and pipes).
RAND MR1486-TS.1
the shipbuilders, RAND researchers had similar discussions with

Type 45 and other MOD ship programme managers and their staffs.
These offices provided considerable supporting information and
data on their ship programmes.
WHAT RAND FOUND OUT

The sole-source option of building all 12 ships in one yard is calculated to be less costly than a direct-allocation strategy based on
capital investment required, labor productivity, work process, and
overhead savings. A direct-allocation strategy would involve dividing
the total production between two shipyards in some predetermined
way. We studied three possible distributionseight, six, or four
shipsto BAE SYSTEMS Marine, with the remaining ships going to
xviii
VT. These cases resulted in an increase in programme costs of 10 to

13 percent over the baseline of BAE SYSTEMS Marine building all 12
ships (see Figure S.1).
A competitive strategy is similar to a direct-allocation strategy in that
the total production would be divided between the two shipbuilders.
However, there are two important differences. First, the final
distribution of work is unknown; that is, any of the distributions
studied is possible. Second, competitive forces would reduce the
price of the ships over those for a direct-allocation strategy. The
question is, Could competitive forces overcome the 10 to 13 percent
increase over the sole-source case? To answer this question, we next
estimated the likelihood that competitive forces sustained
throughout the production programme could overcome these cost
increases. Historical data, collected on 31 ship and missile programmes that involved head-to-head competition on some part of
the production phase, suggest an approximately even chance that
competition would lower production costs by at least 13 percent.
Therefore, we estimate roughly an even chance that competitive
production of the Type 45 at two shipyards would yield about the
same overall cost as sole-source production at one shipyard. We can
give no definite answer on whether competitive or sole-source
production would most likely lead to lower costs for the projected 12ship production programme.
Even with competition, there is the chance that one shipbuilder
would end up building most of the ships or all of the last competitive
batch. An alternative acquisition strategy to ensure that both shipbuilders would remain in the UK warship-building industry is to direct a certain number of ships, or portions of the ships, to each of the
two shipbuilders. Directed buys of whole ships to the shipbuilders
result in the 10- to 13-percent cost increases for the competitive options, but without the potential for lower costs due to competition.
However, allowing each shipbuilder to build the same sections, or
blocks, of all 12 ships not only keeps both companies involved in
building warships but also takes maximum advantage of the lower
production man-hours due to learning.4
______________
4 We recognise that this strategy might not preserve the shipbuilding skills unique to
the blocks produced at the other shipyard.
Summary xix
RAND MR1486-S.1
14
Cost penalty (%)
12
10
8
6
4
2
0
12 to BAE
SYSTEMS
Marine,
0 to VT
8 to BAE
SYSTEMS
Marine,
4 to VT
6 to BAE
SYSTEMS
Marine,
6 to VT
4 to BAE
SYSTEMS
Marine,
8 to VT
Block
Option I
Block
Option II
Allocation strategy
Figure S.1Summary Comparison of Cost Penalties

of Six Procurement Options
We examined two different options involving the distribution of the

various blocks to the different shipyards. These two block options
resulted in cost increases of approximately 4 percent over the solesource option of having BAE SYSTEMS Marine build all 12 ships.
RANDS FINDINGS ALREADY HAVE BEEN PUT TO USE

The analysis described here was presented to senior MOD leadership
during the first half of June 2001 and was a key input to the United
Kingdoms decision on the way forward. On 10 July, the Secretary of
State for Defence, Mr. Geoffrey Hoon, announced to the House of
Commons that BAE SYSTEMS Marine and VT would jointly build
Type 45 destroyers in blocks. Mr. Hoon subsequently announced, on
18 February 2002, that the MOD had made a contractual commitment with the prime contractor for a further three ships (in addition
to the three already on order) and that the prime contractor had, in
turn, committed with BAE SYSTEMS Marine and VT for their work on
the first six Type 45 platforms in the planned class of up to 12.
This solution carries a number of advantages. Spreading the Type 45
work between BAE SYSTEMS Marine and VT helps ensure that both
xx
shipbuilders will remain viable and able to compete on future MOD

programmes such as the Future Surface Combatant (FSC). Having a
single shipbuilder construct all of the same blocks for the Type 45
class takes maximum advantage of the learning effect and, therefore,
reduces costs below those of distributing the construction of complete ships between the two shipbuilders.
Moreover, the revised Type 45 strategy will allow VT to proceed with
plans to invest in a new shipbuilding facility within the Portsmouth
Royal Naval Base. Since spring 2000, the company has been planning to shift shipbuilding operations to Portsmouth, but had delayed
committing to this move pending the MODs determination of VTs
role in the Type 45 programme. Shortly after the MODs decision, VT
announced that it would go ahead with the new facility, to be built
on four existing docks in the Portsmouth Royal Naval Base. VT has
signed a 125-year lease for part of the base. The lease will come into
effect shortly.
The MODs decision is not risk-free. Building ship blocks at multiple
sites has potential disadvantages. Constructing the blocks with
enough rigidity and weatherproofing to permit movement and
transportation will carry additional costs. Structural tolerances must
be managed very closely, since any misalignment of adjacent blocks
can lead to substantial rework costs. Also, block-transportation costs
might be higher than we assumed. Finally, scheduling of the
construction and delivery of the blocks must be closely managed. A
block that arrives late at the assembly yard may cause significant delays. We identified such risks but were unable to quantify their effect. On balance, we found no persuasive evidence to suggest that
the potential penalties of such risks outweighed the potential benefits of the block-construction strategy.
ACKNOWLEDGEMENTS
This work could not have been undertaken without the steadfast
support and encouragement we received from Sir Robert Walmsley,
Chief of Defence Procurement & Chief Executive, Defence Procurement Agency (DPA), and members of his staff. Many individuals
in the UK Ministry of Defence (MOD) provided their time, knowledge, and information to help us perform the analyses discussed in
this book. Their names and contributions would fill several pages.
If we were to single out two persons who participated in and supported this work in extraordinary ways, we would mention Charles
Draper, of the MOD, our action officer, and Brigadier Keith Prentice,
Integrated Project Team (IPT) leader for the Type 45 Destroyer. Their
tireless efforts on our behalf are greatly appreciated, and their
friendship is even more valued.
R Adm Nigel Guild, of the DPA, also deserves special mentionand
our gratitude. His insights and questions were highly valued and
caused us to think more broadly on the issues. Nick Witney, Director
General Equipment, graciously described the programming process
to us. These interactions added to the richness and quality of the
study.
We also want to thank the leadership and staff of MOD, especially Ali
Baghaei, Simon Rusling, Karl Monk, Cdre Dave Maclean RN, Terry
Baldwin, and Roger Russell, the IPT leaders for MODs current and
future ship and submarine programmes, whom we interviewed
during this study. As well, we thank Terry Johns, Group Pricing Manager, Land and Naval Systems; and Arthur Fisher, Cdre Neil Latham
RN , and John Hall, Warship Support Agency. They and their staffs
xxi
xxii
gave generously of their time and advanced our understanding of the

UK shipbuilding programmes and industrial capabilities.
We are also indebted to Brian Phillipson, then Managing Director,
Type 45 Prime Contract Office, BAE SYSTEMS; Nigel Whitehead, then
Managing Director, Astute Class Submarine Prime Contract Office,
BAE SYSTEMS; J. R. Wilson, Managing Director, Appledore
Shipbuilders Ltd.; Andrew Bunney, Managing DirectorShipbuilding,
Vosper Thornycroft (UK) Limited; Simon Kirby, then Managing
Director, BAE SYSTEMS Marine Division; Jaap Kroese, Chairman,
Swan Hunter; and Jim Gregg, Business Development Manager,
Harland and Wolff Heavy Industries Ltd. Each gave us the
opportunity to discuss a broad range of issues with those directly
involved. In addition, all of the firms arranged for us to visit their
facilities. The firms and government offices provided all the data we
requested in a timely manner.
Finally, for their careful and constructive comments on earlier drafts,
we wish to thank Charles Draper and Andy McClelland, MOD, and
RAND colleagues Frank Lacroix and Irv Blickstein. We are additionally indebted to Joan Myers for her deft assistance in organising and
formatting the many drafts, and to Marian Branch for her editing.
ACRONYMS
AAW
Anti-air warfare
ALSL
Alternative Landing Ship Logistics
AO
Auxiliary Oiler
AOR
Fleet Replenishment Ship
ASW
Anti-submarine warfare
CDP
Chief of Defence Procurement
CNGF
Common New Generation Frigate
CPC
Competing Prime Contractor
CVF
Future Aircraft Carrier
D&M
Demonstration and Manufacture
DD
Destroyer
DDH
Devonshire Dock Hall
DFM
Demonstration and First of Class Manufacture
DPA
Defence Procurement Agency (formerly the

Procurement Executive)
FF
Frigate
FOC
First-of-class
FOPV
Future Offshore Patrol Vessel
FPSO
Floating Production, Storage and Offloading vessel
xxiii
xxiv
FSC
Future Surface Combatants
FSED/IP
Full-Scale Engineering Development and Initial

Production
FSL
Fleet Support Limited
G&A
General and Administrative
GEC
General Electric Corporation
GFE
Government-Furnished Equipment
GRT
Gross registry tonnage
HM&E
Hull, Mechanical and Electrical
IPT
Integrated Project Team
JCB
Joint Capability Board
LBP
Length Between Perpendiculars
LCU
Landing Craft Utility
LCVP
Landing Craft Vehicle Personnel
LNG
Liquid nitrogen gas
LPD
Landing Platform Dock
LPD(R)
Landing Platform Dock (Replacement)
LSL
Landing Ship Logistics
MCMV
Mine Countermeasures Vessel
MIL
Marine Industries Limited
MOD
Ministry of Defence
MOU
Memorandum of Understanding
NAPNOC
No Acceptable Price/No Contract
OH
Overhead
OPV
Offshore Patrol Vessel
PAAMS
Principal Anti-Air Missile System
PCO
Prime Contract Office
Acronyms
PFI
Private Finance Initiative
R&D
Research and Development
RAF
Royal Air Force
RFA
Royal Fleet Auxiliary
RN
Royal Navy
RoRo
Roll-onRoll-off ship
SDR
Strategic Defence Review
SSA
Shipbuilders and Shipbreakers Association (UK)
SSBN
Nuclear-powered ballistic-missile submarine
SSK
Conventional submarine
SSN
Nuclear-powered attack submarine
UK
United Kingdom
U.S.
United States
VLCC
Very large crude carrier
VSEL
Vickers Shipbuilding and Engineering Limited
VT
Vosper Thornycroft (UK) Limited
xxv
Chapter One
INTRODUCTION
Between April and September 2001, RAND researchers analysed options open to the Ministry of Defence (MOD) in the United Kingdom
to acquire and produce the Royal Navys (RNs) next-generation
destroyer, the Type 45. Done quickly at the request of the Chief of
Defence Procurement (CDP), this analysis entailed a quantitative
comparison of the advantages and disadvantages of having either
one or two shipbuilding companies produce the Type 45 over the
next 15 years. Options examined included employing either wholeship or modular, so-called block, production techniques, either in a
single shipyard or in multiple shipyards. The analysis aimed to help
MOD policymakers in two ways: first, to gain an understanding of
the costs and benefits of different Type 45 acquisition and
production strategies; and, second, to gauge the effect of those
strategies both on the United Kingdoms shipbuilding industrial base
and on the costs of other current and future MOD ship programmes.
The MODs policy is to pursue competition in defence procurement,
at either the prime-contract or subcontract level, in order to secure
value for money. The MOD generally views acquisition and production strategies that allow for robust, sustained competition in a more
favourable light than alternatives that rely on less competitive arrangements. This preference is based on the premise that competition promotes innovation, reduces costs, and sustains core industrial
capabilities. However, in some situations (detailed in Birkler et al.,
2001), the use of multiple, competitive sources can incur some
penalties, especially in the near term. Consequently, careful, situation-specific analysis to identify the best course of action is needed.
BACKGROUND AND STUDY OBJECTIVES

The MOD has several future ship procurement programmes, including plans for new attack submarines, aircraft carriers, the Type 45
Destroyer, and auxiliary ships. MODs plans call for the Royal Navy
to take delivery of up to 12 of the destroyers from 2007. The MOD
has attempted to make the acquisition of these ships a competitive
process. BAE SYSTEMS is the Prime Contract Office (PCO), with
overall responsibility for the design and manufacture of the Type 45s.
The original conception of the Type 45 warship acquisition process
developed with the PCO called for two companies, BAE SYSTEMS
Marine1 and Vosper Thornycroft (VT),2 to be involved in the
production of the destroyer. These two competitors were to share in
the production of the first three ships: Each would build one
complete Type 45, and each would build one-half of the third, which
would be assembled at BAE SYSTEMS Marine. For the planned next
batch of three ships, competition for assembly would be limited to
BAE SYSTEMS Marine and VT; the loser would have the option of
assembling one of the three ships at the winners price.
The Alliance Forms

This strategy required BAE SYSTEMS Marine and VT to operate together in the design process, ensuring that the ship was buildable by
both shipbuilders. Initially, BAE SYSTEMS Marine and VT formed an
Alliance for the production of the Type 45. The Alliance did pro______________
1 BAE SYSTEMS Marine is a subsidiary of BAE SYSTEMS, a UK-based company that
designs and manufactures civil and military aircraft, surface ships, submarines, space
systems, radar, avionics, communications, electronics, guided-weapon systems, and a
range of other defence products. Formed by the merger of British Aerospace and Marconi Electronic Systems in November 1999, BAE SYSTEMS employs more than 100,000
people worldwide and has annual sales of some 12 billion. One condition of the
merger was that a clear firewall be established between BAE SYSTEMS and BAE
SYSTEMS Marine, to preclude any bias or favouritism in BAE SYSTEMs award of
shipbuilding contracts in a competitive environment.
2 Vosper Thornycroft is a UK-based company that focuses on design, development,
and production of warships and paramilitary craft ranging in size from destroyers and
frigates to inshore patrol craft, and commercial and military marine electronic controls; and provides management, technical, training, and educational support services
to military and civil markets. Together, these activities employ 7,500 people in a network covering the UK, Europe, the United States, and the Middle East.
Introduction
duce a price to the prime contractor.3 However, the Alliance

foundered because, among other things, the shipbuilders could not
agree on prices and terms acceptable to the prime contractor. The
prime contractor was also concerned about taking on more of the
risk and responsibility for the design and build than was originally
envisioned.
The Alliance Dissolves

Late in 2000, BAE SYSTEMS Marine informed the MOD that it had
concluded from a recently completed ten-year business plan that its
three shipyards (Barrow-in-Furness, Scotstoun, and Govan) could
not remain viable without increased throughput and restructuring.4
In December 2000, BAE SYSTEMS Marine sent the Type 45 PCO an
unsolicited proposal to build all 12 Type 45 Destroyers as currently
designed, with participation by VT. However, the design would be
optimised for build at BAE SYSTEMS Marine yards. BAE SYSTEMS
Marine argued that by producing the destroyers in its own shipyards,
it could deliver the ships faster and at lower cost than under the
Alliance strategy.5
The Domestic Industrial Context

Both the Alliance strategy and the unsolicited proposal came about
in a domestic industrial context characterised by excess commercial
and naval shipbuilding capacity. Over the past 20 years, shipyards in
the United Kingdom have seen demand drop as commercial orders
have gone to lower-cost foreign competitors and fewer naval ships
have been ordered. The result: almost every significant UK shipyard
has at one time or another seen its capacity exceed demand.
______________
3 In the United Kingdom, the development of particular weapon systems generally falls
to a PCO, which is responsible for the entire programme from design through
production and, more recently, sometimes a portion of the in-fleet support. This
acquisition strategy reduces the risks borne by the MOD. For the Type 45, the PCO is a
part of BAE SYSTEMS, the parent company of BAE SYSTEMS Marine.
4 BAE SYSTEMS Marine now manages Govan and Scotstoun as one Clyde facility.
5 In addition to the 12 Type 45 ships, the unsolicited proposal included other MOD
contracts at other BAE SYSTEMS Marine sites.
Such excess capacity, along with a string of corporate mergers, has

changed the face of the UK shipbuilding industry, particularly with
respect to naval shipbuilding. The building of Royal Navy vessels,
which in years past involved numerous small and large shipyards, is
today concentrated at two producers, BAE SYSTEMS Marine and VT:
BAE SYSTEMS Marine has some 90 percent of the naval shipbuilding capacity and owns three shipyardsGovan and Scotstoun, in Clyde, Scotland; and Barrow-in-Furness, in northwestern England. Barrow-in-Furness is the sole UK facility for
building nuclear-powered submarines, although substantial new
refueling facilities are being completed at Devonport, in
southern England.
Vosper Thornycroft has supplied 270 warships to 34 countries

over the past 30 years (including mainly smaller warships for the
Royal Navy).
Over the past ten years, several other commercially oriented shipyardsincluding Swan Hunter, Harland and Wolff, Appledore, and
Cammell Lairdhave competed to produce military or auxiliary vessels. The repair shipyards at Devonport, Rosyth, and Portsmouth
have not undertaken new construction for many years, but they play
the major role in the approximately 300-million annual ship refit
and repair programme.
RAND Is Asked to Step In

Faced with different acquisition approaches for the Type 45 (and
aware that the approaches had longer-term implications), and with
excess industrial capacity, the MOD asked RAND to help it analyse
the costs and benefits of alternative acquisition paths. In providing
such assistance, RAND qualitatively and quantitatively addressed
three interwoven questions:
One producer or two producers? Should the MOD acquire Type

45s built by one company or two?
Competitive or directed allocation of work if two producers?

Should the MOD compete some or all of the 12 ships in the Type
45 class, recognising that competition may result in a single pro-
Introduction
ducer, or should it directly allocate work to specific shipbuilders

to ensure that those producers stay involved with the
programme?
Whole-ship or block production? Should the company or companies producing the Type 45 construct the destroyer in its entirety in one shipyard or assemble it from segments, or blocks,
produced in several shipyards?
The answers to these questions have implications not just for the
Type 45 but also for other MOD warship acquisition programmes, for
a number of UK shipyards, and for the UKs pool of skilled shipyard
workers. Depending on how the MOD decides to acquire, produce,
maintain, and repair the Type 45, one or more UK shipbuilders could
see their revenues drop, thereby threatening the future of one or
more shipyards.
TYPE 45 DESTROYER
The Type 45 will be one of the largest UK ship programmes in recent
history (see Figure 1.1). First-of-class, HMS Daring, is due to enter
service in 2007. The total cost for the first six ships, including past
development costs and UKs share of the separately procured Principal Anti-Air Missile System (PAAMS), is some 4.3 billion. The MOD
plans to buy up to 12 of the destroyers.
The Type 45 will replace the Royal Navys current anti-aircraft destroyer, the Type 42, which is ageing and costly to operate. It will be
longer and wider, and will have a larger displacement than the Type
42. As shown in Figure 1.2, it also will have almost twice the displacement of the most recent surface combatant to enter the UK
fleet, the Type 23. This larger displacement limits where the Type 45
can be assembled and launched.
This new destroyer is designed for multiple roles.
On the one hand, it will act as the backbone of the Royal Navys antiair warfare (AAW) force-protection capability through 2040. In this
capacity, its primary mission will be to help UK forces control
airspace during joint operations in littoral theatres. Its top speed of
29 knots, its sophisticated array of PAAMS surface-to-air missiles,
Figure 1.1Artists Rendering of the Type 45 Destroyer
and its SAMPSON radar give it much more robust AAW capabilities
than its predecessors, and its ability to operate a Merlin helicopter
gives it improved antisubmarine warfare (ASW) capabilities.
On the other hand, the Royal Navy also intends to task the Type 45
with protecting maritime trade routes and shipping, and to conduct
open-ocean warfare. The Type 45 must be able to carry out these
missions in the presence of submarine and surface threats and be interoperable with NATO and Allied units. Moreover, the MOD intends
to deploy the Type 45 worldwide in support of broad British interests,
ranging from defence diplomacy to disaster relief to crisis intervention.
Introduction
Figure 1.2Relative Size of the Type 45 Destroyer and the Type 23

Destroyer
A 1.2-billion contract for the Demonstration and First of Class

Manufacture (DFM) of the Type 45 was let in December 2000 to BAE
SYSTEMS PCO. The contract covered the design, development, and
delivery of the first three ships, together with elements of support for
the first three ships of the class. The contract assumes a worksharing arrangement between BAE SYSTEMS Marine and Vosper
Thornycroft to ensure the credibility of future competition in
subsequent batches.
SOURCES OF EVIDENCE WE DREW UPON TO MEET THE

STUDY OBJECTIVES
Our analysis was based on four main sources of information, as
follows:
First, the prime contractor teams for Type 45 and Astute programmes6 provided their own (proprietary) estimates of devel-
______________
6 Data on the Astute programme were needed to understand the total demand for
labour and facilities at Barrow-in-Furness, the BAE SYSTEMS Marine shipyard that
constructs the Astute submarines.
opment and production costs, at a level of detail that enabled us

to estimate costs under different production scenarios.
Second, several MOD authorities provided their own cost estimates and/or overall programmatic information on past and
projected programme schedules, production quantities for each
ship class, etc. These sets of quantitative data were augmented
by extensive discussions with the PCOs for the Type 45 and for
other warships, and with MOD staff and shipyards regarding the
feasibility and desirability of various competition strategies.
Third, each of the shipyards provided extensive historical programme and cost data for all the classes of ships they had produced. They also shared with us their estimates of the costs for
future ship programmes.
Fourth, we conducted a literature review on the results of prior

efforts to introduce competition to a weapons-production programme. Because of the short time available for this study
(about six weeks from go-ahead to first Type 45 briefing), our
analysis of the historical record on the effects of competition on
production cost drew on previous studies of production-cost
changes due to competition.7 A more extensive and thorough
review of UK acquisition programmes would have been desirable; we have no evidence that a useful body of additional data
could be developed.
LIMITATIONS ON THE SCOPE OF THE STUDY

Competition is widely expected to stimulate a variety of actions by
the producers in an attempt to make their product more attractive to
the buyer. This study was almost entirely focused on a single consequence of competition: changes in production costs. Several other
possible consequences of competition are briefly reviewed and factored into the overall conclusions, but only the costs of production
are examined quantitatively.
______________
7 The data on historical programmes are extensively described in Table 5.8 and the
associated text in John Birkler et al., Assessing Competitive Strategies for the Joint Strike
Fighter, Santa Monica, Calif.: RAND, MR-1362-OSD/JSF, 2001, pp. 5356.
Introduction
Unfortunately, RAND has no historical or analytic methods for directly estimating reductions in production cost due to the introduction of competition. Instead, we estimated the incremental costs of
introducing competition. Then, drawing on historical evidence, we
attempted to estimate the likelihood that competition would drive
down the costs enough to permit recovering those incremental costs,
thus allowing the government to at least break even on costs.
ORGANISATION
The book is organised in six chapters. Following this Introduction,
Chapter Two sets the overall context in which the Type 45 programme must be executed. It describes the current status of the UK
shipbuilding industrial base, and the current and future programmes
of the MOD. It then discusses how those programmes are currently
matched, or could be matched, to the shipbuilders that make up the
industrial base. Chapter Three details the choices facing the MOD
for the Type 45 programme and how we quantitatively and qualitatively analysed the various options. Chapter Four presents the results of our analysis of different strategies for distributing the workload for the Type 45 programme between the two shipbuilders. In it
we consider strategies that involve the construction of both complete
ships and major portions, or blocks, at different shipyards. Chapter
Five discusses the Type 45 decision that was made after our analysis
concluded. Chapter Six describes the implications of the Type 45
decision for future MOD shipbuilding programmes and the shipbuilding industrial base.
Several appendices provide more detailed descriptions of the material contained in the body of the book: Appendix A describes a sensitivity analysis of the assumptions made in the study. Appendix B
reports the long-term consequences of a lack of competition. Appendix C discusses the implications for the Type 45 programme of
building ships in blocks at multiple sites.
Chapter Two
THE SHIPBUILDING INDUSTRIAL BASE AND THE

MOD SHIPBUILDING PROGRAMME
Several factors constrain the options available to the MOD for producing the Type 45 Destroyer and other military ships. They include
the present status and condition of the shipbuilding resources in the
UK. Also, the Type 45 programme is but one element of a broader,
long-range programme to upgrade the combatant and support forces
of the Royal Navy. Any policies and decisions regarding the Type 45
programme must be formulated in recognition of that broader programme.
In this chapter, we first describe the past (first 70 years of the twentieth century) and present UK shipbuilding industry structure and capabilities, outline key economic and political forces affecting the industry, and provide a foundation for a subsequent discussion on how
different procurement strategies for the Type 45 programme could
affect future industry capabilities. We then present a brief overview
of the present fleet, summarise the near- and far-term MOD planning assumptions for Royal Navy ships, and outline the shipproduction schedule planned to support those requirements over
the next decade. This overview includes a more detailed description
of the Type 45 system and overall programme than that given earlier
in this book. We conclude the chapter by showing the current and
potential distributions of the MOD programmes to the various shipyards.
11
12
SHIPBUILDING IN THE UNITED KINGDOM: 1900 TO

19701
Whereas the United Kingdom today constructs less than 1 percent of
the worlds ships, the country has had a long connection with shipbuilding. Throughout the nineteenth century and the first two
decades of the twentieth century, UK shipyards produced from onehalf to three-quarters of the worlds shipbuilding output, including a
wide range of commercial and military ships.
World War I closed most export markets to British shipbuilders,
which never recovered that lost market after the war. Exports accounting for nearly 25 percent of UK shipbuilding production before
the war constituted only 8 percent of the industrys output after the
war. By 1918, the decline in British shipyards had accelerated, as
domestic ship owners started to go abroad for their ships. Briefly after 1918, the UK accounted for 50 percent of total world shipbuilding;
then, in the 1920s, the proportion and, more important, the absolute
total, dropped. Gross tonnage produced fell from over 2 million in
1920 to 0.5 million in the mid-1930s. As a result, many UK shipyards
closed during the 1930s.
Just before the start of World War II, British shipbuildings share of
total world output had dropped to 35 percent. After the war, Britains
share fell further, to 15 percent, even though world output more than
doubled. Apart from a very brief period in the late 1950s, the UKs
shipbuilding industry did not really recover. The high percentage of
UK building in 1950 reflected the time it took for European and, in
particular, Japanese shipbuilding to recover from World War II.
______________
1 Much of the material in this section and the next was prepared by Tom Lamb, P.E.,
EUR ING, Technical Associate, Innovative Marine Product Development, LLC, Ann
Arbor, Michigan. It was drawn from Anthony Burton, The Rise and Fall of British
Shipbuilding, London: Constable and Company Limited, 1984; and Robert J.
Winklareth, Naval Shipbuilders of the World: From the Age of Sail to the Present Day,
London: Chatham Publishing, 2000.
The Shipbuilding Industrial Base and the MOD Shipbuilding Programme
13
SHIPBUILDING IN THE UNITED KINGDOM: 1970 TO

2000
The first oil crisis, nationalisation, privatisation, and consolidation
characterised the UK shipbuilding industry during the last three
decades of the twentieth century.
The first Arab oil embargo in 1973 caused a worldwide crisis in shipbuilding, driving up fuel prices and reducing demand for ocean
transport of all kinds. The result of the lower demands was devastating to the UK shipbuilding industry, forcing several UK shipbuilders
close to bankruptcy.
This crisis coincided in the United Kingdom with a change of government. The incoming Labour government was committed to nationalising many industries, a commitment reflecting a view that
government had a major role to play in supporting industry. The
level of security enjoyed by workers had reached its highest point in
the late 1960s. The government decided that the best way to maintain an effective shipbuilding capability in Britain was to nationalise it.
But by the time the industry was nationalised and the structures were
in place for the new British Shipbuilders Corporation, the government was again poised to change. In 1979, the Conservatives again
took power. The newly nationalised industry was now faced with a
government that had the objective of returning shipbuilding to the
private sector as soon as possible.
Although the naval shipbuilding sector was relatively strong from
1980 to 1990, owing to Royal Navy requirements and some export
work, the UK commercial market had almost dried up. As a result,
the government sought to sell or close a number of shipyards. Closures included Cammell Laird, which had been sold by British Shipbuilders to Vickers Shipbuilding and Engineering Limited (VSEL;
now BAE SYSTEMS Marine), and a number of smaller shipyards.
Foreign management took over other shipyards. Kvaerner of
Norway, as part of the development of a large international
shipbuilding group, took on the Govan shipyard. Harland and Wolff
was taken on by a mix of management and Norways Olsen group.
14
Closures continued during the 1990s and included Cochrane in 1993,

Richard Dunston in 1994, and Yorkshire Dry Dock in 1998. Swan
Hunter, a much larger yard, closed in 1993 after it failed to secure a
major naval contract. However, residual demand prompted some of
the closed shipyards to reopen later in the decade. Swan Hunter was
reopened in 1995 by THC Fabricators, a Dutch firm, and now has secured commercial contracts, including the conversion of a Floating
Production, Storage and Offloading (FPSO) vessel, and a naval order
for two Alternative Landing Ships Logistics (ALSL). Yorkshire Dry
Dock reopened as part of the George Prior Group.
The industry had greater success in the repair and conversion market. Cammell Laird reopened and operated with this market focus
from 1993 to 2001, when it went into receivership. At the time of this
writing, Cammell Laird at Tyneside is owned by A&P; its yards on
Wearside and Teeside have closed.
The change in ownership of the three BAE SYSTEMS Marine shipyards is a prime example of the turmoil in the UK shipbuilding environment during the past two decades. In 1985, the Yarrow shipyard
on the Clyde (now Scotstoun) was purchased by GEC-Marconi, the
defence arm of the British General Electric Corporation (GEC). In
1986, Vickers Shipbuilding and Engineering Limited, owner of the
Barrow-in-Furness shipyard and previously a part of the state-owned
British Shipbuilders, was the object of an employee buyout. In 1988,
British Shipbuilders sold the Govan shipyard, formerly known as
Fairfields, to Kvaerner, a Norway-based conglomerate with several
shipyards located throughout the world. In 1994, GEC-Marconi
purchased VSEL and incorporated the Barrow-in-Furness and
Yarrow shipyards as Marconi-Marine. In 1999, Marconi-Marine
bought the Govan yard when Kvaerner announced its exit from the
shipbuilding industry. Finally, in September 1999, British Aerospace
bought GECs Marconi Electronic Systems and reorganised Barrowin-Furness, Govan, and Scotstoun as BAE SYSTEMS Marine.
Sold in 1985, Vosper Thornycroft (VT), the other major UK builder of
naval warships, also was originally part of British Shipbuilders. Over
the past three decades, VT has been successful in the small-warship
export market, and has been successful in building warships to destroyer size for the UK, the last being HMS Gloucester in 1986. To
undertake a project the size of the Type 45, VTs Woolston shipyard
15
near Southampton would have required upgrading. Rather than do

this, VT is planning to move its operation into the Portsmouth Royal
Naval Base and to invest in new shipbuilding facilities there. The
company has signed a 125-year lease for part of the base. The lease
will go into effect shortly. The Type 45 should provide a solid foundation from which the company can sustain its export business and
enter competitions for future naval programmes.
Several other shipyards, including Swan Hunter, Harland and Wolff,
and Appledore, have tended to rely primarily on building commercial ships but have recently entered or re-entered into competition
for naval auxiliaries or warships.
To summarise the current status of the UK shipbuilding industrial
base, only two UK shipbuildersBAE SYSTEMS Marine and VT
have the combination of physical facilities, management and labour
experience, and financial resources necessary to construct Type 45
ships.2 Three other shipyards are now engaged in building ships for
the MOD. Some key descriptors of these five shipyards are displayed
in Table 2.1. Their locations, along with those of other UK shipyards,
are shown in Figure 2.1.
ROYAL NAVY: CURRENT SNAPSHOT 3

The Royal Navy has more than 125 ships and submarines and 182
aircraft. It employs approximately 45,000 uniformed and 20,000
civilian personnel, and operates from three major bases:
Portsmouth, Devonport, and Faslane. Table 2.2 displays the composition of the fleet. There are 107 combatant ships; the remainder
are auxiliary and support ships. Destroyers and frigates make up
about one-third of the ships in the combatant fleet that have displacements in excess of 1,000 metric tons.
______________
2 Other shipbuilders, such as Appledore, Swan Hunter, and Harland and Wolff, could
also construct the Hull, Mechanical and Electrical (HM&E) for the Type 45 if investments were made in their facilities. However, they have limited experience with combatants of the size and complexity of the Type 45.
3 Much of the material in the remainder of this section was drawn from documents
provided by the MOD and from information on the following web sites:
http://www.royal-navy.mod.uk, http://www.mod.uk, and http://www.rfa.mod.uk/
index2.html.
16
Table 2.1
British New-Construction Shipbuilders and Productsa
Company
Shipyard
Appledore
Shipbuilders
BAE SYSTEMS
Marine
Clyde
Non-MOD Products
MOD Products
Irish Navy patrol boats,

Supply Vessels, Fishery
Protection Vessel
3 Offshore Patrol Vessels
(OPVs) for Royal Brunei
Navy
Offshore Supply Vessel,
miscellaneous module
work
Survey Ships
Barrow-inFurness
Harland &
Wolff
Holdings plc
Swan Hunter
Vosper
Thornycroft
(UK) Limited
Type 23 Frigates, various

design work
ALSL, Landing Craft Utility,
and Auxiliary Oiler
Astute submarines, LPD,
and Auxiliary Oiler
Drill Rig, Ferries
Reactivation of 4 Upholderclass submarines for sale

to Canada
2 RoRos
Oil Platform
Design only for Greek
Patrol Craft
2 ALSL
Minehunter, Future OPV
(FOPV)
Excludes the Type 45s that are currently on order.
RAND MR1486-T2.1
The Royal Navy also has access to a fleet of auxiliary and supply vessels. The Royal Fleet Auxiliary (RFA) is a civilian-manned fleet owned
by the MOD. Employing more than 2,000 civilians, the RFA has as its
main task supplying warships of the Royal Navy at sea with fuel,
food, stores, and ammunition. It also provides aviation support for
the Royal Navy, together with amphibious support and secure sea
transport for Army units and their equipment. The 20 ships in the
RFA fleet in 2002 were built to the rules of Lloyds Register
(compartmentation, damage control, habitability) and also meet the
standards of the Shipping Naval Acts of 1911 and of the Maritime and
Coastguard Agency. Since 1994, the head of the RFA has been a flag
officer coequal with other type commanders.
17
RAND MR1486-2.1
Swan Hunter
BAE SYSTEMS Marine
Govan
Clyde Scotstoun
Barrow
Harland
and Wolff
Appledore
Shipbuilders
Vosper
Thornycroft
Figure 2.1UK Shipyards Involved in Major MOD Shipbuilding

Projects2000
MOD NAVAL PROCUREMENT: CURRENT AND FUTURE

PRODUCTION PLANS
The MOD has plans to modernise the RN/RFA along several dimensions. The focus of this study, the Type 45, will be developed amid an
ambitious programme to bring a host of new-generation warships to
the fleet over the next quarter century.
18
Table 2.2
Royal Navy Ships and Royal Fleet Auxiliary Ships: 2002
Type/Class
Number
Tonnage
Fleet
Royal Navy
Aircraft Carrier
Destroyer
3
11
20,600
3,5603,880
Frigate
Landing Platform
Helicopter
Minehunter
Patrol (Coastal)
Patrol (Ice)
21
1
1
3,5004,200
11,060
21,758
25
16
1
450615
49
6,500
Patrol (Offshore)
Submarine (Fleet)
Submarine (Trident Class)
Survey Ship
7
12
4
4
9251,427
5,0005,208
15,900
2513,500

Fleet Tankers Small
Support Tanker
Fleet Replenishment Ship
Aviation Training Ship
Landing Ship
Forward Repair Ship
Roll-onRoll-off Vessel
3
4
4
1
5
1
2
11,522
40,87049,377
23,59133,675
28,080
5,7718,751
10,765
12,350
SOURCE: Information in this table was taken from Ministry of Defence,

Performance Report 1999/2000, Abbey Wood, England, Annex B: Force Structure.
NOTE: Tonnage definitions vary for different types of ships. For passenger ships,
the term is gross tonnage; for tankers and bulk cargo ships, the term is deadweight
tonnage; and for warships, the term is displacement tonnage. These are defined as
follows:
Gross tonnage: a measure of the total volume of enclosed spaces in the ship. The
volume-to-tonnage conversion is 100 cu ft/metric ton.
Deadweight tonnage: a measure of the total volume of the ship dedicated to
carrying cargo, converted to tons of seawater (35 cu ft/metric ton).
Displacement tonnage: the volume of water displaced by the hull beneath the
waterline, converted to tons of seawater (35 cu ft/metric ton).
RAND MR1486-T2.2
19
In July 1998, the MOD completed a Strategic Defence Review (SDR)

that reassessed Britains security interests and defence needs and
considered how the roles, missions, and capabilities of its forces
should be adjusted to meet new strategic realities. 4 For naval forces,
the review suggested continuing the shift in focus away from largescale open-ocean warfare towards a wide range of operations in
littoral areas, with a premium on versatility and deployability. This
direction represents a continuation of trends since the end of the
Cold War.
In addition to the continuation of several procurements already
planned (e.g., new Landing Platforms Dock, Astute submarines, and
Auxiliary Oilers), the SDR introduced several major new programmes, including the Future Aircraft Carrier. The SDR remains
the basis of general MOD policy, but the future procurement programme continues to evolve in line with the strategic environment,
financial imperatives, industrial developments, and new opportunities. A more extensive naval programme is now assumed, with the
addition of two new Landing Ships Logistics and a new class of Offshore Patrol Vessels.
Thus, in addition to the Type 45, ships currently on contract for
manufacture, but which have yet to enter service, include the following:
Two Landing Platforms Dock (Albion and Bulwark)
Two Auxiliary Oilers
Four Landing Ships Logistics
Two survey vessels
Three Astute submarines
Three Offshore Patrol Vessels.
A contract for a Strategic Sealift service, which will make available six
Roll-onRoll-off (RoRo) ships as required under a Private Finance
______________
4 Secretary of State for Defence, Rt. Hon. George Robertson MP, Strategic Defence
Review: Modern Forces for the Modern World, Defence White Paper, 13 July 1998,
available at www.army.mod.uk/servingsoldier/strategicdefrev.html.
20
Initiative (PFI) arrangement, is also expected to be signed shortly. In

a PFI arrangement, the service provider can make the ships available
for the generation of commercial revenue when they are not needed
by the MOD, thus delivering better value for money for the taxpayer.
Under the MODs new acquisition strategy, future programmes will
be focused on capability requirements rather than on specific numbers of platforms. MODs plans for future ship production depend
on many variables, not the least of which is the evolution of the
strategic environment. This caveat needs to be borne in mind where
the book addresses specific ship development and production programmes now being envisioned to satisfy the MODs future requirements. That said, at the time of this study, the MODs future procurement planning assumptions envisioned the following:
Two large aircraft carriers, each capable of operating up to a total

of 50 fixed-wing aircraft and helicopters from all three services at
one time
Two to three more Astute submarines
Up to 20 Future Surface Combatants (FSC; to succeed the current

Frigate force).
The following discussion of current and future plans concludes with

a more detailed description of the Type 45 programme, the primary
subject of the present analysis.5
Astute-Class Submarine6
The MOD plans to acquire a minimum of five new submarines,
known as the Astute-class submarine, to replace the Swiftsure-class
nuclear-powered attack submarines (SSNs). It will take delivery of
the first of these submarines in 2005.
Invitations to tender were issued in July 1994, and competitive bids
were received in June 1995. The MOD identified GEC-Marconi (now
______________
5 This discussion does not include the many different smaller craft, such as the Land-
ing Craft Utility (LCU) and Landing Craft Vehicle Personnel (LCVP), or other capabilities for which the solution might not require extensive shipbuilding work.
6 Formerly known as the Batch 2 Trafalgar-class submarine.
21
BAE SYSTEMS Marine) as its preferred bidder in December of the

same year. Following protracted negotiations, using the policy of No
Acceptable Price/No Contract (NAPNOC), a contract for the first
three submarines was placed with GEC-Marconi as the prime contractor and announced in March 1997.7 The contract put in place the
first whole-boat prime contract for UK nuclear-powered submarines.
The submarines are being built at BAE SYSTEMS Marines Barrowin-Furness shipyard.
The prime contract is for the design, build, and initial support of
three submarines. The prime contractor will undertake the support
task for a total of eight submarine-years.
Future Surface Combatant

It is intended that the FSC will succeed existing Type 22 and Type 23
Frigates. Although timing and numbers are yet to be determined, it
has been assumed that delivery of the first ship of this class will take
place no earlier than 2012. Because the defence review emphasised
that the Royal Navy needs to support expeditionary warfare and joint
operations in littoral waters, the FSC is expected to have a broader,
multifunction role than the primarily ASW missions that were envisioned for the latest versions of the Type 23 that it will replace. The
assumption is that this warship will be operationally versatile and
affordable, and capable of being deployed through life across the full
spectrum of defence missions. This operational adaptability calls for
ships that will be able to acquire capabilities and technologies
throughout their service lives, without recourse to traditional, nonoperational, major refit processes.
Future Aircraft Carrier (CVF)

The MOD is currently in the assessment phases of a programme to
produce two new aircraft carriers to replace the three existing HMS
Invincibleclass aircraft carriers. These ships are currently scheduled
to enter the Royal Navy inventory in 2012 and 2015.
______________
7 The NAPNOC policy requires that, for all noncompetitive procurements with an es-
timated value of 1M and above, an acceptable firm/fixed-price or maximumprice/target-cost incentive package be agreed on before a contract is placed.
22
Twice as large as the Invincible-class carrier, the CVF is expected to

be among the largest warships ever built in the UK. Initial estimates
are that the ships could be 300 metres (m) long, have displacement of
between 40,000 and 60,000 metric tons, and able to carry up to 50
aircraft. Each vessel is expected to have a manpower requirement of
about 1,200 personnel (which includes the Carrier Air Group), which
is roughly equal to that of the Invincible class.
It is intended that the two ships of the new class will replace the three
Invincible-class ships currently in service. This reduction is to be
achieved through modern build-and-support techniques (e.g., better
paints, more-durable piping) that will dispense with the need for
long refit periods and will allow required availability to be achieved
from only two hulls.
The CVF will be a Joint Defence Asset and will primarily be a platform for the Joint Strike Fighter (envisioned as operating in a joint
RN/Royal Air Force [RAF] force), enabling operations from forces of
all three services to contribute to sea, land, and air battles.
Compared with Invincible-class carriers, which originally were
designed for helicopter-based anti-submarine operations, the CVF
will provide greater offensive air power and a larger array of aircraft
able to operate in a wider range of roles.
Currently, two prime contractors are competing for the design and
manufacture of the future carriers: BAE SYSTEMS and Thales
(formerly Thomson CSF). Each Competing Prime Contractor (CPC)
has submitted an initial Demonstration and Manufacture (D&M)
plan to the MOD. The MOD intends to place an order with a single
prime contractor for delivery of two vessels, the first of which is due
to enter service in 2012. Because of the large size of the carriers and
the limited facilities and workforce at most UK shipyards, it is envisioned that the work for the future carrier programme may be spread
among several different shipbuilders.
Landing Platform Dock (Replacement) [LPD(R)]

This project covers the replacement of the amphibious assault ships
HMS Fearless and HMS Intrepid, whose ages exceed 30 years. The
MOD is acquiring replacements for both, which are due in service in
2003. The MOD plans to augment these purchases by acquiring eight
23
associated specialised landing craft, four for each LPD(R), and two
training craft, which successfully ran trials as prototypes in 2000.
These craft are scheduled for delivery between 2001 and 2003.
The two LPD(R) are intended to provide command and control capabilities for amphibious operations. They will have two flight-deck
spots for Merlin-size helicopters and will carry four Mark 10 landing
craft in a floodable well deck and four Mark 5 landing craft in davits.
With 550 linear metres of vehicle parking space, the vessels will have
a cargo capacity of 31 main battle tanks or 16-ton trucks, 36 smaller
vehicles, and 30 metric tons of stores.
The LPD(R)s are being constructed by BAE SYSTEMS Marine at its
Barrow-in-Furness yard.
Future Offshore Patrol Vessels (FOPVs)

Vosper Thornycroft has been awarded a contract to lease three
FOPVs to the Royal Navy and to support those vessels for five years.
The vessels will be used to perform open-ocean patrols and other
missions. All will be built at VTs Woolston shipyard.
Survey Vessels
The Royal Navys Surveying Service, which has been operating
throughout the world since the formation of the Hydrographic Department in 1795, is responsible for hydrographic and oceanographic
surveying. From its survey data, the Royal Navy produces Admiralty
charts and other nautical information used worldwide.
Two shipsHMS Echo and HMS Enterpriseare currently under
construction at Appledore Shipbuilders, with VT as the prime contractor. Delivery is planned for 2002.
Type 45 Destroyer
The MOD intends to acquire up to 12 new-generation destroyers
from 2007. The Type 45s will be the largest and most powerful air
defence destroyers ever operated by the Royal Navy and the largest
general-purpose surface warship (excluding aircraft carriers and am-
24
phibious ships) to join the fleet since designs adapted from World
War II cruisers in the early 1960s.
The original intent was to buy the Common New Generation Frigate
(CNGF), a collaborative programme between the United Kingdom,
France, and Italy to procure a new class of antiair warfare warship
to replace the existing AAW ships. However, the trilateral ship-build
part of the programme was halted in 1999, and the UK started the
Type 45 project.8 When the destroyers enter service later this decade,
they will provide the fleet with an air defence capability significantly
greater than that provided by the existing Type 42s.
The main armament of the class will be the Principal Anti-Air Missile
System, which is being developed and procured jointly with France
and Italy. PAAMS is designed to simultaneously control several
supersonic, stealthy, highly maneuverable missiles that could use
sea-skimming or steep-diving profiles, and each of which could
engage individual targets. It will equip the Type 45 to defend itself
and other ships in company from attack by existing and future antiship missiles, by preventing attackers from swamping the fleets air
defences. Using PAAMS, the Type 45 also will be able to operate
close inshore and provide air cover to British forces engaged in land
battles. The UKs share of the cost of full development and initial
production of the first PAAMS will be approximately l billion.
In addition to a main gun for shore bombardment, the Type 45 will
have either the Merlin or the Lynx helicopter, which will carry
Stingray anti-submarine torpedoes. Should the requirement for a
land-attack capability arise, the Type 45 is large and spacious enough
to accommodate lengthened vertical launchers that could carry
______________
8 The initial Type 45 effort comprised two distinct programmes: the Principal Anti-Air
Missile System (PAAMS) and the ship and its other systems (Horizon). Memoranda of
Understanding (MOU) were signed by the three nations in July 1994 and March 1996.
For Horizon, an initial design and validation phase (Phase 1) started in March 1996.
This was to have been followed by Phase 2, the detailed design and build of three firstof-class (FOC) warships (one for each nation), to be procured under a single prime
contract. For PAAMS, the next major milestone was to be the start of PAAMS FullScale Engineering Development and Initial Production (FSED/IP). In April 1999,
ministers of the three nations announced that it was their intention to place the
PAAMS FSED/IP contract quickly but that it would not be cost-effective to pursue a
single prime contract for the warship. A MOD Integrated Project Team (IPT) then took
work on the warship programme forward, and the Type 45 emerged.
25
cruise missiles. The ship also will be able to embark a force of up to

60 Royal Marine Commandos or other troops and use its aircraft and
boats to support them on operations.
From 1994 to 1999, when the United Kingdom was a partner in the
three-nation Horizon programme to develop a CNGF, GEC, together
with VT and BAE SYSTEMS Marine, led its industry participation.9
That collaboration ended when the Horizon programme was terminated in April 1999, but it undoubtedly provided some foundation
for subsequent alliances on the Type 45 programme.
The initial contract, which was limited to initial design work and was
focused specifically on the Type 45 programme, was with Marconi
Electric Systems (then a part of GEC that would soon merge with BAE
SYSTEMS) in November 1999.10 In May 2000, VT and BAE SYSTEMS
Marine formed an alliance to bid on the final development and
construction of the Type 45. In September 2000, the Alliance
submitted a bid to the MOD for sharing development and construction of the initial three ships. A contract for that work was let to the
PCO in December 2000. In July 2001, the Secretary of State for
Defence, Mr. Geoffrey Hoon, announced that the contract would be
amended to cover the construction of the initial six ships.
Thus, by the end of 2000, it appeared that a structure and process
was emerging for distributing the Type 45 business across at least
three of the major shipyards (BAE SYSTEMS Marines Barrow-inFurness and Clyde [Scotstoun] yards, and VTs yard), with the possibility that BAE SYSTEMS Marine would distribute part of the work
to its Clyde (Govan) facility. These arrangements ensured that some
degree of competition would be feasible between at least VT and BAE
SYSTEMS Marine for later phases of the Type 45 construction programme. They provided a promise of cost control over the life of the
programme, plus the added benefit of sustaining a broad-based in______________
9 No documentation has been found on the exact structure of that team or the relative
participation of the various members.

10On November 28, 1999, Marconi Electronic Systems was appointed as the prime
contractor for the Type 45 programme. This responsibility was passed to BAE
SYSTEMS Electronics when the merger of Marconi Electronic Systems and British
Aerospace took place.
26
dustry that would be modernised and able to compete for future

warship projects.
Those arrangements were complicated in December 2000 when BAE
SYSTEMS Marine made an unsolicited proposal to build all 12 Type
45 ships (and other vessels) by itself. While apparently offering an
attractive price, such an agreement would constrain opportunity for
downstream competition-driven cost control on the Type 45. What
effect it would have on the overall competitiveness of the UK shipbuilding industry was less apparent: While possibly strengthening
BAE SYSTEMS Marines competitive posture, it would weaken that of
VT.
Thus, BAE SYSTEMS Marines unsolicited proposal for the Type 45
programme threw into sharp focus the need for decisions on specific
issues that had not previously been so clearly presented: the longterm benefits to be expected from a concentrated, monopolistic industry versus a more distributed (and, presumably, more competitive) industry.
RFA PROCUREMENTS
There are also plans to expand or modernise the RFAs tanker and
cargo ships over the next 25 years; in particular, the MOD intends to
take delivery of two new Auxiliary Oilers. The MOD also intends to
sign a contract giving it the use of six new RoRo vessels under a PFI
arrangement.
Alternative Landing Ship Logistics

The MOD is procuring four ALSL between 2002 and 2005. The ALSL
will underpin amphibious operations by carrying most of the amphibious force and its equipment into theatres of operation.
Whereas the LPD(R) provides the command and control function for
amphibious operations, ALSL will embark the largest balance of
men, vehicles, and stores needed for such operations.
The ALSL will succeed current Landing Ship Logistics (LSL) ships.
Representing a marked increase in capability over the ships currently
in the fleet, each ALSL will have nearly twice the carrying capacity
and will be much quicker to offload.
27
In mid-2000, three companies bid on the contract to build the first

two ALSL: Appledore Shipbuilders Ltd., BAE SYSTEMS, and Swan
Hunter Ltd. Swan Hunter won that competition. As it assessed the
tenders, MOD looked carefully at the strong operational reasons for
replacing more of the ageing LSL and identified sufficient funding for
two additional ALSL. To get the additional ships introduced into
service more quickly than serial construction by one shipyard would
have allowed, MOD chose Swan Hunter to design and build two
ALSL, with BAE SYSTEMS Marine constructing the other two ALSL at
its Clyde (Govan) shipyard, using Swan Hunters design.
Auxiliary Oilers (AO)

The RFA plans to take delivery of two AOs in 2002. Each will have
nearly 20,000 cubic metres total tankage for diesel and aviation fuel,
lube oil, and cargo water; 500 cubic metres for dry cargo; and space
for eight 20-foot refrigerated provision containers. Both are being
built by BAE SYSTEMS Marine, one at Barrow-in-Furness, the other
at Clyde (Govan).
Roll-OnRoll-Off Vessels
The Strategic Defence Review identified a need for six RoRos to
provide strategic sealift to Joint Rapid Reaction Forces. Operational
experience has demonstrated the difficulties in obtaining suitable
ships to move military equipment in the short timescales demanded
by such specialised forces and for the generalised armed forces
needs in operations worldwide.
Only major operations and exercises will require MOD to use all six
ships. Therefore, the MOD has pursued a contract for a long-term
service under the Private Finance Initiative.
AWSR Shipping Ltd. has been selected as the preferred bidder. The
ships will be fully crewed by British merchant navy personnel while
in MOD use. When combat operations require them, the seafarers
will be eligible for call-out as Sponsored Reserves.
Two of the ships will be built at the Harland and Wolff shipyard in
Belfast, Ireland; the other four ships will be built at the Flensburger
shipyard in Germany.
28
PLANNED SCHEDULE OF FUTURE SHIP PRODUCTION

As noted earlier in this book, the numbers and timing of future ship
programmes depend upon many variables. Table 2.3 presents
illustrative MOD ship-production plans for the vessels described in
the preceding section.
Table 2.3
MOD Naval Procurement: Current and Future Plans
Type/Class
Number
Tonnage
Delivery Years
Fleet
Aircraft Carrier (CVF)
Royal Navy
2
40,00060,000
20122015
Destroyer (Type 45)
12a
6,500
From 2007
Future Surface Combatant
20a
????
Not before 2012
11,060
20032004
Patrol (Offshore)
1,700
20022003
5,0005,208
20052013
3,500
2002
Submarine (Astute)
Survey
5/6
2

Auxiliary Oiler
18,200
2002
Alternative Landing Ship

Logistics
16,160
20022005
Roll-onRoll-off Vessel
10,000
Complete by 2003
Other
SOURCE: The information in this table was taken from Ministry of Defence,
Performance Report 1999/2000, Abbey Wood, England, Annex B: Force Structure.
aMaximum
class size.
RAND MR1486-T2.3
29
DISTRIBUTION OF SHIP CONSTRUCTION ACROSS

SHIPYARDS
We have described the status of the UK shipbuilding industrial base
and outlined the current and future shipbuilding programmes that
have been postulated to satisfy MOD requirements. Figure 2.2 shows
the distribution of present shipbuilding production activity among
the shipyards, as well as an illustrative schedule for the Type 45, CVF,
and FSC. How that future business should be distributed across the
available shipyards is one of the critical issues facing the MOD. MOD
decided, with support by this analysis, that the first six Type 45s will
RAND MR1486-2.2
FSC
CVF
Not yet determined

A0 01
ASTUTE
Barrow
LPD
T-23
Clyde (Scotstoun)
A0 02
ALSL
ALSL
SURVEY
TRIMARAN
SRMH
FOPV
2000
Shipyard
Ship type/class
T-45
Clyde (Govan)
Swan Hunter
Appledore
Vosper Thornycroft
2004
2008
2012
2016
Year
aOnly the first three ships of the FSC class are shown. There may be a total
of 20 FSCs, with the start and delivery dates of the remaining 17 ships spaced
approximately six months apart.
Figure 2.2Royal Navy Warship Production Schedules by Ship Class (left)

and Assigned Shipyard (right). The schedules shown are illustrative.
30
be built at BAE SYSTEMS Marines Clyde (Govan and Scotstoun) and

Barrow facilities and by VT at Portsmouth.
At present (2001), six shipyards have some naval shipbuilding under
way. However, only one yard (Barrow-in-Furness) has an order book
extending beyond the next three or four years. For the Type 45 and
Future Carrier programmes to be focused on one or two shipyards
would likely drive one or more of the remaining shipyards to abandon naval shipbuilding and, in turn, would tend to lessen the opportunities open to the MOD for obtaining the potential benefits of
competition in future programmes. The acquisition policy and strategy options this situation presents to the MOD are explored in the
next chapter.
Chapter Three
IDENTIFYING AND ANALYSING MODS ACQUISITION

CHOICES
In managing the procurement of future naval vessels, most immediately the Type 45 programme, the MOD will face a number of issues
and options. In this chapter, we describe the basic issues and options, and the analytic process used to evaluate them.
ISSUES AND OPTIONS

As noted in Chapter One, the original Alliance plan to share the
workload for the first three Type 45 ships between BAE SYSTEMS
Marine and VT, with competition to decide the distribution of the
remaining ships in the class, was profoundly shaken by the BAE SYSTEMS Marines unsolicited proposal to produce all 12 ships. The
MOD was faced with several options for distributing the business
across those firms and shipyards, each yielding a different set of
advantages and disadvantages. In broad terms, the options are as
follows:
Sole-source procurement: In practice, the source for Type 45

production would be BAE SYSTEMS Marine, because the other
potential builder, VT, does not have the capacity for the entire
project.
Dual-source procurement: Production would be distributed

across two companies, with distribution taking place in a com-
31
32
petitive environment, through a directed-buy1 option, or in a

mixture of the two.
When evaluating these options, the MOD is concerned with at least
three major objectives: controlling the total cost of ship construction; providing incentives for technological innovation and the most
effective use of industrial resources; and contributing to a strengthened shipbuilding industry that would be capable of effectively competing for future naval and commercial business.
In this study, we addressed these options and objectives in two ways.
First, we focused on certain elements of production cost that we believed to be sensitive to an acquisition strategy for the Type 45direct labour costs, labour transition costs, and overhead costsrather
than the total cost of producing the Type 45. Specifically, our examination excluded the cost of weapon systems that will be placed on
the Type 45. Production costs for the Type 45 should not be treated
in isolation. We considered the total work from all programmes at a
shipyard.
Second, having investigated these costs, we subjectively examined
how competition might affect other objectives for the Type 45 programme, for other MOD shipbuilding programmes, and for the
overall health and performance of the UK shipbuilding industrial
base.
QUANTITATIVE EFFECT OF USING TWO SHIPYARDS

How the workload for the 12 ships in the Type 45 programme is distributed between the two shipbuilders affects not only the total
labour hours needed to construct the ships but also how those total
labour hours are spread among the four shipyards.
In the absence of any workload reductions resulting from competition, sole-source procurement will theoretically result in the lowest
number of total labour hours for the programme. This is a result of
the learning effect, a widely accepted principle in manufacturing industries. According to the principle, the workforce becomes more
______________
1 Directed buy entails the buyer defining and directing the distribution of work among
firms, rather than unfettered competition determining the distribution.
Identifying and Analysing MODs Acquisition Choices
33
proficient when activities are done on a repetitive basis, which implies that fewer hours will be required to build each successive ship.
A learning curve describes how the hours to build each ship decrease
with successive ships. For example, a 90-percent learning curve suggests that doubling the number of ships built by a workforce would
reduce the labour hours to construct the ships by 10 percent.
Therefore, if the first ship required, say, 1 million man-hours to
build, the second ship would need only 900,000 hours, and the fourth
ship would require only 810,000 man-hours.2
Because of this learning effect, and temporarily ignoring any possible
offsetting effects of competition, distributing the 12 Type 45s between two producers would most likely result in an increase in the
total hours over those for all units being built by one producer. Figure 3.1 shows the theoretical increase in the total man-hours for the
12 ships for different distributions between the producers and for
different learning-curve slopes. For example, if the 12 ships were
split equally between the two producers (the worst distribution in
terms of total man-hours), the total man-hours would increase by
approximately 9 percent for a 90-percent learning curve and by almost 19 percent for an 80-percent learning curve.
Two important assumptions are inherent in this two-producerinduced increase in man-hours, and they lead to two important
questions. The theoretical calculation behind Figure 3.1 assumes
that the learning curves at the two producers are the same, thus
making the curves symmetrical around the 6/6 split. If one
manufacturer has greater or lesser learning than the other, the curves
would not be symmetrical and would show a different increase in
man-hours.
______________
2 This example follows the unit cost theory (as opposed to the cumulative average
cost theory), for which the average cost of the first two units would be 90 percent of
the first unit and the average cost of the first four units would be 81 percent of the first
unit. The unit theory is generally accepted as appropriate for major weapon-system
production.
34
RAND MR1486-3.1
20
Increase in man-hours (%)
18
16
14
Learning-curve slope 80%
12
85%
10
8
90%
6
95%
4
2
100%
0
Split
0/12
Split
1/11
Split
2/10
Split
3/9
Split
4/8
Split
5/7
Split
6/6
Distribution
Figure 3.1Increase in Total Man-Hours for Different Symmetrical

Learning Curves and Distributions of the 12 Type 45 Ships
Second, although the theoretical increases shown in Figure 3.1 may

be appropriate for a directed-buy option, they may not be correct for
the competition case. In theory, competing firms find ways to
increase learning from ship to ship, or they find ways to reduce the
time to build a ship in order to win subsequent bids. In fact, the expectation is that competition will potentially overcome the increase
in man-hours shown in Figure 3.1 (and other cost effects) and,
therefore, lead to lower costs than those resulting from the solesource option.
This leads to two important questions: Is learning greater for competitive programmes than for noncompetitive programmes? What is
the appropriate learning curve to use?
35
To address these questions, we gathered man-hour data from 25 different United Kingdom and U.S. Navy ship programmes. Table 3.1
summarises our statistical analysis of these programmes.3
The analysis indicates that the average learning curve for the 25 programmes was 87 percent and that the learning curve for competitive
programmes was slightly better (2 percentage points) than that for
noncompetitive programmes, although not a significant difference.4
Figure 3.2 shows the histograms for the complete sample of competitive and noncompetitive programmes.
Table 3.1
Summary of Ship-Production Learning-Curve Statistics
Summary
All programmes
All competitive programmes
All noncompetitive programmes
UK-military competitive programmes
UK-military noncompetitive programmes
Average Slope Standard

Deviation
(percent)
Sample
Size
5.2
5.9
4.6
25
14
13
83
5.3
86
4.9
87
86
88
NOTE: Data for individual programmes are privileged and thus can be displayed
only in a summary format.
RAND MR1486-T3.1
______________
3 This analysis was prepared by Fred Timson at RAND.
4 The standard deviations are large and the sample sizes are small; therefore, it is not
obvious from casual inspection that the differences between the competitive and noncompetitive samples are meaningful. To formally examine the difference, a statistical
test is run to determine whether differences between means are significant. One such
test, a t-test, determines the likelihood that the difference is due to random chance.
If that likelihood is low (usually 5 percent), the difference between the mean values
can be assumed to be meaningfuli.e., the difference is real. The generally accepted levels of significance are 5 or 1 percent. Occasionally, a result may be described as significant at the 10-percent level. In our particular case, the likelihood is
36
RAND MR1486-3.2
Competitive
Noncompetitive
Number of observations
5
4
3
2
1
0
7580
8085
8590
9095
95100
Slope ranges (%)
Figure 3.2Frequency Distribution of Learning-Curve Slopes for

Competitive and Noncompetitive Programmes
The average learning curve of 87 percent suggests that, theoretically,

splitting the 12 ships between the two producers may increase manhours by as much as 11 percent.
OUR ANALYTIC APPROACH

Drawing from and building on previous RAND studies that examined
the economic effects of various procurement strategies (Birkler et al.,
1994, 1998), we developed an analytic model of the UK shipbuilding
industrial base specifically for the Type 45 programme. The model
takes into consideration the many unique aspects of the programme.
It also considers all current and future programmes at BAE SYSTEMS
much greater than 5 percent, indicating that the difference could be entirely due to
chance (incomplete sampling). For the competitive and noncompetitive subsets of all
the ship programmes considered, the likelihood of concluding that the two sets are
different when they are really the same is approximately 30 percent; the t-statistic is
1.05 with 26 degrees of freedom (= t sample size). For the UK military programmes,
the likelihood is about 40 percent, and the t-statistic is 0.89 with 5 degrees of freedom.
Thus, we cannot say that there is a statistically significant difference between
competitive and noncompetitive programmes.
37
Marine and VT shipyards and calculates the workforce, overhead,

and investment costs. The results were validated with contractors
and various MOD offices.
Using the option of having BAE SYSTEMS Marine build all 12 Type
45s as the baselineits unsolicited proposalthe model displays the
relative cost effects on the Type 45 and other programmes for alternative procurement paths. However, as we described in the preceding section on learning curves, we do not know how competition
will affect the learning curves at the shipyards or the number of
hours to build successive ships. No one has been able to adequately
quantify how much competition saves in weapons procurement
how much each competitor is willing to invest to reduce costs, how
much risk each is willing to assume, or how efficient each competitor
can become. In our model, we estimate only the costs of specific distributions of the total workload without taking into account any effect of competition. The results of these calculations show the incremental costs of distributing production among two or more producers compared with the costs of a single producer.
Shipyard Labour Model

Over the past several years, RAND has developed analytic methods to
examine workforce issues at shipyards. The total labour cost for all
the work at a shipyard is more than the sum of its individual components. Labour costs at shipyards have nonlinear aspects whereby a
change in an input to the system does not result in a proportional
change in an output (cost). For example, if we increase the number
of ships that we build by 20 percent, the total production cost does
not increase by 20 percent; it increases by something less than 20
percent. Similarly, a decrease of 20 percent in the number of ships
produced results in a production-cost savings that is less than 20
percent. RANDs research has focused on modelling these nonlinear
aspects of labour cost.
Many factors cause this nonlinearity. Overhead and burden are one
example. Overhead costs are those costs that are related to production activities but that cannot be allocated directly to a particular
product because of either practicality or accounting convention.
Overhead includes the costs of fringe benefits, indirect labour, depreciation, building maintenance and insurance, computer services,
38
supplies, travel, and so forth (Cash, 1999). Burden is the sum of

overheads, G&A (General and Administrative), and profits. The
overhead rate is generally inversely proportional to the total
workload. So, removing work from a shipyard lowers the direct
labour cost but increases the indirect rate for the remaining work.
Another nonlinear component involves worker productivity and
training cost.
Ideally, a production facility has a workload that is fairly stable over
time or that grows moderately. When workload changes rapidly or
fluctuates, additional costs are introduced. For example, a facility
would have to hire many new workers if workload were to increase
rapidly, and it is likely that many of these new workers would be inexperienced. Hence, productivity would suffer. In a situation of increasing workload, therefore, we would expect the workforce to have
lower overall productivity, causing labour costs to rise. Additional
direct hours must be worked to complete the tasks, and there would
be some nonrecurring fixed cost to train these new workers.
The following is a partial list of nonlinear factors in shipbuilding:
Training Costs: costs associated with training new employees,

whether in direct cash outlays, such as external courses, or nonproductive time.
Hiring Costs: costs associated with hiring new employees, which

may be recruitment fees, bonuses, management time, etc.
Termination Costs: compensation that may be owed to a worker

upon termination of employment. This compensation is typically some fraction of his or her annual salary.
Productivity Costs: higher labour costs to cover the extra hours

that new workers require to perform a specific task.5 As described above, new workers hired by a shipyard may be inexperienced and, therefore, take longer to complete tasks than it takes
experienced workers. Over time, new hires gain experience,
______________
5 For our example, we have assumed that the hourly wage rates are the same. In the
United States, the wage rate of a new worker is usually lower than that of an experienced worker. This wage differential somewhat compensates for the lower productivity.
39
lessening the learning effect and increasing productivity.

However, the effect of dramatically increasing the workforce at a
facility can be seen for several years.
Constraints in Ability to Expand or Contract: limits placed on a

shipyard by the available labour pool, mentoring ratios, etc. A
shipyard does not have an infinite labour pool upon which to
draw new workers, so there might be a maximum rate at which a
yard could expand. Even when there may be available workers,
there could also be some limit on how many new employees a
yard can accommodate at one time (training and mentoring ratios may limit this rate). These constraints would cause the shipyard to carry more people than necessary at times in order to
meet peak demands in the future, a situation that, again, would
tend to increase total labour cost.
Learning: proficiency that occurs when building multiple ships

of the same class. A workforce becomes more proficient with
each successive ship; hence, the labour hours and costs decrease.
This learning effect has important implications when we look at
sole-source versus dual-source acquisition strategies. With all
other things being equal (and ignoring competitive pressures),
the cost to build a class of ships should take fewer hours if all the
work is concentrated at a single producer.
Overhead and Burden Rates: costs of fringe benefits, indirect

labour, depreciation, building maintenance and insurance,
computer services, supplies, travel, etc., and the sum of overhead, G&A, and profit. Overhead rates are inversely proportional
to workload. When workload decreases, overhead rates increase,
reducing any savings from removing the work.
Methodology
The goal of the model is to estimate the labour costs, overhead rates,
labour-force transition costs, and learning improvements across
several shipyards for a given shipbuilding strategy.
Once the acquisition plan was determined, we calculated the labour
demands at each shipyard. Every shipyard project that might be
built carries a labour profile that shows the man-hours, or equivalent
workers, per quarter over the build period. These labour profiles are
40
unique to each shipyard.6 After applying an appropriate learning

factor, we summed these labour profiles across all programmes at
the shipyard to determine the total labour demand at a yard. This
total labour is what we term required labour: the minimum labour
necessary to complete all tasks at each shipyard.
However, the required-labour level may not be the actual level that
the shipyard would employ. Hiring and firing constraints (e.g.,
mentoring ratios) may cause the shipyard to hire or retain more
workers than are absolutely required. Given labour constraints that
restrict the rate of workforce expansion from one period to the next,
we used a linear programming model to determine the effective
labour. Effective labour is always equal to or greater than the required labour.
From the effective labour, we established an overall burden rate for
the shipyard. We modelled burden as having a fixed component and
a variable component. The burden rate equation for a shipyard takes
the following form:
Burdenquarter =
A
+B
hoursquarter, trade
(3.1)
trade
where A and B are constants specific to each shipyard. Note that as

the total site hours increase, the burden rate decreases. This elastic
model works well if there are not large, long-term changes in employment levels (particularly, downward ones). In practice, a shipyard would reduce its fixed burden costs in such an environment,
possibly by shedding unneeded facilities and/or reducing the indirect staff and support. To reflect such a situation in the model, we
applied a cap (200 percent) to the burden rate.
Lastly, after determining both the effective hours and burden rates,
we calculated total labour cost, as follows:
______________
6 Labour profiles are developed for three tradesdirect labour, support, and engi-
neering. BAE SYSTEMS Marine and VT provided the build periods, workload profiles,
and learning-curve assumptions.
) rate
Cost quarter = 1 + Burdenquarter
trade
41
hoursquarter, trade (3.2)
trade
where
rate trade is the wage rate for a specific labour type.
Data
Obviously, such a model requires an extensive amount of data about
each shipyard and ship class. RAND researchers prepared comprehensive data-collection forms so that such data would be collected in
a consistent fashion. The types of data gathered were as follows:
Shipyard Capacity: steel throughput, docks, lifting capacity, outfitting berths, etc.
Workforce Profile: experience levels, age, productivity, costs for

hiring and training, termination costs, restrictions on hiring and
termination, current employment levels, and the use of contract
workers.
Production Experience: numbers and types of ships built over the

past five years (including commercial work).
Current and Future Production: current and anticipated production plans (by ship).
Workload Projections: for each activity listed in the current and

future production plan, a listing of the labour profile, by trade
and quarter. Further, these data include design and development workload.
Wage Rates: hourly wage rates for all the labour types.
Burden Rates: overheads (OH), G&A, and profit rates as a function of different sites workloads.
Investment Levels: fixed investments, such as facilities, necessary

for a particular programme or investment required overall.
We requested this information from five shipbuilders: VT, BAE SYSTEMS Marine, Swan Hunter, Appledore, and Harland and Wolff. We
42
also held discussions during site visits with each firm. Beyond the
shipbuilders, RAND had similar discussions with Type 45 and Astute
PCOs, as well as with all current MOD ship programme managers
and their staffs. These offices provided considerable supporting information and data about their ship programmes.
Specific Assumptions
To use the quantitative model to estimate the costs for the various
Type 45 acquisition options, we made several simplifying assumptions (see Appendix A for the results of sensitivity analyses involving
these assumptions). They include:
Shipyards: For this initial analysis, we considered only BAE SYSTEMS Marine shipyards (Barrow-in-Furness and Clyde) and the
VT shipyard (assuming it relocated to Portsmouth).
Time Period: 2001 to 2014. This period covers the planned construction of the Type 45s.
Shipbuilder-Provided Data: site-specific learning curves,

labour-force data, and overhead rates.
Workload Allocation: for the FSC, split evenly between BAE SYSTEMS Marine and VT; for the CVF, one-half to BAE SYSTEMS
Marine and one-quarter to VT.7
Maximum Overhead Rates: 200 percent.
Facility Investment: For the two shipyards (VT and Barrow-inFurness) to compete for the Type 45 production, each site would
require some facility investment/improvement. Given an early,
order-of-magnitude estimate by the Type 45 PCO, we used the
following assumptions in our calculations:
Barrow-in-FurnessThe investment would depend on the
extent of the improvements required. We assumed that this
investment cost at Barrow-in-Furness would be independent
of the number of ships BAE SYSTEMS Marine won. In other
words, this shipyard would build facilities to the full rate of
______________
7 This is RANDs allocation and does not reflect any MOD input or guidance.
43
production, not knowing how much of the production it

would win; so, its investment costs would be roughly the
same under sole-source and competition.
VTThese costs are more difficult to determine, 8 because
there was no detailed estimate on the cost to relocate VT to
Portsmouth. As another rough estimate, the PCO thought
that nearly 2.5 million would be needed to relocate VT to
Portsmouth and another 25 to 50 million would be required to upgrade the Portsmouth site to produce the Type
45. Given the lack of more detailed estimates by VT, and
erring on the conservative side, we assumed that the site investment cost for VT would be 40 million (again, independent of production rate). For block production, we assumed
that the investment cost would be half that value, or 20 million.
Break-Even Approach
Competition is just one form of a multiple-source production strategy. Other strategies can allocate production among several firms.
Unfortunately, there is no reliable method for predicting the savings
from multiple-source procurement (competitive or otherwise). We
cannot predict the behaviour of the firms or their willingness to reduce costs and profits in order to undercut the other firm or firms.
Thus, the basic question is not how much money will be saved but,
rather, whether introducing an additional production source is a reasonable strategy to pursue.
One way to make that reasonableness determination is through a
break-even analysis (Margolis, Bonesteele, and Wilson, 1985;
Hampton, 1984). Such an analysis does not require an explicit estimate of the savings from multiple-source production; rather, it deduces the magnitude of savings needed to justify a second source of
production. In general terms, break-even refers to the point at
which the expected cost to the government of a multiple-source al______________
8 The estimates were made by RAND researchers and do not reflect VTs actual
figures.
44
ternative equals the cost of the sole-source alternative, which, for the
Type 45, is the unsolicited proposal from BAE SYSTEMS Marine.
To calculate a break-even value, we used our shipyard model to determine the production costs for various strategies. Our metric is a
percent change in production cost relative to the BAE SYSTEMS
Marines sole-source case. We include in this percent not only the
effect on the Type 45 programme but also the effect on other contemporaneous programmes. It is a net cost delta for the UK government:
Break - even percent =
All Programs
Cost Dual Source
Cost Sole Source
All Programs
Cost Dual Source
Type 45
(3.3)
For competitive strategies, we can compare this break-even value
with historical values to assess whether competition could reasonably lead to overall cost savings. We make such comparisons in the
next chapter.
Chapter Four
ANALYSIS OF PROCUREMENT ALTERNATIVES
Various acquisition strategies are open to the MOD for the Type 45.
In this chapter, we first describe the five alternative procurement
strategies examined in this study, then show the expected labour
costs of each. Next, we discuss the factors other than labour cost
risks and rewardsthat might be affected by a competitive strategy.
We conclude this discussion with an overall integration of the analysis and observations on all the material in the chapter.
CASES EXAMINED IN THE COST ANALYSIS

A central objective of this study was to examine the cost consequences of different strategies for producing the Type 45. We examined five basic strategies on three procurement paths:
1. Sole-source procurement. All production is performed by a single firm.
2. Dual-source competitive procurement. Production is split between two firms, with the split determined by head-to-head cost
competition. Two variations were examined:
2a.
Competitive procurement of whole ships
2b.
Alliance proposal, involving assigned production of the

first three ships (with the first-of-class built in blocks at
multiple shipyards) and competitive procurement of the
remainder.
45
46
3. Dual-source, directed buy. Production is split between two firms,

but the split is directed by the MOD, without head-to-head price
competition extending throughout the life of the programme.
Two variations were examined:
3a.
Assigned production of whole ships
3b.
Assigned production of individual ship components

(blocks).
COMPARISON OF LABOUR COSTS ACROSS STRATEGIES

As described in the preceding chapter, we calculated costs using
identical assumptions for both competitive and noncompetitive
strategies. That is, we made no attempt to adjust costs to reflect the
fact that firms might perform differently according to the degree of
competition present. Our objective here is to estimate the direct
consequences of spreading the production between two firms rather
than concentrating it at one firm. That cost difference represents the
amount of cost reduction that would have to be created through
competitive pressure in order for the MOD to break even by sustaining a competitive environment for Type 45 production, or that would
have to be justified on some other criterion if a dual-source strategy
were adopted.
The Reference Case: Sole-Source Production

Our analysis approach entails a basic assumption: Any procurement
strategy that concentrates all production at one firm cannot be truly
competitive. A single-source strategy is likely to entail fierce competitive bidding at the start of the construction programme; however, as
soon as an award is made to a single firm, the competitive environment disappears and that firm becomes the monopolist supplier for
the remainder of the construction programme. If that monopolist
supplier fails to perform according to the terms of the contract or is
not otherwise responsive to the needs and preferences of the buyer,
the buyer has limited recourse to effective remedies. The history of
military systems acquisition is replete with examples of sole-source
suppliers who failed to satisfy the terms of the original contract and
the severe problems those failures caused the procuring agency.
Analysis of Procurement Alternatives
47
For these reasons, we define a sole-source acquisition as a noncompetitive strategy. 1 We calculated the production cost for the solesource strategy by assuming the plan described in the BAE SYSTEMS
Marine unsolicited proposal, whereby all ships would be built in BAE
SYSTEMS Marine facilities, thus achieving maximum learning effects
as well as sharing overhead costs with other shipbuilding programmes scheduled for those facilities. The result becomes the reference point against which all dual-source options are compared.
Dual-Source, Competitive Procurement of Whole Ships

We assumed that procurement is divided into four sequential lots of
three ships each, to sustain competition throughout the construction
programme. A price competition would be held for each lot, and the
winner would be awarded production of two ships and the loser
given the option of building one ship at the winners price.
Under this strategy, the final allocation of ships depends on which
shipbuilder wins each of the three-ship competitions and whether
the loser decides to build one of the three ships at the winners price.
To pare the multitude of options to a manageable few, we decided on
the following three cases for the competitive options:
Eight ships to BAE SYSTEMS Marine and four to VT
Six ships to BAE SYSTEMS Marine and six to VT
Four ships to BAE SYSTEMS Marine and eight to VT.
These three specific cases should provide sufficient insights to understand the relative cost differences between the sole-source option
and the directed-buy and competition options.
______________
1 We recognise that the BAE SYSTEMS Marines unsolicited offer was made in a
competitive environment and offered attractive terms. However, under that proposal,
no continuing competition would have existed throughout the production phase.
Experience shows that, under such conditions, production costs tend to rise and the
programme becomes the equivalent of a sole-source negotiated contract.
48
Dual-Source, Competitive Procurement: Alliance Proposal

This strategy is basically the initial plan for the Type 45 (before the
unsolicited proposal from BAE SYSTEMS Marine), whereby each
shipbuilder would be allocated approximately half of the total workload for the first three ships and then compete for the following three
lots of three ships each, with the loser having the option of building
one of the three ships at the winners price. Again, there are a number of potential outcomes, depending on which shipbuilder wins
each of the three competitions and what decision the loser makes.
For simplicity, we limit the Alliance option to the case where BAE
SYSTEMS Marine builds 7.5 ships and VT builds 4.5 ships.
The outcome of this case is similar to the third case in the preceding
strategy whereby BAE SYSTEMS Marine builds eight ships and VT
builds four (only one-half a ship difference), except that here competitive pressure is applied only to the last nine ships. Therefore,
comparing the costs of the specific Alliance case to the case of
eight/four should provide sufficient insights of the other potential
outcomes of the Alliance competition (for example, BAE SYSTEMS
Marine building 6.5 ships and VT building 5.5 ships).
Comparison of Dual-Source, Competitive Procurement with

Sole-Source Procurement
Figure 4.1 shows the relative percent cost increase of the three specific whole-ship cases and the Alliance case compared with the solesource strategy. 2 In percent terms, these increases represent how
much competition would have to save in order for the Type 45 cost
______________
2 When estimating the percent increase in cost due to distributing production between
two producers, we needed to include the same cost elements as were used in the
historical database of competitive and noncompetitive ship production. Unfortunately, we do not know the exact array of cost elements included in the historical data,
and they probably differ somewhat from programme to programme. We do know that
almost all ship procurements include some Government-Furnished Equipment (GFE)
that is probably not included in the stated total cost of the ship, but the stated costs
shown in historical records will always include more than labour and overhead.
Therefore, when calculating the percent increase in cost due to distributing Type 45
production between two producers, we added the cost of material, about 50 million,
which we assume, because it is purchased by the PCO and provided to the shipyards,
is the same across all options.
49
with competition to equal the sole-source cost. Note that the Alliance case involves a directed buy of the first three ships and competition for the last nine ships. Therefore, the cost of the competed
nine ships would have to be reduced about 13 percent in order to
match the cost of the sole-source option. The other three cases involve competition for all 12 ships. Therefore, for the MOD to break
even, competition must yield greater percentage gains in the Alliance
case for the nine competed ships than for the other three cases
(where there is competition for 12 ships).
Figure 4.1 shows that, of the three competitive procurement cases,
the cost relative to the sole-source option increases as VT builds
more of the 12 ships. The highest increase in cost is approximately
13 percent when BAE SYSTEMS Marine builds four ships.
Figure 4.2 provides further details on what contributes to the cost increases over the sole-source option. It categorises cost by Type 45
labour and overhead, initial investment cost, Astute overhead, transportation (of the blocks from VT to Clyde [Scotstoun] in the Alliance
case), and other (a decrease in the costs of other programmes at VT
resulting from increased workload compared with the VT workload
under the sole-source option). The figure shows that the Type 45
RAND MR1486-4.1
14
Cost penalty (%)
12
10
8
6
4
2
0
12
7.5
(Alliance)
Number of ships to BAE SYSTEMS Marine
Figure 4.1Cost Penalties of Five Whole-Ship Procurement Options
50
RAND MR1486-4.2
14
Cost penalty (%)
12
10
8
6
Transport
Astute OH
Investment
T45 labour
and OH
Other
programmes
4
2
0
2
12
7.5
(Alliance)
Number of ships to BAE SYSTEMS Marine

NOTE: For the whole-ship options, the cost of transportation is so small relative to
the other costs that it does not show.
Figure 4.2Composition of Cost Penalties of Five Whole-Ship

Procurement Options
labour and overhead costs at first increase as BAE SYSTEMS Marine

moves from building 12 ships (the sole-source option) to an even
split of the ships between BAE SYSTEMS Marine and VT, then decreases as VT builds more ships (owing to the symmetrical nature of
the learning-curve assumptions as described in the preceding chapter). Also, the Astute overhead increases as BAE SYSTEMS Marine
builds fewer ships.
In addition to providing the cost implications of different acquisition
strategies, the model also provides the size of the labour force required at the various shipyards for the different programmes ongoing
at the shipyard.
Break-Even Analysis for Competition

From the information in Figures 4.1 and 4.2, we can estimate how
much competition would have to lower costs compared with the
costs of the sole-source option. The question is, Are these percent
reductions reasonable to expect from competitive programmes? As
51
described in Chapter Three, we gathered data from a number of historical programmes that involved competition to understand the potential effect of competition on cost. For comparative consistency,
we desired historical data on ship-construction programmes that involved competition. To provide a basis for estimating total costs if
conducted sole-source, we needed programmes that had an initial,
noncompetitive phase, followed by a competitive phase (Birkler et
al., 2001).
We found five such programmes: TAO-187, LCAC, Type 23, CG-47,
and LSD-41. In this small sample, three of the five programmes
resulted in savings of more than 10 percent and two of the five programmes resulted in savings of 20 percent or more. This small,
ship-only dataset suggests that there is about an equal chance that
competition in the Type 45 programme can lead to costs lower than
those of sole-source acquisition as that the costs of competition
could be greater than sole-source.
Because of the small sample size of ship programmes, we augmented
our database of competitive programmes by adding missile programmes that were competitive. The distribution of percent savings
for the new database of 31 programmes is shown in Figure 4.3.
RAND MR1486-4.3
Number of observations
Median: 11-percent cost reduction

Mean: 7-percent cost reduction
8
7
6
5
4
3
2
1
0
40
30
20
10
10
20
30
40
Cost change due to competition (%)
Figure 4.3Cost Changes Resulting from Competition in 31 Missile and

Ship Programmes
52
The figure shows that the average savings from competitive ship and
missile programmes was approximately 7 percent, with a median
savings of approximately 11 percent. Most competitive programmes
resulted in decreased costs; however, a few programmes actually resulted in increased costs of up to 30 percent.
Given that competition in the Type 45 programme would have to reduce costs (compared to the sole-source option) from 10 to 13 percent, depending on the distribution of ships to the two shipbuilders
(see Figure 4.1), the historical competitive programme data suggest
that there is approximately a 50:50 chance that competition, if it can
be sustained over the programme as a whole, will result in equal or
lower costs than the sole source. Therefore, there is no dominant answer to whether competition or sole source would likely lead to lower
costs. 3
Dual-Source, Directed Buy of Whole Ships

One of the important potential advantages of the competitive-buy
option is that two shipbuilding firms would remain in the military
shipbuilding business, thus encouraging innovation, supporting improvement in facilities, and generally strengthening the UK shipbuilding industry. However, the proposed Alliance mechanism involves contracting for four successive blocks of three ships each, with
the winning bidder being awarded two ships but with the loser having the option of building one ship at the winners pricean arrangement offering a distinct possibility that the loser would simply
exit the programme, leaving a sole source to build the remaining
ships.
If the MOD wishes to ensure keeping two firms in the military shipbuilding business, one option would be to simply direct some ships
to one firm and the remainder to the other firm. While forgoing the
possible benefits of competition, such a strategy might be attractive
for a variety of reasons other than cost minimisation, such as retain______________
3 Here, we focused on the effect of competition or sole source on Type 45 costs. There
may also be implications for future programmes, especially if the sole-source option
results in BAE SYSTEMS Marine becoming the sole warship builder in the United
Kingdom. Appendix B provides an initial analysis of the potential costs of future programmes in such a monopoly situation.
53
ing the option of competition in future programmes and stabilising

employment levels at select shipyards. However, there would likely
be a cost penalty compared with sole-source procurement. Barring
the introduction of other considerations, that cost penalty would be
as shown for the three split-buy options displayed in Figure 4.1:
roughly 10 to 13 percent.
Dual-Source, Directed Buy of Ship Blocks

In addition to acquisition options that involve the allocation of whole
ships between the two shipbuilders, we examined two options that
involve the directed allocation of major portions, or blocks, of the
ships as depicted in Figure 4.4.4
Such a strategy would keep both shipbuilders involved in the Type 45
programme, and in future warship programmes, while overcoming
the disadvantages, mentioned in the preceding chapter, of increased
man-hours due to lower learning gains. By having each shipyard
build the same blocks for all 12 ships, the maximum gain to learning
can be achieved. Therefore, acquisition strategies that involve building blocks at shipyards have a number of advantages over the options of building whole ships.5
We examined two specific block options:
Option I:
Blocks B, C, and final assembly at Barrow-in-Furness
Blocks D and E at VT
Blocks A, F, and mast sections on the Clyde.
______________
4 Since the original plans called for the first-of-class to be built in blocks at multiple
shipyards and the whole ship assembled at Clyde (Scotstoun), the block concept was
incorporated into the Type 45 design from the very beginning.
5 In addition to advantages of building the ship in blocks, there are disadvantages, such
as increased costs for transporting the blocks to the final assembly area. Appendix C
describes historical experiences of building ships in blocks at different shipyards and
provides an assessment of the potential advantages and disadvantages.
54
RAND MR1486-4.4
MASTS
BLOCK A
MEGA-BLOCK B + C
BLOCK D
BLOCK E BLOCK F
Figure 4.4Type 45 Blocks
Option II:
Blocks B, C, and final assembly at Barrow-in-Furness
Blocks D, E, F, and mast sections at VT
Block A on the Clyde.
For each of these options, we assumed that the initial investment

would be 20 million less than the investment costs for the solesource option (since VT would require less building/upgrading of
facilities at Portsmouth, and Barrow-in-Furness would require the
same degree of facility building/upgrading as under the sole-source
option) and that the cost to transport the blocks from VT or the Clyde
to Barrow-in-Furness would be 300,000 per shipment.
Figures 4.5 and 4.6 show the cost penalties for these two block options when compared with the sole-source acquisition strategy. For
comparison purposes, we also show in the two figures the cost of the
whole option, whereby BAE SYSTEMS Marine builds eight ships and
VT builds four (the lowest-cost penalty of the whole-ship options).
As Figures 4.5 and 4.6 suggest, both block options significantly reduce the cost penalty of splitting production between two firms
rather than taking the whole-ship options. This reduction results
primarily from the increased learning effect of having each shipyard
build all 12 of the specific blocks. However, this general strategy
55
RAND MR1486-4.5
14
Cost penalty (%)
12
10
8
6
4
2
0
BAE SYSTEMS
Marine Proposal
8 to BAE SYSTEMS
Marine, 4 to VT
Block Option I
Block Option II
Allocation strategy
Figure 4.5Cost Penalties Associated with Four Procurement Options
RAND MR1486-4.6
14
Cost penalty (%)
12
10
8
6
4
Transport
Astute OH
Investment
T45 labour
and OH
Other
programmes
2
0
2
BAE SYSTEMS
Marine Proposal
8 to BAE SYSTEMS
Marine, 4 to VT
Block Option I
Block Option II
Allocation strategy
Figure 4.6Composition of Cost Penalties Associated with Four

Procurement Options
56
provides no opportunity for continuing competition throughout the

production programme, no basis for expecting any reduction from
the costs shown, and minimal incentive for innovation. Therefore,
this strategy must be evaluated on criteria other than that of expecting to break even on costs within the Type 45 programme.
Summary of Cost Analysis

In Figure 4.7, we summarise the estimates of production cost for the
various options relative to the sole-source option.
The figure presents three dual-source strategies. The first such option, represented by the three bars on the left (dark blue), is competitive procurement of whole ships. It is expected that competitive
pressures inherent in this strategy would reduce the actual costs, and
our analysis suggests that there is roughly a 50:50 chance that competitive pressure would overcome the cost penalties of from 10 to
over 12 percent shown in the figure. However, such cost reductions
are not ensured, and the split of production between two competitors could result in a cost increase over the sole-source option.
RAND MR1486-4.7
14
Cost penalty (%)
12
10
8
6
4
2
0
BAE
SYSTEMS
Marine
Proposal
8 to BAE
SYSTEMS
Marine,
4 to VT
6 to BAE
SYSTEMS
Marine,
6 to VT
4 to BAE
SYSTEMS
Marine,
8 to VT
Alliance
Block
Option I
Allocation strategy
Figure 4.7Summary Comparison of Cost Penalties of Seven

Procurement Options
Block
Option II
57
The second procurement option, in which the first three ships would
be a directed split between the two firms, with the remaining nine
ships competed, is represented by the fourth bar from the left
(medium blue). The bar reflects the estimated cost penalty for the
Alliance strategy. If it is expected that the cost penalty of about 10
percent is to be recovered through competition, then a cost reduction of about 13 percent would be required across the nine competed
ships.
The third procurement option, wherein each firm produces a specified set of blocks for the entire 12 ships, provides no opportunity for
competitive recovery of the cost penalty. However, the penalty is
smaller than those for the other options and might be considered a
reasonable cost to pay to obtain other possible benefits of that
option.
OTHER FACTORS AFFECTING THE TYPE 45 PROCUREMENT

DECISION
The analysis and results described above focused entirely on how the
procurement strategy might affect the production cost of the Type
45. However, the choice from among procurement strategies will
have other effects, some within the Type 45 programme and some
outside the programme. Our analysis of these other likely effects
risks and rewardshas been less extensive than that for Type 45
production costs, but it provides some additional insights to support
a decision on the overall Type 45 procurement strategy.
Risks and Rewards to the Type 45 Programme

The main advantage to the Type 45 programme of the sole-source
option is that the costs are within the range deemed affordable by
the MOD. The lower estimated costs of the sole-source option are
primarily due to the economies of scale of having all the ships built
by a single shipbuilder. As discussed earlier in this chapter, any
competitive strategy would involve splitting production between two
shipyards, leading to a directly estimated cost greater than for a single source, but with the expectation that competition would result in
some reduction of those costs. Our calculations show that there is an
even chance that competitive pressures would lead to costs no
58
greater than the sole-source option. However, such an outcome is

not ensured. Thus, selection of a competitive, dual-source strategy is
an attractive option, but not one without cost risks. If competition
yields little or no cost savings, programme execution would be more
difficult within the projected budget.
The option of having specified shipyards build the entire production
run of some blocks should lead to costs that are slightly higher than
the sole-source option but lower than the estimated cost of competitive dual sources for the entire ship, assuming little or no cost reduction through competitive pressures. The cost of a directed buy of
blocks from two shipyards falls within the range deemed affordable
by the MOD, retains the advantages of sustaining two shipyards, and
is not critically dependent on subsequent, and uncertain, reductions
of cost through competitive pressures.
Several other factors must also be considered when deciding on an

acquisition strategy for the Type 45 programme. These are discussed
next.
Innovation. With a guarantee of all 12 ships, the shipbuilder may
have little incentive for innovation or to find ways to reduce costs.
This situation is also true for the block options, since each shipbuilder is guaranteed specific portions of the total programme. Directed buys may influence the shipbuilders to be innovative in
reducing costs if they perceive the opportunity to gain a larger share
of the remaining ships in the programme. Competition should foster
the greatest degree of innovation as each shipbuilder strives to attain
more-efficient building techniques that will result in lower costs and,
it is hoped, larger portions of the total programme.
Multiple Sources. The sole-source option leaves little or no alternative if the sole shipbuilder has problems: There is no other shipbuilder in the programme to turn to. The block options improve the
situation somewhat, because at least two shipbuilders are involved in
the programme. However, each shipbuilder would have to learn
and become proficient in building the blocks that were assigned to
the other shipbuilder. Competitive environments and the directedbuy option result in both shipbuilders being capable of building the
entire ship. These options provide the highest assurance to the MOD
59
that the programme can continue with little disruption if one of the
shipbuilders experiences problems that prevent it from building the
ships in a timely and efficient manner.
Leverage on the Shipyards. Under the directed-buy option, the
MOD and the Type 45 PCO may have little leverage over the shipbuilder once it knows that it will have no competition in the programme. The leverage over the shipbuilder increases with both the
directed-buy option and the block option. With both shipbuilders
involved in the programme, even when the total programme is allocated in some way, the MOD and PCO have the option of reallocation of workloads to use as leverage. The competitive options provide the highest degree of leverage, because the allocation of future
work is uncertain and depends on cost and technical performance.
Commonality of Ships. Using a single shipbuilder, either for the
whole with the sole-source option or with portions of the ship with
the block option, ensures that all 12 ships will have a high degree of
commonality. With the directed-buy and competitive options, two
different shipbuilders are producing ships and there may be differences, or a lack of commonality, across the ships in the Type 45 fleet.
Coordination and Integration of Multiple Shipbuilders. In addition
to commonality of all the ships in the programme, the sole-source
option results in the MOD and the PCO having to deal with only one
shipbuilder. This aspect should result in improved coordination and
integration during the total programme. With the other options, the
MOD and PCO must interact with two shipbuilders, coordinating
and integrating their activities to ensure that the programme stays on
schedule and within budgets.
Coordination and integration become most difficult with the block
options, for which the timing of the construction and transport of the
blocks to the assembly site must be closely managed to ensure that
there are no delays in the build schedule. Also, the dimensional control of the blocks must be closely monitored to ensure that the blocks
fit correctly during assembly and that additional man-hours for rework are minimised or eliminated.
Collocation of Production and Support. Keeping VT involved in the
Type 45 programme through either direct buys (of whole ships or
blocks) or competition will mean that aspects of the production and
60
support of the Type 45s will be collocated at Portsmouth. This

should lead to reduced life-cycle support for the ship.
Risks and Rewards to Other Programmes

In addition to the Type 45 programme, the acquisition strategy chosen by MOD will affect, both positively and negatively, other shipbuilding programmes. As our analysis suggests, the sole-source option will result in reduced overhead costs for the Astute programme
as more workload at BAE SYSTEMS Marine reduces overhead rates.
Directed buys or competition will result in lower reduction of Astute
overhead rates. The sole-source and block options, whereby the
ships are assembled at Barrow-in-Furness, may also alter the Astute
production process. Both the Astute and the Type 45 will be built in
the Devonshire Dock Hall. With all 12 Type 45s going through the
close confines of the DDH, chances will arise for scheduling problems or for minor accidents to disrupt either or both production
lines.
If VT builds more whole ships through the directed buy of whole
ships and the competition options, the potential for problems in the
DDH would be lessened. Also, with the direct buy of whole ships and
competition, VT remains a builder of warships. For future programmes such as the FSC and the CVF, VTs continued presence increases the chance of VT being in the market to compete for these
programmes. Sole-source and, to some degree, the directed buy of
blocks may not allow VT to maintain its warship construction capabilities, leading to future programmes facing a monopoly in warship construction.
Risks and Rewards to the UK Shipbuilding Industrial Base

Sole-source production of the Type 45s might have negatively
influenced the UK shipbuilding industrial base. VT might have
decided to exit the shipbuilding business or might have forgone its
move to Portsmouth.
Directed buys and competition should help the UK shipbuilding industrial base become healthier and more robust. Although the future
of BAE SYSTEMS Marines Clyde shipyards could be threatened if it
61
does not receive a significant portion of the Type 45 work, keeping

VT active in building warships will be positive for future MOD programmes. A caveat must be stated for the option of a directed buy of
blocks, however: Once that paradigm is chosen, it may be difficult to
choose another paradigm for future programmes.
OVERALL INTEGRATION OF ANALYSIS AND

OBSERVATIONS
The various advantages and disadvantages of the different acquisition options are summarised in Figure 4.8.
It is apparent from the figure that none of the options is dominantly
superior to the others for Type 45 procurement. Sole-source procurement has an apparent advantage in near-term costs. However,
those costs might grow in the absence of continuing competition,6
and the option suffers in other areas. Selection from among the
other options depends in large part on the weights the decisionmaker assigns to a criterion relative to the weights for other risks or
rewards in a column. Thus, this analysis is informative but not absolutely conclusive.
______________
6 See Appendix B for an estimate of why and how costs might grow if competition in
naval shipbuilding is lost.
62
RAND MR1486-4.8
Acquisition Strategy
Sole
Source
Risk/Reward
CompetiAlliance
tion
?
Cost
Directed
Buy of
Ships
Directed
Buy of
Blocks
Innovation
Multiple sources
Leverage on yards
Commonality of ships
Coordination and integration of
multiple firms
Collocation of production and
support
Astute costs
VT move to Portsmouth
Production interference in DDH
FSC/CV(F)
Industrial base health and
diversity
Low risk or
high reward
Medium risk or
medium reward
High risk or
low reward
Unknown
outcome
Figure 4.8Summary of Risks and Rewards of Alternative Acquisition

Strategies for the Type 45
Chapter Five
THE APPROVED TYPE 45 PROGRAMME
Members of the RAND research staff presented the analysis described in the preceding chapters to senior managers in the MOD
during the first half of June 2001. On 10 July, the Secretary of State
for Defence, Mr. Geoffrey Hoon, announced the government decision on the programme to the House of Commons:
. . . Working with the companies, we have developed a revised strategy, which allocates work on the ships between the two shipbuilders
for the whole class of type 45 destroyers. The first-of-class ship will
be assembled and launched at Scotstoun. The focus of design
support to the whole class will remain there, with continuing
participation by both shipbuilders. The remaining ships will be assembled and launched at Barrow-in-Furness.
Vosper Thornycroft at Portsmouth, and BAE Systems Marineon
the Clyde and at Barrow-in-Furnesswill both build and outfit
substantial sections of each ship. The yards will continue to build
the same sections throughout the programme, to increase efficiency
and produce better value for money for the taxpayer.
Under the strategy, a commitment has now been made to six ships
in a planned class of up to 12 ships. That commitment has therefore doubled the number on order. This larger volume of guaranteed work, and a stable foundation to the project, will allow industry
to make long-term investment decisions.
Subject to negotiations being completed satisfactorily, I propose to
adopt this revised approach, through which we are confident that
63
64
we can secure demonstrable value for money. We are seeking demanding efficiency improvements from industry. The initial findings of the RAND study support this new approach. It reflects the
best features of the BAE Systems Marine bid in terms of learning
from experience from one ship to the next, but it also preserves the
possibility of competition for a number of subsequent defence programmes. The new strategy gives a welcome level of stability to our
warship building industry. Above all, it offers the best prospect of
achieving the in-service date for the type 45 destroyer, with deliveries to the Royal Navy starting in 2007. Any significant delay in that
date would have significant operational and cost penalties.
The planned distribution of work between the shipyards is shown in

Figure 5.1. The ship will be made up of six blocks plus the superstructure. Vosper Thornycroft will build two blocks (E and F) plus
the superstructure; the BAE SYSTEMS Marine Clyde shipyards will
build two blocks (A and D); the BAE SYSTEMS Marine Barrow shipyard will build two blocks (B and C). VT and the Clyde shipyards will
transport their blocks to the Barrow shipyard, where the whole ship
will be assembled (the first-of-class will be assembled at Clyde
[Scotstoun]). BAE SYSTEMS Marine Barrow will conduct the ship
trials, and Vosper will conduct the combat system trials. Table 5.1
details this allocation of work.
This solution offers a number of economic advantages. Spreading
the Type 45 work between BAE SYSTEMS Marine and VT helps
RAND MR1486-5.1
MASTS
Vosper Thornycroft
BAE SYSTEMS
Marine Barrow
BAE SYSTEMS
Marine Clyde
BLOCK A
MEGA-BLOCK B + C
BLOCK D
BLOCK E BLOCK F
Figure 5.1MODs Type 45 Block-Production Allocation
The Approved Type 45 Programme
65
Table 5.1
MOD Type 45 Work-Allocation Strategy Announced 10 July 2001
BAE SYSTEMS Marine Barrow
BAE SYSTEMS Marine Clyde
Vosper Thornycroft
First-of-class
Perform design work
Design centre
Produce all basic steelwork
for Marine
Build Clyde blocks
outfitted to about 40%
Ship all Clyde blocks to
Clyde (Scotstoun)
Build Barrow blocks
(main machinery spaces)
from steel units from Clyde
(Govan)
Perform design work

Build forward section
of the ship and
masts/funnels,
outfitted to about 80%,
and barge to Clyde
(Scotstoun)
Conduct Stage 2 trials
(combat system
performance)
Receive VT elements;
assemble and launch ship
Conduct Stage 1 trials (ship
performance)
Rest of class
Build Barrow blocks (main
machinery spaces) from
steel units from Clyde
(Govan)
Receive Clyde blocks from
Clyde (Govan) and forward
section and masts/funnels
from VT; assemble and
launch ships
Design centre
Perform design work
Produce all basic steelwork

for Marine
Build forward section

of the ship and
masts/funnels,
outfitted to about 86%,
and barge to Barrow
Build Clyde blocks

outfitted to about 80%
Ship all Clyde blocks to
Barrow
Conduct Stage 2 trials

(combat system
performance)
Conduct Stage 1 trials (ship

performance)
RAND MR1486-T5.1
ensure that both shipbuilders will remain viable and able to compete
on future MOD programmes. The strategy also facilitates the VT
move to Portsmouth while helping secure the future of the Clyde
66
shipyards. Having a single shipbuilder construct all of the same

blocks for the Type 45 class takes maximum advantage of the learning effect and, therefore, reduces costs from what they would be if
the construction of complete ships were distributed between the two
shipbuilders.
Building blocks at multiple sites does have potential disadvantages.
Additional costs will be associated with constructing the blocks with
enough rigidity and weatherproofing to permit movements and
transportation in such a way that the structural tolerances and integrity of the blocks will not be impaired. Structural tolerances must
be managed very closely; misalignment of adjacent blocks can lead
to substantial rework costs. Also, costs will be associated with transporting the blocks. Finally, scheduling of the construction and delivery of the blocks must be closely managed. A block that arrives late
at the assembly yard may cause significant delays in not only the
Type 45 programme but also the Astute programme.
This MOD decision will have direct consequences for shipyard modernisation actions. BAE SYSTEMS Marine will need to make investments at each of its facilities to accommodate the distribution of
work shown in the table: The first Type 45, HMS Daring, will be assembled and launched at Clyde (Scotstoun). All subsequent Type 45s
will be assembled and launched at BAE SYSTEMS Marine Barrow.
The companys Clyde (Govan) shipyard will produce major steelwork
for all of the ships.
In an interview shortly after the governments announcement, BAE
SYSTEMS Marines managing director, Simon Kirby, spelled out
these investments (Industry Update, September 2001, p. 42):
Barrowupgraded facilities for construction of Blocks B and C
Clydefabrication facilities and block-transfer facilities (Govan);

new door and cranes in the module hall, upgrades to the pipe
shop, and new bending machinery (Scotstoun).
The revised Type 45 strategy will allow VT to proceed with plans to

invest in a new shipbuilding facility within Portsmouth Royal Naval
Base. Since spring 2000, the company had been planning to shift
shipbuilding operations to Portsmouth, but had delayed committing
to this move until the MODs determination of its role in the Type 45
The Approved Type 45 Programme
67
programme. Shortly after its role as a producer of major modules of

the warship was clarified, VT announced that it would go ahead with
the new facility, which will be built on four existing docks in the
Portsmouth base. This construction will have the added benefit of
reducing overhead costs on ship repair work to be performed at
Portsmouth.
Chapter Six
ISSUES REQUIRING FURTHER STUDY
The Type 45 programme is one part of a larger acquisition programme laid out by the MOD for the Royal Navy and expected to
extend over the next two decades. The decisions on the Type 45
procurement strategy will affect those future programmes, and the
MOD will face a number of additional issues as the larger programme evolves. In this final chapter, we outline the following key
issues remaining for the MOD to address over the coming years:
What is the future manpower demand and supply picture?
Is the distribution of ownership and management responsibility

among the shipyards, PCO, and MOD appropriate?
Which acquisition strategies are most viable for future programmes?
How can innovation be encouraged in the absence of competition?
How can core industrial base capability best be sustained?
What are the issues when block production is performed separately from final assembly?
We explore these issues in turn in the following sections.
69
70
WHAT IS THE FUTURE MANPOWER DEMAND AND SUPPLY

PICTURE?
The current and potential UK military shipbuilders with whom we
spoke expressed confidence that they could rapidly expand their
workforces to accommodate increased workload. They also said that
this expanded workforce would require minimal, if any, training to
become as proficient as current shipyard workers. We are less sanguine than they. Although there is a surplus of qualified shipyard
workers today, the ambitious ship construction programme envisioned by MOD may very well exhaust the pool of experienced
workers.
As discussed in Chapter Two, the MOD is embarking on an ambitious shipbuilding schedule in the coming years. After 2006, most of
the military shipbuilding will be concentrated in just four programmes: Type 45, CVF, Astute, and FSC (see Figure 2.2). This concentration contrasts with the many smaller ship programmes being
undertaken at present and will have profound implications for UK
military shipbuilding. The smaller shipyards, which are incapable of
producing these larger ships, will find themselves without a direct
base of MOD work. To get smaller portions of work, such as modules, they will need to subcontract with the PCOs (not the MOD).
How this subcontracting will play out is unclear at present, because
one of the major PCOs has direct ties to several shipyards. By contrast, the larger shipyards should see strong workloads for the next
decade, particularly BAE SYSTEMS Marine Barrow and VT.
Figure 6.1 depicts the demand for direct-worker labour, engineering,
and support in military shipbuilding. The figure shows that the demand in that industry segment for direct workers will be fairly steady
for engineering and support. Note, however, that because some
shipyards include these workers as indirect labour, our values likely
undercount real demand. Yet, the trend is clear. Demand for bluecollar workers doing direct labour declines slightly between now and
2006, after which demand increases rapidly to a peak of nearly 7,000
direct workers in 2010almost double the level in 2005. Clearly, the
main risk for the MODs future programmes is maintaining and expanding the labour force in the next several years while managing a
small dip in demand.
Issues Requiring Further Study
RAND MR1486-6.1
8,000
Engineering
Labour
Support
7,000
Number of direct workers
71
6,000
5,000
4,000
3,000
2,000
1,000
0
2000
2005
2010
2015
2020
Year
Figure 6.1Estimated Future Demand for Direct Shipyard Workers

to Produce UK Military Ships
Data collected show that, as of fall 2001, direct-labour employment

stood at a little more than 5,000 workers. Given the projection from
Figure 6.1 that shipyards will need nearly 7,000 workers in 2010, there
will be a shortfall of nearly 1,800 workers if the shipyards do not increase staffing from current levels. Further complication the matter
is whether workers will be available in the geographic locations
where they will be needed. The shipyards were confident of being
able to attract workers from all over the country. We are less confident. Certain shipyards may find it difficult to staff appropriately.
Moreover, it is unclear how many workers will be available to move
to military shipbuilding from commercial shipyard work. In recent
years, commercial shipbuilding capacity in the United Kingdom has
fluctuated, with each succeeding peak lower by about 200,000 gross
registry tons (GRTs). Figure 6.2 shows commercial ship tonnage
produced by UK shipyards over the past two decades.
This figure does not necessarily reflect all UK shipbuilding. For example, it does not include warship building. But it illustrates the
problem: The tonnage being produced is declining. Estimates vary
72
RAND MR1486-6.2
600
Thousands of GRT
500
400
300
200
100
0
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
Year of build
SOURCE: The values are derived from the Lloyds Maritime Information
System database of all Lloyds Register ships built in the UK since 1980. Data are
current through October 2000.
Figure 6.2UK Commercial Shipbuilding Gross Registry Tonnage

Versus Year
for the number of workers remaining in shipbuilding trades at this

time. Armstrong Associates estimated recently that there are only
about 4,000 workers left in the nonmilitary shipbuilding trades, and
that there are another 10,000 working in building Royal Navy vessels
(Thales, 2001). The latter value corresponds to our shipyard survey
total of over 9,300. The UK Shipbuilders and Shipbreakers Association (SSA, 2001) is more optimistic, estimating that there are about
26,000 workers engaged in shipbuilding and ship repair.
Given such workforce issues, we recommend that the MOD consider
the following four strategies:
1. Encouraging PCOs to include options whereby the smaller shipyards can compete for subcontracts through the PCOs or from the
larger shipyards.
2. Requesting companies to keep industry informed on future plans
so that the government can implement shipyard training
programmes that will ensure adequate numbers of qualified
73
workers for the planned work; and assessing the effect of new
workers on overall productivity and projecting that effect onto the
cost and schedule of future programmes.
3. Detailing issues regarding the availability and cost of labour as a
function of the construction site selected and specifying how regional difficulties, if any, are to be overcome. Labour shortages
could drive up labour rates substantially.
4. Changing the start dates of future programmes to shift or lessen
the peak demand for workers.
IS THE DISTRIBUTION OF OWNERSHIP AND

MANAGEMENT RESPONSIBILITY AMONG THE SHIPYARDS,
PCO, AND MOD APPROPRIATE?
Both the Type 45 and Astute PCOs are part of BAE SYSTEMS, as is
one of the PCOs competing for the CVF. The PCO concept envisions
an organisation independent of shipyards among which the PCO is
expected to conduct a vigourous competition. BAE SYSTEMS also
controls three shipyards that perform the majority of military ship
construction. Many of the nonBAE SYSTEMS personnel and shipyard management teams with whom we spoke were quite critical of
having the PCOs housed in the same company as the shipyards, and
clearly did not have confidence that sufficient internal company
firewalls could be erected and/or sustained between the BAE SYSTEMS PCOs and shipyards.
The appearance of a conflict of interest and possible compromise to
fairness, if not an actual conflict of interest, raises an issue that needs
additional consideration and vigilance by the MOD.
In addition, the PCO approach is a new acquisition strategy for the
MOD. As experience is gained, further modifications/improvements
in this strategy may be required.
WHICH ACQUISITION STRATEGIES ARE MOST VIABLE FOR

FUTURE PROGRAMMES?
After examining the relative costs and other consequences of four
different acquisition strategies applied to the Type 45 procurement
74
sole-source, directed buy of whole ships across two firms, competitive procurement of whole ships, and directed buy of ship blocks
we drew two conclusions in view of the characteristics of the Type 45
programme and the present status of the naval shipbuilding industry
in the UK: No single strategy was predominantly superior, and selection of a preferred strategy depends on the judgement of senior managers in assigning relative importance to various strategy attributes.
However, these conclusions do not necessarily apply to other programmes to be conducted in the future. No particular procurement
strategy can be claimed as best until the particular situation has been
examined.
The selection of a preferred strategy for future programmes will depend, in large part, on two factors: number of ships to be procured
and distribution of other business across the available shipyards.
Number of Ships to Be Procured

Because of loss of learning effects, splitting large production runs
among two or more producers can incur large cost penalties that
might, or might not, be recovered through competitive pressures.
For very small production quantities, say, two or three ships, the absolute cost of learning effects is smaller; other shipyard costs, such as
start-up investment and sharing of overhead, become more important.
Distribution of Other Business Across Available Shipyards

Ship construction is labour-intensive, but labour is not readily
transferred from one shipyard to another. Thus, the costs and benefits of placing a particular ship-construction project at a particular
shipyard can depend heavily on the shipyards other current and
projected business.
HOW CAN INNOVATION BE ENCOURAGED IN THE

ABSENCE OF COMPETITION?
Consolidation within the military shipyards, which reduced overhead and facility costs, has resulted in significant savings for MOD.
However, a more concentrated defence sector may be less innova-
75
tive, the smaller number of remaining firms perceiving a reduced

need for new ideas in order to win contracts. Contractors may also
be reluctant to pursue innovations that may cannibalise lucrative
existing markets (i.e., they do not want to develop competing
products within the existing market). Increased MOD emphasis on
cost control may discourage firms from undertaking technically risky
options out of fear of possible failures and consequent adverse costs.
Also, with less emphasis on innovation, contractors may assign their
best staff and most fertile minds to tasks other than advanced
defence R&D; the best minds, in turn, may seek out smaller, more
entrepreneurial firms in other industries, thereby leaving the defence
industry altogether.
Arguably, reduced acquisition budgets make innovation more important than ever. Smaller military forces must perform more demanding roles. Hence, MOD must be, and is, interested in strategies
for encouraging innovation in this changing defence sector.
RAND studies of military contractors behaviour (Birkler et al., 1994,
1998) suggest that, in the past, innovation has come from two
sources: (1) a large number of independent contractors engaged in
dynamic and intense competition to win the major contracts and (2)
a second tier or marginal set of prime contractors willing to take
more technological and financial risks in order to break into the first
tier of prime contractors. As shipbuilding has consolidated, these
traditional sources of innovation have disappeared or are in the
process of disappearing. Even more worrisome, the new consolidated primes usually reduce R&D spending and the number of
suppliers even more.
Many industrial sectors in the United States have seen significant
consolidation in recent years, but without a concomitant reduction
in innovation. Production (and marketing) activities are increasingly
consolidated in the pharmaceutical and biotechnology industries, in
the telecommunications sector, and in some segments of the computer hardware and software industries. In all of these sectors, technological innovation has nonetheless continued at a very rapid pace.
Popular wisdom suggests that much of this innovation can be attributed to the separation of production from R&D in each of these
sectors for practicality. Small, R&D-oriented firms are easily established and compete fiercely to produce the next technological ad-
76
vance. Successful innovators license or sell their discoveries or are

themselves bought by the few consolidated firms that dominate production and marketing.
A version of this pattern of innovation has been seen in the cruiseship industry, in which subsystem contractors are competed, with
the effect that technical risks and costs to the primes are reduced.
The MOD has an interest in capturing the efficiencies that come
from consolidation of the military shipbuilding production base and
from sustaining innovation. The MOD would be well served by
gaining a better understanding of the effects of defence industry consolidation on innovation and of how other rapidly consolidating industries have managed to sustain and encourage robust innovation.
HOW CAN CORE INDUSTRIAL BASE CAPABILITY BEST BE

SUSTAINED?
Although investments are still required, the Type 45 decision sustains
the industrial base over the next decade for its size and class of warship. BAE SYSTEMS Marine is in the process of making capital investments and reorganising and optimising facilities for its current
workload. By moving to Portsmouth, VT can construct larger ships,
on the scale of the Type 45, and blocks than were possible at its
Woolston facility and keeps open the opportunity to participate in
Type 45 production and to compete for other programmes in the
future. Thus, the Type 45 solutions enhance and preserve vital elements of the industrial base. However, the availability of a qualified
workforce as the programme moves to rate production is still an active issue, especially in light of the workforce required for the CVF
programme.
The proposed size of the future carriers is believed to be beyond the
current production capability of any single UK shipyard. The current
plan is that major portions, or blocks, of the carriers will be constructed in several shipyards and transported to one shipyard for final assembly. But only a few facilities are large enough to assemble
the ships, and each has shortfalls and constraints.
In addition to the problem of facilities and capacities within the current shipbuilding industrial base, any construction plans for future
77
carriers must include an assessment of how other current and future

shipbuilding programmes, both military and commercial, will curtail
or enhance the availability of workers and facilities.
Finally, given the potential need to involve and integrate several
shipyards in carrier construction, the MOD must resolve several
major issues: the scheduled start and completion of the carriers, the
gap between the start of the first and second ships, and the need for
an ensured facility for subsequent life-cycle support.
WHAT ARE THE ISSUES WHEN BLOCK PRODUCTION IS

PERFORMED SEPARATELY FROM FINAL ASSEMBLY?
Shipyards worldwide routinely build, outfit, and join together modules of 1,000 metric tons every day. Many very large marine structures have been built in ultra-large modules, which are subsequently
assembled to form one structure. But all of the examples of which we
are aware are far simpler structures than a surface combatant or aircraft carrier such as CVF. For both the Type 45 and CVF, subassembly fabrication and block construction will be done at locations remote from the final assembly point. In our judgement, the larger and
more complex the modules to be joined, the greater will be the difficulty of ensuring adequate mating and the risk of costly rework. The
shipyards must provide enough planning and analysis detail to
support confidence on three issues connected with modular
construction: block fabrication and assembly, out-of-sequence delivery of blocks, and time on assembly berth. Table 6.1 outlines these
issues and the actions that the MOD should require of the PCO.
78
Table 6.1
MOD-Required PCO Actions for Type 45 Block Construction
Issue
MOD-Required Actions
Block fabrication and

assembly
Demonstrate that the detailed design process,

construction process, and attendant transportation,
unloading, and mating processes for each block ensure
that the blocks will fit together; additional rework will be
minimal; and that successful delivery of the ships will
result within estimated costs and budgets.
Out-of-sequence
a
delivery of blocks
Describe the delivery sequence of blocks intended for

the PCOs selected production scheme and how delivery
delays of any component will be accommodated. This
description shall include how out-of-sequence delivery
will be accommodated at the final assembly yard and how
the accommodations affect cost and final ship delivery.
Time on assembly
berth
Show, by detailed analysis of the work required, how the

assembly in-dock will be accomplished in the time
allowed in the dock.
aUnderstanding out-of-sequence delivery of blocks is especially important for the

Type 45 programme, since the Type 45 will be assembled in the DDH, along with the
Astute-class submarines.
RAND MR1486-T6.1
Appendix A
SENSITIVITY ANALYSIS
Throughout our evaluation on the acquisition options for the Type

45, we made many assumptions about the UK industrial base (as described in the main text). Without these assumptions, it would not
have been possible to make quantitative evaluations. However, the
reader might question the degree to which our results would change
if different assumptions were made. How robust are the results? Do
small changes in the assumptions dramatically change the conclusions? To determine how robust the results are, we change a number
of key assumptions about number of ships acquired, learning slopes,
CVF workload, and workforce productivity to see whether the outcomes are dramatically different.
NUMBER OF SHIPS ACQUIRED

We have assumed throughout the main body of the analysis that 12
Type 45 ships will be built over the next two decades. However, the
total number is far from certain. It will depend on many factors.
Typical of military acquisitions, budget shortfalls result in fewer systems being purchased than originally planned. Likewise, priorities
may change over the long production time, leading to fewer or more
systems being required. To understand the effect of changing the
number of ships acquired during the Type 45 programme, we varied
the number from three to 15 total ships.
Figure A.1 shows the cost penalty (over the entire Type 45 programme) to break even for an acquisition programme of nine ships.
The values are somewhat lower than those observed for the 12-ship
programme (see Figure 4.7).
79
80
Figure A.2 shows the cost penalty for the Alliance strategy at various
build levels for the Type 45 only and for all programmes. As can be
seen in the figure, the cost penalty increases with the number of
ships, but nonlinearly. It tends to flatten as the number of ships
increases.
This does not mean that it is easier to achieve savings through competition as the number of ships in the programme decreases. In fact,
the reverse is true for the Alliance strategy. Recall that, under the Alliance strategy, the first three ships are allocated and the remaining
ships are competed in lots of three. Thus, the benefits of competition are not fully gained until the second and subsequent lots. If we
adjust the cost-penalty levels to account for the number of ships
actually competed, the trend is rather flat but decreases somewhat
with the quantity (Figure A.3; see also Appendix B).
LEARNING SLOPES
For the production of the Type 45, we assumed the learning slopes to
be those that BAE SYSTEMS Marine and Vosper Thornycroft re-
RAND MR1486-A.1
14
Cost penalty (%)
12
10
8
6
4
2
0
BAE SYSTEMS
Marine Only
Block Option I
Block Option II
Alliance
Type 45 procurement options
Figure A.1Cost Penalty for a Production Run of Nine Ships
Sensitivity Analysis
RAND MR1486-A.2
12
All programmes
Type 45 only
10
Cost penalty (%)
81
8
6
4
2
0
3
12
15
Number of Type 45 ships produced
Figure A.2Cost Penalty for the Alliance Strategy for Varying Levels of
Procurement
Competition savings required

on procured lots (%)
RAND MR1486-A.3
18
16
14
12
10
8
6
4
2
0
6
12
15
Number of Type 45 ships produced

NOTE: We included a production run of 15 Type 45s to better understand the cost
sensitivity connected with a large total programme size.
Figure A.3Competition Cost Penalty for the Alliance Strategy for Varying
Levels of Procurement
82
ported, which were not identical. The difference was not surprising,
considering that the two yards have different work-breakdown and
cost structures. Would our results change if we assumed an equivalent learning slope at each yard?
Figure A.4 shows the cost penalty for each of the seven strategies and
different learning assumptions. For the common learning slope, we
assumed a 90-percent slope, which is slightly higher than the average
of the values from the two yards. For this common case, the cost
penalty is lower by approximately 1 to 4 percentage points,
depending on the strategy. All values are within 8 percentage points.
This lowering of the cost penalty is consistent with the overall higher
learning slope discussed in Chapter Four.
CVF WORKLOAD
Another assumption we made was that VT would perform onequarter of the CVF production work. In the data provided to RAND,
VT did not speculate on the level of work for CVF production,
RAND MR1486-A.4
14
Unique slopes
Cost penalty (%)
12
Common slope
10
8
6
4
2
0
BAE
SYSTEMS
Marine
Proposal
8 to BAE
SYSTEMS
Marine,
4 to VT
6 to BAE
SYSTEMS
Marine,
6 to VT
4 to BAE
SYSTEMS
Marine,
8 to VT
Block
Option I
Block
Option II
Alliance
Figure A.4Cost Penalty with Different Learning Assumptions
Sensitivity Analysis
83
because the programme was in the early design phases. Do the

results change if we remove the CVF workload from VT? According
to Figure A.5, the answer is, Very little.
WORKFORCE PRODUCTIVITY
In other work by RAND (Birkler et al., 1994, 1998), the authors have
found that productivity changes in the workforce can dramatically
alter the cost of production for ships. For the study on the Type 45,
we assumed that new, unskilled workers start at 67 percent
proficiency and linearly improve over time to 100 percent at the end
of three years. Furthermore, we assumed that the pool of new hires
has an equal distribution of workers over the proficiency range of 67
to 100 percent.
However, the shipbuilders opined that getting additional, fully productive workers would not be difficult. In essence, they anticipate
that all new hires will be fully productive. Figure A.6 shows that
these assumptions on productivity do not change our results substantially.
RAND MR1486-A.5
14
Cost penalty (%)
12
CVF at VT
No CVF at VT
10
8
6
4
2
0
BAE
SYSTEMS
Marine
Proposal
8 to BAE
SYSTEMS
Marine,
4 to VT
6 to BAE
SYSTEMS
Marine,
6 to VT
4 to BAE
SYSTEMS
Marine,
8 to VT
Block
Option I
Block
Option II
Alliance
Figure A.5Cost Penalty with Different CVF Workload Assumptions

at VT
RAND MR1486-A.6
14
No productivity
penalties
12
Cost penalty (%)
84
10
Including assumed
productivity penalties
8
6
4
2
0
BAE
SYSTEMS
Marine
Proposal
8 to BAE
SYSTEMS
Marine,
4 to VT
6 to BAE
SYSTEMS
Marine,
6 to VT
4 to BAE
SYSTEMS
Marine,
8 to VT
Block
Option I
Block
Option II
Figure A.6Cost Penalty with Different Workforce-Productivity

Assumptions
Alliance
Appendix B
LONG-TERM IMPLICATIONS OF A LACK OF

COMPETITION
Our analysis of the cost and benefits of the various acquisition

strategies for the Type 45 programme looked only at the short-term
implications. For the BAE SYSTEMS Marine unsolicited offer (solesource), the production cost was lower than that for the other
directed-allocation strategies. However, although the sole-source
might appear to be the least expensive procurement strategy in the
short term, it poses considerable long-term drawbacks. The primary
drawback is that selecting the sole-source strategy could result in a
lack of competition in UK surface combatant production (BAE
SYSTEMS Marine will have a virtual monopoly). How might a lack of
competition affect ship procurement costs in the long term?
In Chapter Four, we observed that competitive ship and missile programmes, on average, cost about 7 percent less than noncompetitive
programmes (see Figure 4.3).
Another cost of noncompetitive industries is a higher rate of price escalation. Figure B.1 shows the consumer price growth in five different industries: drugs and pharmaceuticals, cars, electricity,
petroleum (gasoline), and airfare. As a reference, we also show the
Consumer Price Index (CPI). We obtained industry and CPI data
from the U.S. Bureau of Labor Statistics. As is readily evident, the
drug and pharmaceuticals, and airfare indices have the highest rate
of escalation over the 11-year period. They averaged 1.7-percent
greater annual escalation than the CPI. We use this 1.7-percent
additional escalation rate as an example to illustrate the possible
consequences of reduced competition in future UK shipbuilding.
85
86
RAND MR1486-B.1
1.9
Airfare
Relative price growth since 1989
1.8
1.7
Drugs and
pharmaceuticals
1.6
1.5
Petroleum
(gasoline)
CPI
1.4
1.3
1.2
Cars
Electricity
1.1
1
1985
1990
1995
2000
2005
Year
SOURCE: U.S. Bureau of Labor Statistics.
Figure B.1Relative Price Growth in Various Industries from 1989 to 2000
To quantify the effects of greater procurement cost and greater price

escalation on the MOD, we assumed that warship production after
completion of the Type 45 programme (about 2014) would continue
at the annual investment rate of that year and would be
noncompetitive. All procurement costs from that point would be 7
percent higher and escalate at a 1.7-percent higher rate. When doing
a cost-benefit analysis of potential future savings, an accepted
practice is to use techniques such as discounting (net present value)
as a metric to quantify benefits. This method balances the fact that
cash flows do not occur in the same year. Because there is an
opportunity cost of capital, receiving 1 today is more desirable than
receiving 1 five years from now. This desirability does not relate to
inflation; rather, it reflects the fact that if we forgo use of capital, we
lose the ability to spend it on something else. Likewise, we gain
benefit if we have the ability to use it sooner. Therefore, it is an
opportunity cost.
Figure B.2 shows how the net cost savings from sole-source production erodes relative to a strategy whereby each shipyard builds six
Long-Term Implications of a Lack of Competition
87
ships, thereby maintaining two shipyards that could compete for

future programmes. The net production savings for the sole-source
case is the cumulative savings through a sole-source procurement
(negative means savings). It is generally cheaper to produce in the
sole-source case, but that savings gets eroded through the greater
escalation and the future greater procurement costs (the 1.7 percent
and 7 percent, respectively). Undiscounted means that no discount
rate was applied (0 percent). The 4-percent discount rate means that
a cost of capital (discount rate) of 4 percent per annum was applied.
Two curves are shown. The black curve displays the net savings
undiscounted; the blue curve shows the net savings discounted at 4
percent per year.
Notice that the most savings is achieved by 20122013 (the end of the
Type 45 production). After that point, ship-production cost becomes
more expensive (owing to a lack of competition). The break-even
points, at which all the savings from sole-source production is
eroded, are about 2021 undiscounted and 2026 discounted. From
then on, the sole-source strategy for the Type 45 results in higher
costs for future programmes.
RAND MR1486-B.2
Net cost with sole-source strategy (M)
400
300
Undiscounted
200
100
0
4% discount
rate
100
200
2000
2010
2020
Year
Figure B.2Net Cost Resulting from a Lack of Competition
2030
Appendix C
BUILDING SHIPS IN BLOCKS AT MULTIPLE SITES:

IMPLICATIONS FOR THE TYPE 45 PROGRAMME 1
One of our research tasks was to better understand the extent to

which a strategy of building ships in large blocks at multiple sites is
used in commercial and military shipbuilding. Another task was to
describe the potential advantages and disadvantages of block construction for the Type 45 programme. This appendix presents the results of our survey of current shipbuilding practices, describing
precedents for block and multiple-site construction.
This appendix uses the following terminology:
Interim product: a level of the product structure; it is the output

of a work stage and is complete in and of itself.
Part: the lowest level of the interim product structure. It can be

either manufactured by the shipyard or purchased. Examples of
shipyard-manufactured parts are a part cut from a plate, a stiffener cut from a tee profile, and a pipe piece cut from a pipe
length. Examples of purchased parts are main engine, steering
gear, and pumps.
Subassembly: a small interim product, relative to an assembly,

that is made from parts. It generally consists of one or a few
plate parts with stiffener flanges connected to them.
______________
1 This appendix summarises material prepared for the study by Tom Lamb, P.E., EUR
ING, Technical Associate, Innovative Marine Product Development, LLC, Ann Arbor,
Michigan.
89
90
Assembly/panel: an interim product that is made from subassemblies and parts. It generally consists of a single skin
(plating) that has stiffeners and web frames connected to it.
Unit: a structural interim product made from assemblies, subassemblies, and parts. It generally is a three-dimensional structure, such as the shell, bulkheads, and decks, having assemblies
joined perpendicularly to each other.
Block: a number of units joined together.
Grand block: a large ship section made from two or more blocks
before it is erected in the building berth.
Ring section: a block or grand block that extends from the keel to
the main deck and from one side of the ship to the other.
Machinery module: a group of outfit items made into a selfcontained package consisting of the support framework, grating,
equipment, pipe, controls, etc.
Zone: a geographical portion of the ship, such as the bow or a

package of machinery.
In the next section, we employ these terms to describe the advantages and disadvantages of block construction.
BLOCK SHIP-CONSTRUCTION STRATEGIES

Until about 1950, ships were generally built piece by piece (similar to
a building site), an approach with a very low investment cost, minimum crane requirements, and small-scale transport. Today, the
piece-by-piece approach is rarely used, except in undeveloped
countries, where the labour cost is low, and in developed countries
for some small craft and one-of-a-kind ships. Figure C.1 shows the
piece-by-piece construction of a small craft.
Over the past half century, shipbuilders have recognised the advantages of building larger portions of ships in covered production facilities, then assembling those portions in a dry dock or on a slipway.
As larger numbers of similar ships are built, standardisation, repetition, and automation lead to economies of scale and production ef-
Building Ships in Blocks at Multiple Sites
91
Figure C.1Piece-by-Piece Construction of a Small Craft
ficiencies, and, in turn, to lower costs and reduced schedules. But, as

the size of the modules built in production facilities grows, more
investment is needed for larger cranes and transporters, and much
more attention is required for configuration control to maintain the
build tolerances between the modules.
Assembly Construction
The ship-construction method that erects assemblies on building
berths, shown in Figure C.2, is common for small- to medium-sized
ships. It has the advantages of being somewhat faster and less expensive than piece-by-piece construction, requires minimal investment, and provides greater flexibility in the sequencing of construction. It has the disadvantages of still being fairly slow, having low
productivity and, hence, relatively high labour costs, and having the
need for rework built in.
92
Figure C.2Assemblies Being Erected on Berth
Two sites where assembly ship construction takes place are
Appledore Shipbuilders in Devon: one of the first modern

covered shipyards, it uses the assembly build technology typical
of 1970 (when it was developed).
ASEA shipyard in Sestao, Bilbao, Spain: an old riverbank shipyard, it has been modernised and extended for building larger,
modern ships. It is limited by the need to construct from relatively small assemblies.
Block Construction
Block construction, common for large vessels, also is used for some
smaller ones. Blocks can vary in size from approximately 50 tons for
93
small vessels to up to 400 tons for large vessels such as very large
crude carriers (VLCCs). A typical block is shown in Figure C.3.
Block construction has advantages that are more dramatic than
those for assemblies : higher productivity and, therefore, lower
labour costs. But blocks also entail disadvantages : the need for
highly accurate assembly, a larger investment cost in facilities, and a
very high reliance on control of accuracy and on on-time delivery of
materials.
Frigates being designed and constructed in Spain for the Spanish and
Norwegian navies use the block-construction approach, as shown in
Figure C.4. However, these ships are employing a more traditional
block approach, in which many small blocks (up to 100 tons) are
fabricated in the same shipyard.
Figure C.3Typical Structural Block
94
Figure C.4Spanish Frigate Structural-Block Breakdown
Other examples of block construction include the following:
Kvaerner Masa, Helsinki: a very effective builder of passenger

ships that uses substantial blocks (with significant pre-outfitting)
to exploit its covered dock.
Astilleros Espagnoles, Cadiz: a classic large, greenfield shipyard2

originally developed for VLCCs. It is equipped with Goliath
cranes capable of lifting 600 metric tons each.
Grand-Block Construction
Grand blocks can be built from either assemblies or smaller blocks.
Grand blocks are usually outfitted and painted in advance (see Figure C.5). The benefits attributed to grand blocks are reduced building-berth erection time and welding, easier access to blocks being
assembled into the grand blocks, and no need for staging on the
building berth.
One disadvantage of grand blocks is that they need to be moved to
the building berth via large-lift-capacity cranes or other means. They
also need to be aligned to other grand blocks.
______________
2A greenfield shipyard is a shipyard constructed from scratch on a site that did not
previously have a shipyard.
95
Figure C.5A Typical Grand Block, Comprising Seven Blocks
Ring Construction
Generally used with mid-sized vessels, the ring approach is not as
common as block construction. It has been used for some large
ships as well, as shown in Figure C.6.
The several advantages of ring construction are that it can be substantially completed under cover; it improves productivity and lowers labour costs; and it allows production from assemblies, which
provide flexibility to recover from inaccuracies, or from blocks. The
disadvantages of ring construction are that it requires substantial investment costs and has an inflexible sequence of work, making it totally dependent on timely deliveries.
96
Figure C.6A Typical Ring Section
Sites employing ring construction include the following:
Electric Boat and Newport News Shipbuilding in the United

States, and Barrow in the United Kingdom: both build submarines by the ring approach.
Litton Industries, Pascagoula, Mississippi: developed in the

1960s specifically to exploit this method of building for frigates
and destroyers.
BAE SYSTEMS Marine Clyde (Govan): built liquid-nitrogen gas

(LNG) ships from ring sections of up to 3,000 metric tons. BAE
SYSTEMS Marine Barrow has since adopted ring construction for
the Fleet Replenishment Ship (AOR) and Landing Platform Dock
(LPD) contracts for the UK MOD.
EXAMPLES OF MULTISITE CONSTRUCTION

Once the techniques of building ships in large blocks were mastered,
several shipyards took the next step of building blocks at multiple
shipyards, then transporting them to a single shipyard for assembly
97
into a whole ship. This technique of assembling ships from large

blocks produced at different locations is more common than might
be expected, and it has both potential benefits and problems.
World War II Liberty ships may have been the first example of ships
constructed from pieces built at multiple shipyards. To produce
large numbers of ships in a short timescale, conventional shipbuilding techniques were not adequate. As the demand for naval ships
burgeoned, existing shipyard capacity could not keep pace. The use
of welding and prefabrication delivered the solutionassemblies
and pre-outfitting. Liberty ship sections were pre-assembled and
transported to an existing shipyard for final assembly and launching.
Despite the huge publicity about the success of these methods, the
costs were high and rework was common. However, these ships
demonstrated the feasibility of the method and met demanding ship
supply and delivery schedules.
After the end of World War II, shipbuilding returned to more
traditional practices. Occasionally, programmes used multisite
block-construction techniques. For example, in the early 1970s,
Seatrain Shipbuilding started the construction of large tankers while
the shipyard in the old Brooklyn Navy Yard was being refurbished.
To maintain its desired production schedule, Seatrain had all the
structural blocks for the cargo tanks for the first ship, and many for
the second ship, constructed by US Steel in Orange, Texas, and
barged to New York. Also at this time, Litton built the bow and stern
of a Great Lakes bulk carrier, joined them together, and sailed the
joined sections to Lake Erie, where they were cut apart and attached
to the cargo mid-body, which was built in a Great Lakes shipyard.
Other examples of building sections of ships at multiple shipyards
include the following:
Swan Hunter used multisite production to manage a large programme of tanker production in the mid-1970s, during which
time the company had several shipyards on the River Tyne
downstream from Newcastle, each specialising in different sizes
and types of ships. For the tanker programme, which had a large
number of ships and a short timescale, Swan Hunter pooled its
resources through multisite construction. It developed new
facilities at the Hepburn shipyard to provide a major new
98
steelwork facility built around an existing dry dock. It also used

existing, smaller steel facilities at the nearby Walker shipyard. To
meet the programme schedule, the tanker aft ends were
constructed and outfitted at Walker and towed to Hepburn, to be
joined to steel-intensive cargo sections and a bow that had been
constructed there. The forebody was built in the dry dock, taking
advantage of the new and extensive steelworking facilities. No
major problems were reported in aligning or in completing the
ships once the dock had been drained.
In the 1980s, Mitsubishi Koyagi, Hitachi Ariake, and Ishikawajima-Harima Kagoshima built eight large floating crude-oil
reservoir tanks. The IHI Kagoshima Works supplied all the large
blocks for the project, and final assembly was done at Koyagi and
Ariake, facilities located near each other in Kyushu. The blocks
were barged from the Kagoshima manufacturing facility to the
Ariake and Koyagi erection sites.
In Spain, the Sestao shipyard within the old Astilleros Espagnoles

group set up subcontractor facilities on its land, where it could
build structural blocks, thereby enabling it to reduce the number
of permanent employees.
BAE SYSTEMS Marine (formerly Marconi-Marine) has used multisite construction to build conventional (SSK) and nuclear
(SSBN) submarines. Specialised production facilities were available in two sites for the important pressure hull rings (of differing
sizes and configurations). To avoid wasting time or capacity,
BAE SYSTEMS Marine decided to manufacture the hull-ring
units at the most convenient site and transport them 100 miles
by road to the other facility, as required. The solution was successful, but probably at some additional cost, and the finished
submarines were satisfactory. Most important, the peaks and
troughs in production were avoided and the programmed
schedule was maintained.
Jumboising is fairly common in ship conversion. Typically, a

vessel to be enlarged is dry-docked and cut in half transversely
through all the structure and systems. Thereafter, the ship halves
may be separated and moved via hydraulic or other groundtransfer systems to other parts of the dry dock. The new midbody is built unit by unit, depending on crane capacity and the
99
space between the ship halves filled in by the newly built assemblies.
The Canadian frigate programme, which took place from the

late-1980s through the mid-1990s, used some multisite work.
The main contract was placed with St. John Shipbuilding in New
Brunswick. However, the programme had to have a second
source and St. John had to subcontract some of the ships to Marine Industries Limited (MIL), based in Quebec. MIL had two
sites at Quebec City and Sorel-Tracy. To meet the tightprogrammed requirements, MIL used both sites for blocks, but
completed final construction at Davie. Transport was of substantial, floating hull sections, which moved approximately 150
miles on the St. Lawrence River.
A programme around 1990 to construct new ferries for British

Columbia required a capacity for ship construction that was not
available locally in existing shipyards. To fulfill the strong desire
to retain the programme locally, an imaginative solution was developed. Three sites were identified, and the ferry construction
was divided into three ship sections. The hull was split into fore
and aft parts, and the superstructure formed the third major section. Each section was constructed at a separate site, after which
the hull halves were floated together and the superstructure was
put in place. The process was achieved with a limited expenditure for a single, two-ship contract. In fact, simple technologies
were used imaginatively to move substantial ship sections.
However, the approach was not cost-effective: The original
contract price was significantly more than the single-shipyard
bids from other Canadian and U.S. shipyards, and the projects
final cost was more than 216 percent of the contract price.
Newport News Shipbuilding subcontracted all of the deckhouses

for the Double Eagle product tankers it built in the 1990s. However, substantial rework costs were incurred when the deckhouses did not align properly with the ship body.
To build the Disney cruise ships, Fincantieri selected two shipyards previously not involved in the cruise-ship business. The
shipyard near Venice had both block-construction and finalassembly responsibilities.
100 The Royal Navys New-Generation Type 45 Destroyer
The Netherlands shipyards organise in geographical clusters of

shipyards that have considerable internal competition for domestic orders. However, they have a high degree of collaboration
for export orders. Such collaboration involves large-scale interyard subcontracting. As a result, relatively small shipyards are
able to obtain or share in significant orders, especially in terms of
the numbers of ships constructed. The shipyards are also able to
offer rapid delivery dates. One shipyard may accept an order,
then subcontract various blocks to other shipyards within the
cluster. Such subcontracting enables each to specialise in a particular block type. The process is applied to relatively small,
simple ships, but is extremely effective in minimising costs and
reducing timescales.
DANYARD in Denmark established a process whereby it constructed the mid-bodies of its ships in the Aarhus shipyard and
towed them after launch to the Fredrikshaven shipyard, which
had built the bows and sterns in parallel (Figure C.7). The deckhouses were subcontracted to a Polish shipyard (Figure C.8). The
Fredrikshaven shipyard connected the parts and completed and
delivered the ships.
Imabaris new, large Saijo shipyard opened in March 1995 as a

dedicated block-production centre with no final assembly capability. It was at that point a new generation factory built
with the idea of realising increased productivity and accuracy by
complete introduction of most advanced automatic equipment
and apparatus (sic) (Cooperative Association of Japanese Shipbuilders, 1996). Saijos building dock became operational in
2000, and the plant is now a full-fledged shipyard.
There are other examples of using multiple shipyards in the construction of ships. The following describe cases where one shipyard
builds the basic ship (i.e., the Hull, Mechanical and Electrical
[HM&E]) and a different shipyard performs many of the systemoutfitting functions:
The hull for the UK helicopter carrier, HMS Ocean, was constructed in BAE SYSTEMS Marine Clyde (Govan) shipyard and
sailed to BAE SYSTEMS Marine Barrow for combat system outfit-
Building Ships in Blocks at Multiple Sites 101
Figure C.7Stern Grand Block at DANYARD in Denmark
ting. This arrangement was reportedly based on a cost decision,

but was also necessitated because Clyde (Govan) had the better
steel-production facilities for this type of ship.
Many Floating Production, Storage and Offloading (FPSO) ships

are built in Asia but are sailed to the United States or Europe to
have all of the topside equipment installed. The Asian shipyards
do not have the experience to install such complex topside
equipment for the offshore oil-drilling industry.
Mitsubishi in Japan constructs hulls for warships in a small, traditional shipyard with end launching ways. It then tows the hulls
about 15 miles to the main Nagasaki shipyard, where they are
completed. This process serves to keep Mitsubishis warship
building as separate as it can be from its other shipbuilding
activities.
Figure C.8Subcontracted Deckhouse Constructed in Poland for

Denmarks DANYARD
Now that we have seen how shipbuilders spread work around multiple shipyards, we look at why they do so.
RATIONALE FOR MULTISITE CONSTRUCTION

Shipbuilders choose to spread work among various shipyards for
many reasons. Political motivations to spread work among different
areas of the country or to maintain an adequate number of firms in
the industrial base may be one reason. A number of recent U.S. Navy
shipbuilding programmes, the DDG 51 and LPD 17 classes and the
Virginia-class submarine programme, for example, have been split
between two shipyards.
Demanding delivery schedules may also force shipyards to build
various portions of ships at multiple sites simultaneously. In the
shipbuilding industry, demand is highly variable; so, too, are the
odds that a shipyard will win an order. As a result, shipyards are reluctant to maintain workforces sized to meet the highest demand.
The flexibility to obtain blocks from other fabricators in such market

conditions offers shipbuilders a definite advantage.3
The most compelling reason for shipbuilders to build portions of
ships at multiple shipyards is the potential for reduced costs. Theoretically, it is possible to generate cost savings by concentrating
specific blocks in one company so that duplication of skills and facilities can be eliminated. The result is lower overall overhead costs,
as well as the benefits of learning extended over a longer production
run. Some recent decisions on the subcontracting of structural
blocks have been driven by this cost-savings focus. Some very
efficient new facilities focused on building only blocks for other
shipyards to assemble have been introduced in Japan and Europe.4
The basic motivation for this strategy is to achieve economies of
scale and thus be able to better compete in the global commercialshipbuilding market. Concentration could create the scale necessary
to enable an investment in new fabrication/assembly technologies
that would not be possible to justify at a lower scale (lower level of
throughput).5
Another reason for multisite construction is to most effectively utilise
a shipyards existing assets as the global shipbuilding markets fluctuate. In shipbuilding, as in the other capital-intensive heavy
manufacturing industries, efficient capacity utilisation is a key driver
of business effectiveness. To this end, partial outsourcing is an
effective tool in Japans shipbuilding environment. What makes it
feasible is the somewhat distributed nature of the Japanese
shipbuilding industry. Seven major builders all have multiple
facilities, and there are more than a dozen medium-sized or secondtier builders (which build large ships just as do the majors). Many
medium-sized companies have multiple yards as well. In Japan,
labour mobility is low, so subcontractors are used to a much greater
extent than in the United States and United Kingdom, to avoid
______________
3 In Japan, this approach has been named Tactical Level Flexible Outsourcing.
4 In Japan, this is called the Strategic Level Concentration of Investment and Capa-
bility.
5 For example, the new laser steel cutting facility at Bender Shipbuilding (Mobile, Al-
abama) has productive capacity in excess of the shipyards needs. Bender was able to
make this investment, which improves its shipbuilding capability, by setting up the
new facility to do a substantial amount of its work for outside clients.
having to lay people off during periodic slow periods. Some of this
subcontracting involves blocks being built by other companies,
including fabricators that are not shipyards.
The combination of a multiplicity of yards and other fabricators, and
a subcontracting environment, provides a useful degree of operational flexibility at a tactical level that is to handle market fluctuations. When a yard is very busy, it can sub-out more blocks and
thereby push more ships through its final assembly stage. When
business is slower, a yard can retain more of the added value of each
ship contract in-house. This balancing act tends to work because the
multiplicity of shipbuilding companies, shipyards, and steel fabricators confers a degree of flexibility at the total-industry level. It is one
of the benefits of Japans two-tier structure in shipbuilding (Koenig,
Narita, and Baba, 2001).
POTENTIAL DISADVANTAGES OF MULTISITE

CONSTRUCTION
Multisite construction also involves possible disadvantages or additional costs.
First, the problems of accuracy control become more acute because
design and build tolerances must be maintained at several shipyards.
Common nomenclature, techniques, and software packages must be
used to ensure that the blocks built at different shipyards align correctly during assembly. Problems with alignment can lead to potential significant rework costs.
Blocks must be constructed or reinforced in a way to ensure that dimensional tolerances are maintained during transportation. They
also may require additional bracing or structures for the transportation process, which will incur additional costs. Additional costs will
also be incurred by maintaining separate trades and workforce for
transporting blocks between sites.
Finally, since processes must be coordinated among several shipyards, management of the schedule for construction and delivery of
the blocks becomes more difficult. Delays in block construction at
one shipyard, or delays in delivery caused by transportation problems, can seriously throw off the schedule for the delivery of the ship.
CONCERNS FOR THE TYPE 45 PROGRAMME

To summarise what we have learned from the survey of past and current ship construction processes, building the Type 45 ships from
large blocks at multiple shipyards is technically feasible. As
described in the body of this book, overall programme costs with the
block strategy should be lower than having each shipbuilder
construct entire ships and only approximately 4 percent greater than
having sole-source production for the Type 45 class.6 Both BAE
SYSTEMS Marine and VT will maintain their ability to build warships
and, therefore, should provide competition for future MOD
programmes. Finally, the block strategy will facilitate the VT move to
Portsmouth, thereby reducing overhead costs at the Portsmouth
repair facility and providing new facilities for the construction of
ships in the United Kingdom.
However, building, transporting, and assembling blocks of the size
and complexity envisioned for the Type 45 programme have never
been undertaken by either BAE SYSTEMS Marine or VT. Furthermore, the two companies have not previously worked together on
the construction of a ship. Management control and coordination
between the two organisations are extremely important to ensure
that structural tolerances of the blocks are maintained at the two
shipyards, both during production and after transporting the blocks.
Also, most previous examples of multisite construction have been for
commercial shipbuilding programmes. Naval ships, especially warships, are more complex and require more system and communication interfaces. These requirements will magnify the problems with
aligning blocks.
Having both BAE SYSTEMS Marine and VT participate in the design
of the Type 45 is an important determinant of success. This joint design effort must ensure that the necessary engineering and build information is prepared in the correct sequence and in a manner that
is understood by all the shipyards involved in the programme. The
information must also specify the materials and equipment needed
______________
6 Note that our cost estimates for the block options did not include significant rework if
blocks did not align properly. Therefore, our cost estimates for the block options are
somewhat conservative.
by the various shipyards for the construction of their blocks and describe the source and schedule for the material and equipment.
During construction of the blocks at each shipyard, quality-control
functions must examine the blocks during construction and identify
any potential distortion problems when the blocks are completed.
Detailed finite-element analysis is also required to understand the
requirements for and potential effects of proposed lifting and transportation plans. All software tools, procedures, nomenclature, and
methods must be coordinated among all the shipyards to avoid
problems in matching the blocks during final assembly.
All appropriate costs for building and transporting the blocks must
be identified and factored into the total Type 45 programme costs.
For example, the weight of the blocks being transported is much
larger than has typically been moved over water. Special
barges/ships will be required, which could add substantial costs.
The outfitted blocks will have to be made weather-tight, not only for
the sea voyage but also in the shipyards while they are being outfitted
initially and prepared for joining together. An aspect that must be
considered is the change in block size due to temperature. BAE SYSTEMS Marine will be required to measure the joining edges of the
block received at Barrow to eliminate any inaccuracy due to temperature difference between Portsmouth and Barrow. With the
planned block division for the Type 45, such inaccuracies could be
up to an inch in breadth and depth. Finally, the shipyards receiving
and assembling the blocks must have the capability to receive, transport, and load the blocks onto the assembly berth. Creating this capability at shipyards that do not have it could add significant costs to
the programme, thereby negating any total cost benefit.
BIBLIOGRAPHY
Baker, A. D., U.S. Naval Institute Guide to Combat Fleets of the World,
20002001, Annapolis, Md.: U.S. Naval Institute, 2000.
Birkler, J. L., Edmund Dews, and Joseph P. Large, Issues Associated
with Second-Source Procurement Decisions, Santa Monica, Calif.:
RAND, R-3996-RC, 1990.
Birkler, John, et al., Assessing Competitive Strategies for the Joint
Strike Fighter, Santa Monica, Calif.: RAND, MR-1362-OSD/JSF,
2001.
Birkler, John, et al., The U.S. Aircraft Carrier Industrial Base: Force
Structure, Cost, Schedule, and Technology Issues for the CVN 77,
Santa Monica, Calif.: RAND, MR-948-NAVY/OSD, 1998.
Birkler, John, et al., The U.S. Submarine Production Base: An Analysis
of Cost, Schedule, and Risk for Selected Force Structures, Santa
Monica, Calif.: RAND, MR-456-OSD, 1994.
Burton, Anthony, The Rise and Fall of British Shipbuilding, London:
Constable and Company Limited, 1984.
Cash, Jack (Defense Systems Management College), Indirect Cost
Management Guide: Navigating the Sea of Overhead, 2nd ed., Fort
Belvoir, Va.: Defense Systems Management College Press, July
1999, at http://www.dsmc.dsm.mil/pubs/gdbks/icmguide.htm.
The Cooperative Association of Japanese Shipbuilders, Modern Ships
in Japan, Tokyo: n.d. [probably compiled in 1996].
107
Hampton, Richard J., A Price Competition in Weapons Production: A

Framework to Analyze Its Cost-Effectiveness, Boulder, Colo.: Air
University Press, UA-ARI-84-6, June 1984.
Industry Update, Janes Navy International, September 2001, p. 42.
Koenig, Philip C., Hitoshi Narita, and Koichi Baba, Strategies and
Outcomes in the Two Sectors of the Japanese Shipbuilding Industry, Proceedings of the 2001 Ship Production Symposium,
Ypsilanti, Michigan, 1315 June 2001.
Margolis, Milton A., Raymond G. Bonesteele, and James L. Wilson, A
Method for Analyzing Competitive, Dual Source Production Programs, presented at the 19th Annual DoD Cost Analysis Symposium, September 1985.
Ministry of Defence, Performance Report 1999/2000, Abbey Wood,
England, Annex B: Force Structure.
Robertson, Rt. Hon. George, MP, Secretary of State for Defence,
Strategic Defence Review: Modern Forces for the Modern
World, 13 July 1998, at www.army.mod.uk/servingsoldier/
strategydefrev.html.
SSA Industry Overview, 29 August 2001, at www.ssa.org.uk.
Thales, Implications of Armstrong Technology Industry Review,
Paris, 24 August 2001.
Winklareth, Robert J., Naval Shipbuilders of the World: From the Age
of Sail to the Present Day, London: Chatham Publishing, 2000.

Type 45 Destroyer

Uploaded by

Copyright:

Available Formats

Type 45 Destroyer

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Type 45 Destroyer

Uploaded by

Copyright:

Available Formats

TYPE 45

Prepared for the United Kingdoms Ministry of Defence

RAND is a nonprofit institution that helps improve policy and

All rights reserved. No part of this book may be reproduced in any

In April 2001, the United Kingdoms Ministry of Defence (MOD)

The Royal Navys New-Generation Type 45 Destroyer

Royal Navy ships, to commercial work and Foreign Military Sales,

The Royal Navys New-Generation Type 45 Destroyer

Royal Navy: Current Snapshot ....................

Dual-Source, Directed Buy of Whole Ships ..........

B. LONG-TERM IMPLICATIONS OF A LACK OF

C. BUILDING SHIPS IN BLOCKS AT MULTIPLE SITES:

S.1. Summary Comparison of Cost Penalties of Six

The Royal Navys New-Generation Type 45 Destroyer

4.8. Summary of Risks and Rewards of Alternative

S.1. Data Used in RAND Shipyard Model .............

The Royal Navys New-Generation Type 45 Destroyer

HOW CAN MOD MOST EFFECTIVELY MAINTAIN A

This strategy required BAE SYSTEMS Marine and VT to cooperate in

RANDS THREE INTERWOVEN QUESTIONS

Competitive or directed allocation of work if two producers?

The Royal Navys New-Generation Type 45 Destroyer

HOW RAND STUDIED THE PROBLEM

Steel throughput, docks, lifting capacity, outfittinga

Experience levels, age, productivity, costs for hiring and

Current and future

Current and anticipated production plans (by ship)

For each activity listed in the current and future

Wage rates for all the labour types

Overheads, General and Administrative, and profit rates

Fixed investments, such as facilities, necessary for a

aOutfitting includes installation of furnishings (e.g., desks and chairs) and

the shipbuilders, RAND researchers had similar discussions with

WHAT RAND FOUND OUT

The Royal Navys New-Generation Type 45 Destroyer

VT. These cases resulted in an increase in programme costs of 10 to

the blocks produced at the other shipyard.

Cost penalty (%)

Figure S.1Summary Comparison of Cost Penalties

We examined two different options involving the distribution of the

RANDS FINDINGS ALREADY HAVE BEEN PUT TO USE

The Royal Navys New-Generation Type 45 Destroyer

shipbuilders will remain viable and able to compete on future MOD

The Royal Navys New-Generation Type 45 Destroyer

gave generously of their time and advanced our understanding of the

Alternative Landing Ship Logistics

Fleet Replenishment Ship

Chief of Defence Procurement

Common New Generation Frigate

Competing Prime Contractor

Future Aircraft Carrier

Demonstration and Manufacture

Devonshire Dock Hall

Demonstration and First of Class Manufacture

Defence Procurement Agency (formerly the

Future Offshore Patrol Vessel

Floating Production, Storage and Offloading vessel

The Royal Navys New-Generation Type 45 Destroyer

Future Surface Combatants

Full-Scale Engineering Development and Initial