Navy Ships: Lessons of Prior Programs May Reduce New Attack Submarine Cost Increases and Delays (Letter Report, 10/20/94, GAO/NSIAD-95-4)
Lessons learned from prior submarine programs, particularly the Seawolf,
could help the Navy save as much as $100 million in acquisition costs
while improving design quality and construction for the SSN, the latest
class of nuclear-power attack submarine. Specifically, the Navy may want
to (1) contract with a single shipyard to both design and build the lead
submarine, (2) delay lead ship construction until the ship's design is
substantially mature, (3) strengthen the specification development and
approval process, (4) identify critical components and supply vendors
early in the program, and (5) reduce submarine combat system development
risks. Because of the importance of applying both management and
technical lessons, GAO believes that the formal Defense Department
approved acquisition strategy should spell out specifically how the Navy
will avoid repeating the problems of earlier programs.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: NSIAD-95-4
TITLE: Navy Ships: Lessons of Prior Programs May Reduce New Attack
Submarine Cost Increases and Delays
DATE: 10/20/94
SUBJECT: Navy procurement
Nuclear powered submarines
Construction (process)
Construction contracts
Cost analysis
Contract administration
Shipbuilding industry
Cost overruns
Defense cost control
Construction costs
IDENTIFIER: Seawolf Attack Submarine
SSN-21 Submarine
NSSN Attack Submarine
SSN-22 Submarine
**************************************************************************
* This file contains an ASCII representation of the text of a GAO *
* report. Delineations within the text indicating chapter titles, *
* headings, and bullets are preserved. Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are *
* identified by double and single lines. The numbers on the right end *
* of these lines indicate the position of each of the subsections in the *
* document outline. These numbers do NOT correspond with the page *
* numbers of the printed product. *
* *
* No attempt has been made to display graphic images, although figure *
* captions are reproduced. Tables are included, but may not resemble *
* those in the printed version. *
* *
* A printed copy of this report may be obtained from the GAO Document *
* Distribution Facility by calling (202) 512-6000, by faxing your *
* request to (301) 258-4066, or by writing to P.O. Box 6015, *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time. *
**************************************************************************
Cover
================================================================ COVER
Report to Congressional Requesters
October 1994
NAVY SHIPS - LESSONS OF PRIOR
PROGRAMS MAY REDUCE NEW ATTACK
SUBMARINE COST INCREASES AND
DELAYS
GAO/NSIAD-95-4
Navy Ships
Abbreviations
=============================================================== ABBREV
DOD - Department of Defense
Letter
=============================================================== LETTER
B-257866
October 20, 1994
The Honorable Edward M. Kennedy
Chairman, Subcommittee on Regional
Defense and Contingency Forces
Committee on Armed Services
United States Senate
The Honorable John Conyers, Jr.
Chairman, Legislation and National
Security Subcommittee
Committee on Government Operations
House of Representatives
As you requested, we assessed the Navy's plans to incorporate lessons
learned from prior submarine programs, particularly the Seawolf
SSN-21 program into the design and construction of the NSSN, a new
class of nuclear-powered attack submarine. According to the Navy,
the NSSN is to be smaller, less costly and less capable than the
Navy's Seawolf submarine; it will also be expected to perform a
variety of missions.
Several factors make the NSSN both an excellent opportunity and a
challenge for the Navy to control acquisition costs and to improve
the quality of the design and construction process. These factors
are (1) a reduced antisubmarine warfare threat, which has resulted
from the breakup of the former Soviet Union; (2) the U.S. defense
budget, which has been more tightly constrained each year; and (3)
the early stages of the NSSN acquisition cycle, which allow an agency
to apply lessons of past programs to future programs.
BACKGROUND
------------------------------------------------------------ Letter :1
Believing that high Seawolf submarine program costs would lead to
inadequate force levels, the Department of Defense (DOD), in October
1991, established a requirement for a more affordable new attack
submarine. According to the Navy, the NSSN's estimated displacement
weight will be about 7,100 tons, 2,000 tons less than the Seawolf's.
The NSSN's missions include battlegroup support, covert strike
warfare, covert intelligence, special warfare, covert mine warfare,
antisubmarine warfare, and antisurface warfare operating in both open
ocean and littoral (coastal) areas.
In August 1992, the Defense Acquisition Board authorized the Navy to
initiate concept exploration and definition (milestone 0) studies. A
project office was established to set out the basic design and to
develop an acquisition strategy that included the schedule of detail
design and production. The Navy initially planned for the Board to
approve the NSSN acquisition strategy in August 1993, as part of the
milestone I decision to enter the demonstration and validation phase.
However, the milestone I meeting slipped until January 1994. That
meeting resulted in requesting the Navy to perform additional studies
and analyses. These were completed and submitted to the Board.
On August 1, 1994, the Board approved milestone I, and on August 18,
1994, issued an Acquisition Decision Memorandum. The memorandum
directed the Navy to submit an updated documentation package for the
Board's approval within 60 days. The package is to include an
acquisition strategy report, reflecting the Navy's plan to initiate
detail design and lead ship construction at Electric Boat. The Board
also directed the Navy to initiate (1) advanced procurement of the
lead ship's nuclear reactor in fiscal year 1996 and (2) lead ship
construction in fiscal year 1998. Further, the Board directed the
Navy to update the submarine's combat system acquisition strategy to
reflect "a significant degree of private sector involvement in
planning an open system architecture," which contains commercially
available hardware and software that meet broad industry standards.
A September 1993 cost and operational effectiveness analysis prepared
by the Center for Naval Analysis estimated the cost for comparison
purposes of procuring 30 NSSNs and procuring 30 Seawolf submarines at
1 ship per year. In constant fiscal year 1994 dollars, the
procurement cost for the NSSN was about $45 billion ($1.5 billion
each) and for the Seawolf about $56 billion ($1.9 billion each).
RESULTS IN BRIEF
------------------------------------------------------------ Letter :2
The Navy may be able to avoid some design and construction costs and
schedule delays by applying the following five management lessons
from prior submarine construction programs, which we have distilled
from the reports of our reviews over the past decade: (1)
contracting with a single shipyard to both design and construct the
lead submarine, (2) delaying lead ship construction until the ship's
design is substantially mature, (3) strengthening the specification
development and approval process, (4) identifying critical components
and supply vendors early in the program, and (5) reducing submarine
combat system development risks. While the Navy's project manager
has said that he intends to incorporate these management lessons into
the program's acquisition strategy, the extent to which they will be
applied cannot be assessed until DOD approves the strategy and makes
it public for evaluation.
The Navy has evaluated a number of technical lessons learned from
past submarine construction programs and has approved plans to
incorporate some of them in the NSSN program. The Navy estimates
that NSSN procurement cost savings from these lessons could range
from $90 million to $100 million.
Because of the importance of applying both management and technical
lessons, we believe the formal DOD approved acquisition strategy
should explicitly address how the Navy will avoid repeating the
problems of prior programs.
APPLYING MANAGEMENT LESSONS MAY
REDUCE COSTS AND AVOID SCHEDULE
DELAYS
------------------------------------------------------------ Letter :3
By incorporating management lessons into the NSSN program, the Navy
may avoid repeating many of the problems that caused Seawolf detail
design and lead ship construction cost increases and schedule delays.
In recognition of Seawolf problems, the NSSN project manager told us
he intends, subject to DOD approval, to incorporate the five
management lessons into the multibillion dollar NSSN program.
However, because of the absence of a DOD approved acquisition
strategy, the extent to which the NSSN acquisition strategy will
include these lessons cannot be assessed now.
USE A SINGLE SHIPYARD TO
DESIGN AND CONSTRUCT THE
LEAD SUBMARINE
---------------------------------------------------------- Letter :3.1
Under the split design/construction strategy used for the Seawolf
program, Tenneco's Newport News Shipbuilding and Drydock Company was
responsible for the overall design and detail design of the
submarine's forward end, while General Dynamics' Electric Boat
Division was responsible for designing the submarine's aft end and
for constructing the SSN-21 and the SSN-22. The split design
approach, with a requirement that design data be suitable for use at
either shipyard, was originally instituted to instill competition for
building 29 SSN-21 class submarines. This approach, which required
additional time and resources as well as a high degree of
coordination between the two shipbuilders, caused design and
construction cost increases and additional time to approve design
data and to resolve design drawing problems.
Electric Boat, to construct the SSN-21, still had to convert Newport
News Shipyard's generic design data into Electric Boat specific work
packages (instructions and materials). According to Seawolf program
office officials, the two shipyards were unwilling to open their
operations to one another. In addition, the Navy's Seawolf program
office occasionally had to mediate the resolution of design drawing
problems between the two shipyards. Confusion between the two
shipyards over design drawing delivery schedules was one factor that
led to late delivery of design drawings to Electric Boat, the
shipbuilder, in 1990 and in the first 6 months of 1991.
A Seawolf program office official noted that the Seawolf program
office has learned that having one shipbuilder design and construct
the submarine can save time and money. The August NSSN Acquisition
Decision Memorandum shows that one shipyard will design and build the
lead NSSN.
DELAY LEAD SHIP CONSTRUCTION
UNTIL DESIGN MATURES
---------------------------------------------------------- Letter :3.2
The high degree of concurrent development and lead ship construction
caused cost increases on Seawolf. The Navy awarded Newport News
Shipbuilding the overall Seawolf detail design contract in April
1987. Construction of the first Seawolf, the SSN-21, started in
October 1989, with delivery originally scheduled for May 1995.
In some cases, this concurrency required developing and issuing
drawings before system designs were fully mature. Although this
approach provided the shipbuilder with design data earlier, it also
caused a higher degree of design rework and, in some cases,
construction rework. For example, the Navy's data requirement lists
developed during the early phase of Seawolf design were based, as was
the case with prior submarine efforts, on providing the shipbuilder
with engineering drawings as the basis for performing construction
tasks. It was later discovered that because Seawolf submarines
required a significantly greater level of modular construction and
outfitting, new and more detailed sectional construction drawings
were needed to initiate modular construction tasks. As a result, in
June 1990, 8 months after SSN-21 construction started and about 37
months after detail design started, the Navy rebaselined and
increased Newport News' original $303 million detail design contract
by $168 million. The rebaselining was for Newport News to prepare
and to provide Electric Boat with more detail design data and
incorporate final submarine specifications into the detail design.
A September 1993 NSSN cost and operational effectiveness analysis
found that an additional investment of between $105 million and $175
million in research and development funds to review the NSSN's
specifications and to complete design before lead ship construction
contract award could reduce procurement costs by $141 million to $173
million per ship.
Starting lead NSSN ship construction with a more mature detail design
could result in a more cost-effective and efficient approach than
that used under the Seawolf submarine program. In June 1994, the
NSSN project manager stated that the Navy plans to begin lead NSSN
construction when the detail design matures. However, lead ship
construction will still begin in fiscal year 1998, despite the 1-year
slip of milestone I. Under the current NSSN schedule, detail design
is scheduled to begin in July 1995, with lead ship construction
beginning about 27 months later in October 1997. However, we
question whether the detail design will be mature enough to avoid
repeating similar problems the Navy experienced with the Seawolf
program. The Seawolf program experienced design and construction
rework, significant cost increases, and schedule delays, despite a
30-month interval between starting detail design and lead ship
(SSN-21) construction.
STRENGTHEN SPECIFICATION
DEVELOPMENT AND APPROVAL
PROCESS
---------------------------------------------------------- Letter :3.3
Deficient government specifications for welding HY-100 strength
steel\1 have increased SSN-21 construction costs and have delayed the
submarine's delivery from May 1995 until May 1996. In June 1991,
Electric Boat experienced problems welding this new steel. As a
result, Electric Boat notified the Navy that it had discovered weld
cracks where two hull rings were joined together. Further
investigation revealed additional unacceptable welds on the SSN-21's
pressure hull and on at least 21 government-and contractor-furnished
items. By August 1991, all HY-100 welding had been stopped.
The chemical composition of the welding metal, among other things,
had resulted in cracking and unacceptable metal yield strengths and
ductility. Ultimately, however, the welding cracks were traced to
deficient government HY-100 welding specifications. Electric Boat
and the Navy took corrective action; all welding resumed by December
1991. As a result of this problem, the Navy paid Electric Boat $77.8
million (in then-year dollars) to fix the cracks. It also caused a
1-year delay in the SSN-21's delivery.
During the determination of defective government specifications, the
Commander, Naval Sea Systems Command, requested an independent
assessment of the system for developing, preparing, and approving
specifications. The assessment, completed in March 1992, showed that
weaknesses in developing and qualifying specifications were caused by
a lack of management priority and oversight, inadequate and untimely
availability of funds, and a shortage of personnel needed to develop
and update specifications. In addition, the assessment showed that
only 39 percent of the specification parameters were supported by
historical performance data and less than 5 percent of the parameters
were supported by test data.
The NSSN project manager said he plans to incorporate a review
process that supports developing specifications. In addition, he
indicated that the Navy plans to work with critical NSSN vendors
early during design to coordinate specifications, including
revisions, whenever necessary. Moreover, according to the NSSN
project manager, the NSSN, to the extent possible, will incorporate
existing systems and components from prior submarine programs and
off-the-shelf, commercially available technology. Nevertheless, some
existing systems may require varying degrees of reengineering for
installation into the NSSN.
--------------------
\1 HY-100, a high-yield steel used to construct the SSN-21's pressure
hull, allows the submarine to achieve deeper diving depths. Prior to
the Seawolf program, a U.S. submarine's pressure hull was
constructed using primarily HY-80 strength steel.
EARLIER IDENTIFICATION OF
CRITICAL COMPONENTS AND
SUPPLY VENDORS
---------------------------------------------------------- Letter :3.4
DOD has identified the decline of the submarine industrial base and
the resulting uncertainty surrounding submarine component vendors as
key factors contributing to Seawolf cost and schedule delays.
Early identification of critical components and supply vendors can
help determine whether to buy or manufacture some components in-house
and can help reduce potential procurement problems. For the SSN-21,
Electric Boat had to manufacture certain systems and components that
it was originally planning to buy. This was due either to a lack of
qualified vendors or the cost and schedule risks inherent in using a
vendor for complex components that were under development (i.e., the
weapons storage and handling system). Collectively, the absence of
sufficient vendors contributed to Seawolf design and construction
cost increases and schedule delays.
The Navy's March 1992 assessment showed an apparent incomplete
coordination with industry and inadequate notification to and
consultation with industry regarding major changes in Seawolf
specifications as required by the Naval Sea System Command's
specification process. The assessment also showed that vendors were
generally dissatisfied with government feedback to their comments
during specification development and modification.
According to the NSSN project manager, Electric Boat will identify
and obtain critical suppliers earlier than was done on the Seawolf
program. To improve coordination with vendors and to identify issues
that can affect the NSSN's design and construction, Electric Boat has
assembled a team of 100 designers, construction trade people, and key
vendors. However, the commitment to and the success of this effort
will not be assessable until a later phase of design.
REDUCE COMBAT SYSTEM
DEVELOPMENT RISKS
---------------------------------------------------------- Letter :3.5
The Navy experienced problems developing the AN/BSY-1 combat system\2
for the Improved SSN-688 class submarine and the AN/BSY-2 combat
system for the Seawolf submarine. Because the time to correct
AN/BSY-1 combat system design and development problems was
insufficient, the AN/BSY-1 became the major factor in delays to the
Improved SSN-688 construction program. These problems resulted in an
additional $82 million in contract costs for five Improved SSN-688s.
In addition, the first nine Improved SSN-688s equipped with AN/BSY-1
systems were delivered to the Navy an average of 17 months late.
The AN/BSY-2 combat system scheduled for installation on the SSN-21
experienced cost increases and schedule delays. Changes to the
system's design caused a portion of the submarine to be redesigned at
an additional cost.\3 The Navy originally provided Newport News with
general space and weight information for the system that the shipyard
used to begin designing its portion of the Seawolf. The Navy later
provided the shipyard with more specific information that caused
considerable redesign of the submarine and increased design costs,
according to Newport News.
The Navy estimated in August 1994 that system development would cost
$123 million more than the original contract target cost of $1
billion. Our November 1992 report\4 showed that delivery of the
system's first phase capabilities (all hardware and the majority of
software) had been delayed from its original November 1993 delivery
to between late March and June 1994. Because the HY-100 welding
crack problem delayed the submarine's delivery 1 year, until May
1996, the Navy revised the AN/BSY-2 system's first phase delivery to
February 1995. According to a February 1994 Defense Acquisition
Executive Summary prepared by the Seawolf program office, maintaining
the AN/BSY-2 software development schedule to support lead ship
delivery remains a challenge. The AN/BSY-2 hardware is complete and
ready for delivery to Electric Boat.
To reduce combat system cost, schedule, and technical risks the Navy
encountered developing the systems for the Improved SSN-688
(AN/BSY-1) and Seawolf (AN/BSY-2) class submarines, the NSSN project
manager stated that whenever possible, the NSSN's combat system will
be developed using what he termed an open systems architecture, which
consists of commercial, off-the-shelf hardware and software. The
Acquisition Decision Memorandum specifies a combat system acquisition
strategy that involves "a significant degree of private sector
involvement in planning an open system architecture." Nevertheless,
some existing systems may require varying degrees of reengineering
for installation into the NSSN.
--------------------
\2 A submarine combat system detects, classifies, localizes, tracks,
and destroys enemy targets.
\3 Status of SSN-21 Ship Construction Program (GAO/NSIAD-90-163, Apr.
19, 1990).
\4 Status of SSN-21 Design and Lead Ship Construction Program
(GAO/NSIAD-93-34, Nov. 17, 1992).
APPLYING TECHNICAL LESSONS MAY
SAVE MILLIONS OF DOLLARS
------------------------------------------------------------ Letter :4
The NSSN project office compiled a database that identified about
1,350 primarily technical lessons from prior Navy programs. Electric
Boat, Newport News, and other Navy organizations provided the input
for this database. Personnel transferring into the NSSN project
office from earlier submarine programs also provided some lessons.
After consolidating duplicate lessons, the Navy reduced the database
to 954 lessons. The NSSN project office's evaluation process is
ongoing, and new lessons are added to the database periodically.
Of these 954 lessons, 290 had been approved for incorporation into
the NSSN design as of May 1994. Examples of approved lessons are (1)
centralizing the ship's service hydraulic power plant, (2)
simplifying the ship's deck design, and (3) simplifying the ship's
pipe hangers. These three lessons are expected to save over $10
million in acquisition costs. The Navy estimates that NSSN savings
from all approved lessons could range from $90 million to $100
million. However, because individual lessons' costs can offset each
other, savings must be assessed on a lesson-by-lesson basis. The
potential exists for additional savings because the project office
has not completed its review of almost 600 lessons. (See table 1 for
status of the lessons.)
Table 1
Navy's Disposition of Technical Lessons
Learned as of April 1994
Status Number of lessons
---------------------------------------- ------------------
Approved for incorporation 290
into NSSN design
Deferred until a later stage of 84
development
Open until additional 504
reviews are completed
Rejected for NSSN purposes 76
============================================================
Total 954
------------------------------------------------------------
The NSSN's project manager noted that the Navy plans to incorporate
the 290 technical lessons into the submarine's preliminary design
during the submarine's demonstration and validation phase.
RECOMMENDATION
------------------------------------------------------------ Letter :5
To allow an assessment of how the Navy will avoid a repetition of
past problems, we recommend that the Secretary of Defense ensure that
the formal NSSN acquisition strategy explicitly documents how the
Navy is to address and incorporate the management and technical
lessons from prior submarine programs.
AGENCY COMMENTS AND OUR
EVALUATION
------------------------------------------------------------ Letter :6
In commenting on a draft of this report, DOD generally concurred with
the report and indicated that the Navy intended to apply the lessons
learned from the prior programs. However, DOD did not believe it was
necessary to explicitly document in a formal acquisition strategy how
the Navy is to address and incorporate those lessons. DOD stated it
was confident the current process provides adequate emphasis on
lessons learned from prior programs.
After considering DOD's position, we continue to believe that
implementation of our recommendation is warranted. This is a
multibillion dollar program; the lessons that should have been
learned have already been identified; therefore, it seems that
documenting how they are to be incorporated is merely a completion of
the cycle--a way of better assuring that the Navy avoids a repetition
of cost and scheduling difficulties. Further, we believe such
documentation will serve as a valuable tool for guiding the
implementation of the program.
DOD comments are presented in their entirety in appendix I. DOD's
suggestions for improving the clarity of the report have been
incorporated in the text where appropriate.
SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :7
To determine the types of experiences the Navy should apply to the
NSSN effort, we reviewed our prior products on the SSN-21, SSN-688,
Trident, the combat systems, and other organizations' reports on
lessons learned. We held discussions with Navy program officials for
the Seawolf program, the AN/BSY-1 and the AN/BSY-2 combat system
programs, and the SSN-688 and the Trident submarine programs. We
held discussions with the Supervisors of Shipbuilding in Groton,
Connecticut, and with Naval Undersea Warfare Center officials in
Newport, Rhode Island. We also held discussions with Navy officials
responsible for planning the NSSN's development in Washington, D.C.
We reviewed the NSSN project office's database of technical lessons
learned experiences or suggestions and reviewed and analyzed Navy
studies and assessments. We obtained, reviewed, and assessed
suggestions provided by Electric Boat, Groton, Connecticut; and
Newport News Shipbuilding, Newport News, Virginia. Electric Boat
provided more detailed information on views on selected lessons
learned that should be applied to the NSSN program, but Newport News
Shipbuilding did not because of other business pressures.
We conducted our review from June 1993 to June 1994 in accordance
with generally accepted government auditing standards.
---------------------------------------------------------- Letter :7.1
We are sending copies of this report to the Secretaries of Defense
and the Navy and to congressional oversight committees. We will also
make copies available to others upon request.
Please contact me at (202) 512-4841 if you or your staff have any
questions concerning this report. The major contributors to this
report are listed in appendix II.
Brad Hathaway
Associate Director, Systems Development
and Production Issues
(See figure in printed edition.)Appendix I
COMMENTS FROM THE DEPARTMENT OF
DEFENSE
============================================================== Letter
(See figure in printed edition.)
(See figure in printed edition.)
MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II
NATIONAL SECURITY AND
INTERNATIONAL AFFAIRS DIVISION,
WASHINGTON, D.C.
John D'Esopo
David Fisher
BOSTON REGIONAL OFFICE
Jeffrey Rose
Ralph Tavares
Alson Castonguay
LIST OF SSN-21 GAO-RELATED
PRODUCTS SINCE 1991
============================================================ Chapter 0
GAO has performed work on the Seawolf program since 1985. The
following chronology presents products issued since 1991.
Navy Ships: Seawolf Cost Increases and Schedule Delays Continue
(GAO/NSIAD-94-201BR, June 30, 1994).
Navy Ships: Problems Continue to Plague the Seawolf Submarine
Program (GAO/NSIAD-93-171, Aug. 4, 1993).
Navy Ships: Plans and Anticipated Liabilities to Terminate SSN-21
Program Contracts (GAO/NSIAD-93-32BR, Nov. 27, 1992).
Navy Ships: Status of SSN-21 Design and Lead Ship Construction
Program (GAO/NSIAD-93-34, Nov. 17, 1992).
SSN-21, Seawolf Contract Terminations (GAO/NSIAD-93-41R, Nov. 6,
1992).
Navy Shipbuilding: Effects of Reduced SSN-21 Procurement Rates on
Industrial Base and Cost of Program (GAO/NSIAD-92-140, Apr. 8,
1992).
Submarine Combat System: BSY-2 Development Risks Must Be Addressed
and Production Schedule Reassessed (GAO/IMTEC-91-30, Aug. 22,
1991).
Submarine Combat System: Status of Selected Technical Risks in the
BSY-2 Development (GAO/IMTEC-91-46BR, May 24, 1991).
A May 24, 1991, letter to Representative Herbert H. Bateman
discussing cost projections of several SSN-21 procurement
scenarios.
|
NEWSLETTER
|
|
Join the GlobalSecurity.org mailing list
|
|
|
