New Attack Submarine: Program Status (Letter Report, 12/03/96,
GAO/NSIAD-97-25).

GAO reviewed the status of the Navy's New Attack Submarine (NSSN)
Program to identify areas of potential risk for rising costs and to
provide information on the submarines' potential performance.

GAO found that: (1) the NSSN program is not likely to meet the objective
of producing a submarine that is significantly less costly than the
Seawolf; (2) based on Navy estimates for a 30-ship, single shipbuilder
program, the Seawolf's average acquisition cost was estimated to be
about $1.85 billion compared to the NSSN's estimate of about $1.5
billion; (3) based on a 30-ship, two shipbuilder program, the Navy's
current estimated acquisition cost for the fifth ship of the NSSN class
has already risen from about $1.5 billion to about $1.8 billion; (4)
Public Law 104-106 directed the Navy to accelerate construction of the
first two submarines and to use two shipyards instead of one to build
the first four submarines; (5) according to the Navy, this change has
increased the estimated cost of developing and building 30 NSSNs by $3
billion; (6) although GAO believes the cost categories seem reasonable,
GAO has no basis to agree or disagree with the total program estimate
because the Navy has not provided support for the costs associated with
the individual categories; (7) according to the Department of Defense, a
$3.8-billion increase in budget authority for the Fiscal Year 1997-2001
Future Years Defense Plan period will be needed to acquire two
submarines earlier than originally planned as directed by the law; (8)
GAO believes there is a potential for other cost increases because of a
variety of program risks: (a) anticipated changes in the ship's design
or the addition of new technologies are likely; (b) the NSSN command,
control, communication, and intelligence combat system development and
integration program is highly complex and optimistic; (c) development of
some prototype equipment is concurrent with ship construction; and (d)
the cost of transferring the submarine's design from the first to the
second shipbuilder is based on an optimistic estimate that is less than
the actual cost of the last major design transfer; and (9) there is a
divergence of views on the start-up costs for the second shipbuilder.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  NSIAD-97-25
     TITLE:  New Attack Submarine: Program Status
      DATE:  12/03/96
   SUBJECT:  Submarines
             Navy procurement
             Future budget projections
             Defense cost control
             Defense capabilities
             Shipbuilding industry
             Command/control/communications systems
             Concurrency
             Advanced weapons systems
IDENTIFIER:  Los Angeles Class Attack Submarine
             NSSN Attack Submarine
             Navy New Attack Submarine Program
             Seawolf Attack Submarine
             DOD Bottom-Up Review
             AN/BSY-1 Submarine Combat System
             AN/BSY-2 Submarine Combat System
             DOD Future Years Defense Program
             DDG-51 Destroyer
             
******************************************************************
** 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.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Committees

December 1996

NEW ATTACK SUBMARINE - PROGRAM
STATUS

GAO/NSIAD-97-25

New Attack Submarine

(707112)


Abbreviations
=============================================================== ABBREV

  ATM - Asynchronous Transfer Mode
  C3I - command, control, communication, and intelligence
  COTS - commercial off-the-shelf
  DOD - Department of Defense
  INPP - improved nonpenetrating periscope
  NSSN - new attack submarine
  OPTEVFOR - Operational Test and Evaluation Force
  SCN - shipbuilding and conversion

Letter
=============================================================== LETTER


B-272638

December 3, 1996

Congressional Committees

The Congress, the Department of Defense (DOD), and the Navy agree
that the Navy needs a new attack submarine (NSSN) that is
significantly less costly than the previous class of attack
submarines, the Seawolf.  However, the estimated cost of the NSSN
program is rising.  We evaluated the status of the NSSN program to
identify areas of potential risk for rising costs and to provide
information on the submarines' potential performance.  We conducted
this review under our basic legislative responsibilities and are
addressing it to you because the matter discussed in this report
falls within your Committee's jurisdiction. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

The NSSN program is not likely to meet the objective of producing a
submarine that is significantly less costly than the Seawolf.  Based
on Navy estimates for a 30-ship, single shipbuilder program, the
Seawolf's average acquisition cost was estimated to be about $1.85
billion compared to the NSSN's estimate of about $1.5 billion.\1
Based on a 30-ship, two shipbuilder program, the Navy's current
estimated acquisition cost for the fifth ship of the NSSN class has
already risen from about $1.5 billion to about $1.8 billion as of
March 1996.  In addition, there is a potential for other cost
increases because of a variety of program risks. 

The National Defense Authorization Act for fiscal year 1996, Public
Law 104-106, directed the Navy to accelerate construction of the
first two submarines and to use two shipyards instead of one to build
the first four submarines.  According to the Navy, this change has
increased the estimated cost of developing and building 30 NSSNs by
$3 billion.  Although we believe the cost categories seem reasonable,
we have no basis to agree or disagree with the total program estimate
because the Navy has not provided support for the costs associated
with the individual categories.  Historical evidence shows that some
cost increases may occur.  The impact of competition is generally
expected to result in decreases in production costs.  Such results
will depend on the Navy's acquisition strategy, which has not yet
been determined. 

According to DOD, a $3.8-billion increase in budget authority for the
Fiscal Year 1997-2001 Future Years Defense Plan period will be needed
to acquire two submarines earlier than originally planned as directed
by the law. 

We believe there is a potential for other cost increases because of a
variety of program risks. 

  -- Anticipated changes in the ship's design or the addition of new
     technologies are likely. 

  -- The NSSN command, control, communication, and intelligence (C3I)
     combat system development and integration program is highly
     complex and optimistic. 

  -- Development of some prototype equipment is concurrent with ship
     construction. 

  -- The cost of transferring the submarine's design from the first
     to the second shipbuilder is based on an optimistic estimate
     that is less than the actual cost of the last major design
     transfer. 

There is also a divergence of views on the start-up costs for the
second shipbuilder. 

DOD and the Navy believe the baseline NSSN satisfies military
requirements.  However, in an April 1995 report, the Commander,
Operational Test and Evaluation Force (OPTEVFOR), stated if the NSSN
were to just meet design thresholds for survivability, the NSSN might
not be operationally effective against the most capable threat.  He
noted that there are too many variables and unknowns about the
systems involved to assess whether the design will meet the
requirements.  The Navy is addressing the concerns raised by this
report.  OPTEVFOR is currently reassessing the NSSN design for a
program review scheduled in early 1997. 


--------------------
\1 Unless otherwise noted, all costs are in constant 1995 dollars. 


   BACKGROUND
------------------------------------------------------------ Letter :2

According to Navy documents, producing a significantly less costly
submarine would allow the Navy to maintain a force level of 45 to 55
submarines and maintain the current submarine industrial base.  The
NSSN program is intended to address the Navy's requirement for 10 to
12 new attack submarines by the year 2012 that are as quiet as the
Seawolf, but at lower cost and without compromising military utility. 
Before design evaluation, the Chief of Naval Operations set the
attributes and requirements for the NSSN based on the Navy's
prospective need to defeat a very sophisticated future Russian threat
and to operate in littoral (coastal) areas.  The submarine's
attributes and requirements are the NSSN's major cost drivers; these
include speed, quietness, diving depth, weapons load, and sensor
performance.  According to the Navy, the NSSN will be a highly
effective multimission platform capable of performing antisubmarine
and antisurface ship missions and land attack strikes as well as mine
missions, special operations, battle group support, and surveillance. 

During submarine design and development, the Navy initiated several
acquisition reform measures intended to keep costs under control. 
According to the Navy, the cost of the submarine will be reduced by
about one third, because of the use of Integrated Product and Process
Development teams, computer-aided design, commercial off-the-shelf
technology, and a performance-based acquisition strategy.  The Navy
also claims it can save as much as $100 million per ship by
incorporating lessons learned from previous submarine programs.  The
savings expected from these initiatives are included in the Navy's
estimated acquisition cost for the fifth ship NSSN. 


   NSSN COSTS HAVE RISEN
------------------------------------------------------------ Letter :3

Based on Navy estimates for a 30-ship, single shipyard program, the
Seawolf's average acquisition cost was estimated to be about $1.85
billion compared to the NSSN's estimate of about $1.5 billion.  Based
on a 30-ship, two shipbuilder program, the Navy's estimated
acquisition cost for the fifth\2 ship of the NSSN class has already
risen from about $1.5 billion to about $1.8 billion as of March 1996. 

According to the Navy, most of the $250 million increase in the
estimated cost of the fifth NSSN is the result of the direction in
Public Law 104-106 to add a second shipbuilder to the program and to
accelerate the procurement of two of the first four submarines.  The
Navy's original plan, approved in May 1995, was to build one ship in
fiscal year 1998, a second ship in fiscal year 2000, and two ships
per year beginning in fiscal year 2002--all at Electric Boat
Corporation, Groton, Connecticut.  The 1-year gap between
construction of the first and second ships, according to Navy
officials, would allow the Navy to "mature" the NSSN design and gain
experience constructing the first ship before beginning construction
of the second ship. 

The Navy currently maintains two nuclear-capable shipyards--one to
build submarines (Electric Boat, Groton, Connecticut) and one to
build nuclear-powered aircraft carriers (Newport News Shipbuilding
and Drydock Company, Newport News, Virginia).  According to Navy
officials, the use of the single contractor to build the NSSN would
increase savings by avoiding overhead and design transfer costs and
achieving the benefits of the experience gained by building 30 ships
at one shipyard.  The Navy adopted this strategy in response to a
recommendation in DOD's Bottom-Up Review to use a single contractor. 

However, Public Law 104-106 directed the Navy to start construction
of an NSSN at Newport News Shipbuilding and Drydock Company (Newport
News) in 1999 and submit a plan for building four NSSNs between
fiscal year 1998 and 2001, two of which were to be built by Electric
Boat (one in 1998 and one in 2000) and two by Newport News (one in
1999 and one in 2001).  The intent of the legislation in introducing
competition was to obtain a cost benefit.  The act requires that each
of the first four submarines develop and demonstrate new technologies
that will make each more capable and more affordable than its
predecessor. 

In its required March 1996 report, DOD indicated it would need an
additional $3.8 billion for fiscal years 1997 through 2001 in
shipbuilding and conversion (SCN) funds to implement the act.  Even
with this additional funding, new submarine construction would
require about 23 to 38 percent of the available SCN funds versus the
historical 22 to 25 percent.  The additional funding is to build the
baseline design only and does not include funding for new technology
insertion or nonrecurring costs related to design changes.  DOD said
that it would also need an additional $787 million in research,
development, test, and evaluation funds to accelerate and develop new
technology not included in the current design.  According to DOD and
Navy officials, the NSSN, as currently designed, is expected to meet
all requirements for carrying out its missions. 

DOD noted in its report that it would be difficult to afford the plan
directed by Public Law 104-106 in the context of other modernization
programs.  The Department developed and proposed several alternative
construction schedules that would add a second shipbuilder and lead
to competition.  The report states, however, that DOD would face
major near-term affordability issues if it pursued the congressional
plan or any of the alternatives presented in its report. 

DOD did not report the individual cost for each of the first four
ships, which we provide in table 1. 



                                Table 1
                
                 NSSN Shipbuilding and Conversion Costs

                    (Then-year dollars in billions)

Submarine                                          Cost
----------------------------------  ----------------------------------
1                                                 $3.272
2                                                 2.543
3                                                 2.093
4                                                 2.112
======================================================================
Total                                            $10.020
----------------------------------------------------------------------
In addition to construction costs, the Navy estimates it will cost
about $3.8 billion (then-year dollars) to design and develop the
NSSN.  As a result, the Navy will spend a total of about $13.8
billion to develop and buy the first four NSSNs under the revised
shipbuilding plan. 

Under the strategy to use the two shipbuilders, the Navy estimates
that the cost of the fifth ship of the NSSN class in fiscal year 1995
dollars will rise from about $1.5 billion to about $1.8 billion as of
March 1996.  According to the Navy, this change in strategy will
result in increased costs such as increased overhead and the loss of
experience in building both the lead and follow-on ships when using
two shipbuilders and decreased costs such as lower production costs
due to competition.  The Navy estimates that the net effect for the
total program is a cost increase of about $3 billion.  Although we
believe the cost categories seem reasonable, we have no basis to
agree or disagree with the total program estimate because the Navy
has not provided support for the costs associated with the individual
categories.  Historical evidence shows that some cost increases may
occur.  The impact of competition is generally expected to result in
decreases in production costs.  Such results will depend on the
Navy's acquisition strategy, which has not yet been determined. 


--------------------
\2 According to Navy documents, the fifth ship cost is used because
it provides a convenient reference point for cost comparisons because
the influences of learning and cost improvement have already
occurred.  Other programs, such as the Los Angeles attack submarine
and the Arleigh Burke destroyer also used the fifth ship cost for
estimating unit costs.  The difference between the previous average
acquisition costs of the NSSN and the new fifth ship NSSN cost is
estimated to be about $50 million. 


   POTENTIAL FOR COST INCREASES
   FOR VARIETY OF PROGRAM REASONS
------------------------------------------------------------ Letter :4

Even without adding a second shipbuilder to the program, there is a
potential for cost increases, because of a variety of program risks. 
A highly complex and optimistic C3I system development and
integration program, concurrent system development and ship design
and construction, and the lack of fallback systems in the event a
system fails suggest that major cost increases are likely.  Moreover,
major systems historically encounter unforeseen problems during
development, resulting in cost increases of about 20 to 40 percent.\3
Further, the Navy's estimates for transferring the design to the
second shipbuilder may be understated by hundreds of millions of
dollars.  The estimated start-up costs for the second shipbuilder
range from $1 million to $1 billion. 

The Congress authorized and appropriated funds for advance
procurement of the first two NSSNs--the 1998 start of construction at
Electric Boat and the 1999 start at Newport News.  The National
Defense Authorization Conference Report for fiscal year 1997 modified
the House National Security Committee's recommended provision, which
would have authorized funds for both shipyards to (1) design
improvements for incorporation into the first four NSSNs and (2)
design another new attack submarine that would be more capable but
less costly than the NSSN.  Instead, the National Defense
Authorization Act for fiscal year 1997 provides that the shipbuilders
can propose any design improvements to the first four submarines to
the Secretary of the Navy.  The Secretary will be required to submit
an annual report to the authorization committees on actions taken on
the proposed design improvements.  If any shipbuilder design
improvements for the first four submarines are proposed and accepted,
it would further increase the cost of the NSSN program for the new
design work and the technology insertion.  A new estimated cost would
need to be determined if another new design is proposed and accepted. 
If a new design is proposed and accepted, the estimated cost would
depend on the submarine's attributes and requirements. 


--------------------
\3 Weapons Acquisition:  A Rare Opportunity for Lasting Change
(GAO/NSIAD-93-15, Dec.  1992). 


      COMBAT SYSTEMS RISKS
---------------------------------------------------------- Letter :4.1

The Navy is developing a C3I combat system that will integrate
15 subsystems upon delivery to the shipyard.  The integration will be
at the system level using an open system architecture.  For this
integration, the Navy plans to use new and existing technology,
commercial off-the-shelf (COTS) equipment, and reengineered systems. 
The ship's prime contractor will provide four of the
subsystems--exterior communications, interior communications,
nontactical data processing, and ship monitoring.  The government
will provide the remaining subsystems, including those for radar,
navigation, navigation data distribution and display, imaging, and
electronics support measures.  The government will also provide three
subsystems from a C3I system prime contractor--sonar, combat control,
and architecture and the system-level integration of all subsystems. 

In its April 1995 early operational assessment report on the NSSN,
OPTEVFOR considered the C3I open system architecture a high risk
because it is a very extensive and ambitious effort with a very short
development and integration time frame.  The report noted that the
development schedule is extremely optimistic for the complexity and
scope of the effort.  OPTEVFOR officials stated that the Navy has
never attempted such a large-scale integration effort on a submarine. 
While the BSY-1 and BSY-2 systems did have some level of integration,
the NSSN combat system will have to be totally integrated.  Of
special concern to these officials were the absence of an established
system design, a new open system architecture being used for the
first time on submarines, network security plans that are still
incomplete, a wide aperture sensor system that is still considered
immature, COTS technology whose military utility and supportability
has yet to be proved, and software development and reuse. 

The report noted several areas of risk.  Software development and
reuse were rated high risk because of historical problems associated
with the use of COTS real-time database management systems and
multisource data fusion capabilities.  Problems with the database
management system could cause significant problems with the combat
system.  The extremely short developmental time line, in comparison
with the time lines for developing past C3I systems, specifically the
BSY-2, contributes to the risk.  The combat and control system
architecture, the command workstation, and subsystem integration are
concepts unique to the NSSN C3I combat control system.  The short
developmental time line, incomplete design, and COTS equipment
uncertainties add to the risk. 

According to the Navy's May 1995 NSSN risk assessment, the ship
construction schedule has little flexibility to accommodate
unanticipated development, test, or integration problems or delays in
the C3I system delivery to the shipyard.  Traditionally, however,
problems have arisen in developing similar large systems.  For
example, both the BSY-1 combat system for the Improved Los
Angeles-class and the BSY-2 combat system for the Seawolf-class
submarines had problems that resulted in late delivery and increased
costs.  The BSY-2 combat system was to be delivered in two phases
with all of the hardware and 86 percent of the software in November
1993 and the remaining software in November 1994.  However, the BSY-2
experienced development problems, and the first phase was not
delivered to the shipbuilder until July 1995.  The second phase will
not be delivered until after the ship, which was previously scheduled
for delivery to the Navy in October 1996, or 8-1/2 years after the
award of the BSY-2 contract in March 1988. 

The NSSN combat system is scheduled to be delivered to the shipyard
by November 2000, about 4-1/2 years after award of the combat system
contract.  After delivery of a fully functional C3I system in
November 2000, there will be another 1-1/2 years for integration and
testing activity for the NSSN lead ship.  The total time for
development, integration, and tests prior to ship installation is 6
years, about the same as the original schedule for the BSY-2.  Navy
officials believe the schedule is achievable because the use of COTS
hardware and software will reduce the need and time to develop both
hardware and software. 

With fleet introduction of the NSSN, the Navy will need to support
three different submarine combat systems--the BSY-1 on the 688I
class, the BSY-2 on the three Seawolfs, and the new NSSN fully
integrated C3I system.  This involves separate logistics, training,
and life-cycle support programs. 

DOD and the Navy believe that the risk associated with the C3I system
is low and noted that significant design maturity has occurred since
the April 1995 OPTEVFOR report, which has mitigated many of the
identified risks.  The Navy is currently addressing the concerns
raised by the report.  OPTEVFOR will reassess the NSSN for an early
1997 program review and will issue a follow-up report in December
1996. 


      OTHER AREAS OF RISK
---------------------------------------------------------- Letter :4.2

According to OPTEVFOR officials, subsystems being developed outside
the NSSN program, such as the photonics mast, for inclusion in the
C3I system also pose a potential problem because any cost increases,
technical problems, or delays in these programs could have a major
impact on the NSSN program. 

In addition, the Navy, on the recommendation of industry experts in
open systems, has selected Asynchronous Transfer Mode (ATM) as its
networking technology for integrating the 15 combat subsystems
included in the C3I system.  ATM is considered by some to be a
maturing telecommunications technology; the standard-setting bodies\4
and industries in the telecommunications market have yet to agree on
ATM implementation standards.  Even though the Navy will use
equipment and software from the BSY-1 or BSY-2, some subsystem
interfaces will have to be modified to accommodate the new network
interface requirements to achieve interoperability.  Meanwhile, any
changes to interim ATM standards could result in additional interface
redesign and slow the combat system's testing and integration. 

According to Navy officials, the standard-setting bodies have agreed
to
47 interface standards, and an additional 21 will be finalized by the
end of 1996.  They also stated that the interface standards needed to
integrate the C3I combat system are already well defined.  According
to Navy officials, the C3I development schedule includes a
"technology refresh" concept that provides an opportunity to update
the C3I system before the system is delivered to the Navy.  Any
changes to ATM standards could be incorporated as part of this
process. 

In addition to the ATM, the combat system's lightweight wide aperture
array, which is vital to the ship's combat system performance, is not
yet under development for use on the NSSN. 


--------------------
\4 Telecommunication standards are developed by groups representing
industry, government, and academia that reach consensus on the
implementation of technology to achieve interoperability within the
industry. 


      CONCURRENT DEVELOPMENT AND
      THE LACK OF ALTERNATIVES
---------------------------------------------------------- Letter :4.3

The NSSN will use several new technologies and prototype systems that
are being developed concurrently with ship design and construction
and for which the Navy has no alternative technologies available
should problems arise.  Failure of any one of these systems would
likely result in performance problems, cost increases, and delays in
delivery of the submarine to the Navy. 

An example of a system that could pose a problem is the submarine's
air conditioning system, which requires the use of a new refrigerant
and design of a new air conditioning unit.  This new unit will still
be under development while the ship is being built.  The
technological challenges include the size of the unit and achievement
of the necessary level of quietness.  Prototype testing will overlap
with the construction of the lead ship, and if problems arise,
extensive rework could be necessary.  Since the Navy has no
alternatives for this development, such rework could result in
delayed delivery of the submarine to the Navy. 

The propulsor (propeller) also presents a major technological
challenge.  While the propulsor is being designed for power, speed,
and quieting efficiency, an OPTEVFOR early operational assessment
noted that a full-scale model of the propulsor is needed to determine
whether it will meet its quieting requirements.  Because of the
concurrent development of the propulsor and the ship, however, the
Navy will have to select a final propulsor design before full-scale
testing is complete.  According to the Program Manager, the results
of large-scale testing are promising, but meeting the propulsor
requirements remains a definite technological challenge. 

Another example of a potential problem involves the electronic
warfare support measures (countermeasure surveillance system). 
Because of other Navy priorities, funding for this program was
reduced and the program had to be restructured to meet the needs of
the NSSN.  According to Navy program officials, the system will now
use COTS and existing systems to meet the performance and schedule
requirements of the NSSN. 

The photonics mast program has also been restructured because of a
$10-million increase in cost.  A prototype of the mast and its
photonics sensor are currently at sea being tested on the USS
Phoenix.  According to the NSSN Program Manager, the NSSN is designed
to incorporate a nonpenetrating imaging (i.e., photonics) system as
opposed to the traditional submarine periscope.  If the photonics
mast were unavailable, the Navy would need an alternative sensor that
is compatible with the overall ship design, such as the improved
nonpenetrating periscope (INPP), which is currently being tested on
the USS Phoenix.  The mast is one of the systems being developed
outside of the NSSN program for the C3I system.  If any problems with
sensor performance and delivery arise, there would be some program
impact. 


      DESIGN TRANSFER COSTS ARE
      OPTIMISTIC
---------------------------------------------------------- Letter :4.4

According to program documentation and the Program Manager, the
estimated $2.5 billion (then-year dollars) to build the first Newport
News ship includes $154 million (then-year dollars) to transfer the
submarine design data from Electric Boat to Newport News.  However,
this cost may be understated, according to program officials, since
the last major design transfer for a complex ship design between
shipyards--for the DDG-51--cost about $400 million to $500 million. 
The Program Manager stated that, while optimistic, the $154-million
estimate was based on the use of integrated product and process
development teams, close coordination between the contractors, and
other acquisition reform initiatives. 


      RANGE OF POTENTIAL START-UP
      COSTS
---------------------------------------------------------- Letter :4.5

There is a range of estimates for start-up costs at Newport News. 
According to Navy program officials, costs are estimated to be $250
million.  According to a 1993 U.S.  submarine production base
report,\5 start-up costs at Newport News could range from $607
million to over $1 billion, depending on the length of time various
areas of the shipyard and submarine production lines have been
closed.  According to Newport News officials, the only start-up cost
identified to date is $1 millon to restart the shell and ring
assembly area.  These officials said that there may be other start-up
costs but Newport News has only been involved in the program since
February 1996, and it is too early to develop cost estimates since
the scope of the NSSN program is still evolving. 


--------------------
\5 The U.S.  Submarine Production Base, An Analysis of Cost Schedule
and Risk for Selected Force Structures, Rand, National Defense
Research Institute, 1993. 


   POTENTIAL OPERATIONAL
   EFFECTIVENESS
------------------------------------------------------------ Letter :5

According to Navy officials and documents we obtained, the Navy has
traded performance for cost savings.  Compared with the Seawolf, the
NSSN is slower, carries fewer weapons, and is less capable in diving
depth and arctic operations.  On the other hand, the NSSN is expected
to be as quiet as the Seawolf, will incorporate a vertical launch
system and have improved surveillance as well as special operations
characteristics to enhance littoral warfare capability. 

DOD and the Navy believe the baseline NSSN satisfies military
requirements.  However, an April 1995 report by the OPTEVFOR
Commander expressed concern that if the NSSN were just to meet design
thresholds for survivability, the NSSN may not be operationally
effective against the most capable threat.  The report noted that
there were too many variables and unknowns about the systems involved
to assess whether the design will meet requirements. 

The Navy is addressing the concerns raised by that report.  OPTEVFOR
officials said they are currently involved in reassessing the NSSN
design for an early 1997 program review and expect to issue a
follow-up assessment in December 1996. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :6

We reviewed and analyzed Navy and DOD documents and studies and
discussed the status of the new NSSN, Seawolf submarine, and C3I
combat system programs with Navy program officials in Washington,
D.C., and at the Naval Undersea Warfare Center, Newport, Rhode
Island.  We held discussions about these programs with
representatives from the offices of the Defense Advanced Research
Projects Agency; the Chief of Naval Operations; the Assistant
Secretary of the Navy for Research, Development, and Acquisition; and
the Secretary of Defense.  We also discussed the programs with
representatives from Electric Boat Corporation, Groton, Connecticut,
and Newport News Shipbuilding and Drydock Company, Newport News,
Virginia, and the Supervisors of Shipbuilding at these respective
shipyards.  In addition, we analyzed the results of the NSSN's early
operational assessment and discussed the results with the Commander,
Operational Test and Evaluation Force, Norfolk, Virginia. 

In our cost comparison, we did not calculate any potential benefits
that might result through competitive pressures by introducing a
second shipyard.  Although competition, with a resulting cost benefit
is possible, it is by no means certain.  Achieving any cost benefit
will be determined by how the Navy implements the acquisition
strategy, which has not yet been determined. 

We conducted our review from June 1995 to August 1996 in accordance
with generally accepted government auditing standards. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :7

DOD did not concur with statements in our draft report and said that
(1) the NSSN will not be more costly than the Seawolf; (2) NSSN unit
costs will not continue to rise and significant cost increases should
not be expected due to risk associated with the C3I system or design
transfer between the two shipbuilders; and (3) the NSSN will be able
to defeat the most capable threat. 

Regarding costs, DOD said that on a lead ship basis in fiscal year
1998 dollars, the NSSN would be approximately $600 million less than
construction costs of the Seawolf and on a follow-on ship per unit
basis, the NSSN was estimated to cost at least $300 million less in
fiscal year 1995 dollars.  DOD also said life-cycle costs of the NSSN
are expected to be approximately 15 percent less than those of
Seawolf. 

We did not evaluate life-cycle costs.  In its comments on acquisition
costs, we believe DOD has used inappropriate comparisons that deviate
from previous assessments.  First, a lead ship cost comparison is not
normally used because design, start-up, and other related costs are
included that would not be included in follow-on unit ship cost
comparisons.  For example, the Navy's estimate for the construction
costs of the lead Seawolf includes cost increases related to late and
incomplete design drawings, late development and delivery of
government- furnished equipment such as the BSY-2 combat system,
welding cracks, and problems with the torpedo doors.  Second, the
Navy used a fifth ship acquisition cost for comparison purposes on
both the Los Angeles and Arleigh Burke class programs.  The Navy
believes a fifth ship unit cost is more appropriate for cost
comparison because the influences of learning curves and cost
improvements have occurred and can be realistically reflected in
estimated unit costs.  Use of a lead ship unit cost basis for
comparisons ignores these influences and is different from the
commonly accepted practices DOD and the Navy use in unit construction
costs for comparison purposes. 

DOD said we used the $1.85-billion average Seawolf cost cited in the
Navy's NSSN cost and operational effectiveness analysis and that this
was the average cost of the 2nd through 30th Seawolf built at a
single shipyard and expressed in 1995 dollars.  We did use this basis
to determine the previous acquisition cost of both the Seawolf and
the NSSN.  Our comparison of the average acquisition costs of the
Seawolf and the NSSN, under the original program, is on a comparable
basis (30-ship program, single shipbuilder, and constant 1995
dollars).  We then compared the Navy's cost estimates of the fifth
ship of the NSSN class using a single shipyard and using two
shipyards to show the impact of Public Law 104-106, which
significantly altered the acquisition strategy. 

Regarding the risks associated with the combat systems and concurrent
technology development, DOD said that the risk associated with the
C3I system is low and that significant design maturity has occurred
since the April 1995 OPTEVFOR report, which has mitigated many of the
identified risks.  In providing a status of the NSSN program, our
intent is to identify areas of risk because experience has shown that
integrating new technologies and concurrent development and
production schedules usually result in cost increases.  As we point
out in our report, the C3I system is a highly complex development and
integration effort and there is prototype equipment development
concurrent with ship construction.  The Navy's intent to use COTS
items may prove successful, but plans to do so do not remove risk
from the program. 

DOD did not concur with our statement that design transfer costs are
highly optimistic.  DOD believes that the current estimate is
adequate and that it is not appropriate to use the cost of
transferring the design of the DDG-51 to estimate the cost of
transferring the design of the NSSN.  Despite the introduction of
more sophisticated technology, we continue to believe that the Navy's
estimate may be optimistic because the shipyards use different
computer systems.  The use of the DDG-51 design transfer as a
comparison is appropriate because it is the most recent major design
transfer between shipyards, albeit through a paper-based process. 

Finally, DOD did not concur with our statement that the NSSN has
potential performance limitations.  We have modified our presentation
regarding the NSSN's ability to defeat the most capable threat to
clarify the concerns raised in the OPTEVFOR report and to present
DOD's and the Navy's belief that the baseline NSSN satisfies military
requirements. 

We have made other changes, as appropriate, based on DOD's technical
comments.  (DOD's comments are in app.  I.)


---------------------------------------------------------- Letter :7.1

We are sending copies of this report to the Secretary of the Navy and
the Assistant Secretary of the Navy for Research, Development, and
Acquisition.  Upon request, we will make copies available to other
interested parties. 

Please contact me on (202) 512-4841 if you or your staff have any
questions concerning this report.  Major contributors to this report
are listed in appendix II. 

Louis J.  Rodrigues
Director, Defense
 Acquisitions Issues

List of Congressional Committees

The Honorable Strom Thurmond
Chairman
The Honorable Sam Nunn
Ranking Minority Member
Committee on Armed Services
Unites States Senate

The Honorable Ted Stevens
Chairman
The Honorable Daniel K.  Inouye
Ranking Minority Member
Subcommittee on Defense
Committee on Appropriations
United States Senate

The Honorable Floyd Spence
Chairman
The Honorable Ronald V.  Dellums
Ranking Minority Member
Committee on National Security
House of Representatives

The Honorable C.  W.  Bill Young
Chairman
The Honorable John P.  Murtha
Ranking Minority Member
Subcommitttee on National Security
Committee on Appropriations
House of Representatives




(See figure in printed edition.)Appendix I
COMMENTS FROM THE SECRETARY OF
DEFENSE
============================================================== Letter 



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II

NATIONAL SECURITY AND
INTERNATIONAL AFFAIRS DIVISION,
WASHINGTON, D.C. 

Katherine V.  Schinasi, Associate Director
Richard J.  Price, Assistant Director
Jack G.  Perrigo, Jr., Senior Evaluator

NEW YORK/BOSTON FIELD OFFICE

Richard E.  Silveira, Evaluator-in-Charge
Joseph Rizzo, Jr., Evaluator


*** End of document. ***