Defense Acquisitions: The Expeditionary Fighting Vehicle	 
Encountered Difficulties in Design Demonstration and Faces Future
Risks (01-MAY-06, GAO-06-349).					 
                                                                 
The Marine Corps' Expeditionary Fighting Vehicle (EFV) is the	 
Corps' number-one priority ground system acquisition program and 
accounts for 25.5 percent of the Corps' total acquisition budget 
for fiscal years 2006 through 2011. It will replace the current  
amphibious assault craft and is intended to provide significant  
increases in mobility, lethality, and reliability. We reviewed	 
the program under the Comptroller General's authority to examine 
(1) the cost, schedule, and performance of the EFV program during
system development and demonstration; (2) factors that have	 
contributed to this performance; and (3) future risks the program
faces as it approaches production.				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-06-349 					        
    ACCNO:   A52981						        
  TITLE:     Defense Acquisitions: The Expeditionary Fighting Vehicle 
Encountered Difficulties in Design Demonstration and Faces Future
Risks								 
     DATE:   05/01/2006 
  SUBJECT:   Cost analysis					 
	     Defense procurement				 
	     Developmental testing				 
	     Performance measures				 
	     Procurement evaluation				 
	     Procurement planning				 
	     Procurement practices				 
	     Program management 				 
	     Risk management					 
	     Schedule slippages 				 
	     Weapons research and development			 
	     Weapons systems					 
	     Marine Corps Expeditionary Fighting		 
	     Vehicle						 
                                                                 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO 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.                                         **
**                                                              **
******************************************************************
GAO-06-349

                 United States Government Accountability Office

Report to Congressional Committees

GAO

May 2006

DEFENSE ACQUISITIONS

     The Expeditionary Fighting Vehicle Encountered Difficulties in Design
                      Demonstration and Faces Future Risks

GAO-06-349

DEFENSE ACQUISITIONS

The Expeditionary Fighting Vehicle Encountered Difficulties in Design
Demonstration and Faces Future Risks

  What GAO Found

Although the EFV program had followed a knowledge-based approach early in
development, its buying power has eroded during System Development and
Demonstration (SDD). Since beginning this final phase of development in
December 2000, cost has increased 45 percent as shown in figure 1.

Figure 1: EFV Acquisition Cost Growth Since the Start of SDD Dollars in
billions

                            2001 2002 2003 2004 2005

Source: GAO analysis of program office data.

Unit costs have increased from $8.5 million to $12.3 million. The program
schedule has grown 35 percent or 4 years, and its reliability requirement
has been reduced from 70 hours of continuous operation to 43.5 hours.
Program difficulties occurred in part because not enough time was allowed
to demonstrate maturity of the EFV design during SDD. The SDD schedule of
about 3 years proved too short to conduct all necessary planning and to
incorporate the results of tests into design changes, resulting in
schedule slippages. In addition, several significant technical problems
surfaced, including problems with the hull electronic unit, the bow flap,
and the hydraulics. Reliability also remains a challenge.

Three areas of significant risk remain for demonstrating design and
production maturity that have potential significant cost and schedule
consequences. First, EFV plans are to enter low-rate initial production
without requiring the contractor to demonstrate that the EFV's
manufacturing processes are under control. Second, the EFV program will
begin low-rate initial production without the knowledge that software
development capabilities are sufficiently mature. Third, two key
performance parameters-reliability and interoperability-are not scheduled
to be demonstrated until the initial test and evaluation phase in fiscal
year 2010-about 4 years after low-rate initial production has begun.

                 United States Government Accountability Office

Contents

  Letter 1

Results in Brief 3 Background 5 Cost, Schedule, and Other Problems Have
Reduced EFV Buying

Power 7 Difficulty of Demonstrating Design Maturity Was Underestimated 11
Risks Remain for Demonstrating Design and Production Maturity 21
Conclusions 27 Recommendations for Executive Actions 28 Agency Comments
and Our Evaluation 28

Appendix I Scope and Methodology

Appendix II Comments from the Department of Defense

Appendix III GAO Contact and Staff Acknowledgement

  Tables

Table 1: Program Office Rationales for Rebaselining the EFV

Program Since Entering SDD 9 Table 2: Comparison of Key Events Timing 10

  Figures

Figure 1: Current EFV under Development 5 Figure 2: Comparison of EFV
Acquisition Cost to the Marine Corps'

Total Acquisition Cost for Fiscal Years 2006-2011 (Then

year dollars) 6 Figure 3: EFV Acquisition Cost Growth Since the Start of
System

Development and Demonstration 8 Figure 4: Best Practices for Demonstrating
Design Maturity 12 Figure 5: EFV Hull Electronics Unit 15 Figure 6: EFV
Bow Flap 16 Figure 7: Original Reliability Growth Plan 19 Figure 8:
Current Reliability Growth Plan 20

Abbreviations

DOD          Department of Defense                       
DOT&E        Director, Operational Test and Evaluation   
EFV          Expeditionary Fighting Vehicle              
GDAMS        General Dynamics Amphibious Systems         
GDLS         General Dynamics Land Systems               
HEU          Hull Electronic Unit                        
SDD          System Development and Demonstration        

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office Washington, DC 20548

May 1, 2006

The Honorable John Warner Chairman The Honorable Carl Levin Ranking
Minority Member Committee on Armed Services United States Senate

The Honorable Duncan L. Hunter Chairman The Honorable Ike Skelton Ranking
Minority Member Committee on Armed Service House of Representatives

Congress continues to express concerns over both the costs and the cost
growth of the Department of Defense's (DOD) major acquisition programs1
and not following its own acquisition policies. In the November 2005,
hearing on DOD Acquisition Reform, the House Armed Services Committee
noted that DOD's acquisition costs and capabilities were increasing so
much for individual systems that the nation will not be able to afford
enough of them to support its missions; it also observed that the symptoms
of this problem include increasing costs and programs ignoring internal
regulations and processes.

We have reported on widespread and persistent cost, schedule, and
performance problems with major weapon system developments and DOD's
inability to resolve them. Over the last 9 years, we have benchmarked
successful commercial and defense development programs and identified the
key characteristics for getting better outcomes as being knowledge-based.
Successful programs insist on having key product knowledge demonstrated at
key points in a new development.

We have found that a sound business case at the beginning of the system
development and demonstration (SDD) phase is essential for the

1

Major defense acquisition programs are defined by DOD as those estimated
as requiring an eventual total expenditure for research, development,
test, and evaluation of more than $365 million or for procurement of more
than $2.190 billion in fiscal year 2000 constant dollars.

Page 1 GAO-06-349 Defense Acquisitions

successful completion of a weapon system program.2 Demonstrated knowledge
at key junctures is at the core of the business case. The basic elements
of a sound business case at the start of SDD include:

     o A match must be made between the customer's needs and mature
       technology. We refer to this as knowledge point 1.
          * The acquisition strategy for SDD should provide for
            demonstrating:
               o Design stability at the time of the critical design review
                 (knowledge point 2).
               o The design meets performance requirements, is reliable, and
                 can be produced within cost, schedule, and quality targets
                 before production begins (knowledge point 3).
     o A realistic cost estimate is made to support the acquisition strategy.
     o Sufficient funds are available to cover realistic program costs.

In sum, successful programs insist on having key product knowledge
demonstrated at key points in a new development.

Starting in October 2000, DOD incorporated a knowledge-based approach in
its policy that guides major acquisitions and expanded this approach in
its May 2003 policy.3 The way to implement this policy is through
decisions on individual programs. As we have reported, most individual
programs do not follow a knowledge-based approach, preferring instead to
proceed without adequate knowledge and to accept the consequences of lost
buying power that attend subsequent cost increases.4

The Marine Corps' Expeditionary Fighting Vehicle (EFV) is a major
acquisition program that did show indications of following a
knowledgebased approach and other best practices. For example, the program
earlier adopted best practices in its implementation of Integrated Product
Teams and has trained its program office staff on this acquisition
improvement initiative. In addition, as we have reported, the earlier EFV
program has been a leader both in the use of Integrated Product Teams and
Cost as an

2GAO, Tactical Aircraft: F/A-22 and JSF Acquisition Plans and Implications
for Tactical Aircraft Modernization GAO-05-519T, (Washington, D.C.: April
6, 2005).

Department of Defense Instruction 5000.2, Subject: Operation of the
Defense Acquisition System (May 12, 2003).

4GAO, Defense Acquisitions: Assessments of Selected Major Weapon Programs,
GAO-05-301(Washington, D.C.: March 2005).

Page 2 GAO-06-349 Defense Acquisitions

                                Results in Brief

Independent Variable.5 The EFV program has since been used by the Defense
Acquisition University as a lessons-learned case study for training
acquisition program managers.

We reviewed the EFV program under the Comptroller General's authority to
determine how it is performing against its business case. Specifically,
this report addresses:

     o the cost, schedule, and performance of the EFV program during SDD;
     o factors that have contributed to this performance; and
     o future risk the program faces as it approaches production.

In conducting our review, we used knowledge-based acquisition strategy
principles as a framework. Appendix I contains details of our approach. We
conducted our work from May 2005 to May 2006 in accordance with generally
accepted government auditing standards.

Since the EFV program began the System Development and Demonstration (SDD)
phase, its return on investment has eroded as costs have increased,
deliveries have been delayed, and expected reliability has been lowered.
Since December 2000, the EFV's total cost has grown by about $3.9 billion
or 45 percent, to $12.6 billion. Cost per vehicle has increased from $8.5
million to $12.3 million. Deliveries of vehicles to the warfighter have
been delayed, as planned production quantities have been reduced by about
55 percent over fiscal years 2006-2011, and the development schedule has
grown by about 4 years, or 35 percent. Furthermore, a key requirement has
been lowered. EFV reliability-a key performance parameter-has been reduced
from 70 hours of continuous operation to 43.5 hours.

Program difficulties occurred in part because not enough time was allowed
to demonstrate maturity of the EFV design during SDD. Best practices (and
current DOD acquisition policy) call for system integration work to be
conducted before the critical design review is held. This review
represents the commitment to building full-scale SDD prototypes that are
representative of the production vehicle. In the case of the EFV, however,

5GAO, Best Practices: DOD Training Can Do More to Help Weapon System
Programs Implement Best Practices, GAO/NSIAD-99-206 (Washington, D.C.:
March 1999).

Page 3 GAO-06-349 Defense Acquisitions

the SDD critical design review was held before the system integration work
had been fully completed. While testing of early prototypes began 1 year
before SDD critical design review, it continued for 3 more years after the
decision to begin building the SDD prototypes. The SDD schedule of about 3
years proved too short to conduct all necessary planning and to
incorporate the results into design changes, resulting in schedule delays
and cost increases. Lessons learned from testing the early prototypes
necessitated design changes in the SDD prototypes, which delayed their
delivery and testing. The schedule was delayed further to allow more time
to demonstrate the reliability of the EFV using the SDD prototypes. Even
with the delays, it is clear that the actual test hours accumulated are
significantly less than planned. While the original plan called for
conducting 12,000 hours of testing by September 2005, the current plan
will not achieve this level until after 2008. Also, several significant
problems have surfaced in testing the SDD prototypes, including problems
with the hull electronic unit (HEU), the bow flap, and the hydraulics.

Three areas of risk remain for demonstrating design and production
maturity, which have potential cost and schedule consequences-risks for
the EFV's business case. First, while the EFV program has taken steps and
made plans to reduce risk in the production phase, production risk remains
in the program. Current plans are to enter low-rate initial production
without requiring the contractor to ensure that all key EFV manufacturing
processes are under control. Second, the EFV program will transition to
low-rate initial production without the knowledge that software
development capabilities are mature. Third, two key performance
parameters-reliability and interoperability-are not scheduled to be
demonstrated until the initial test and evaluation phase in fiscal year
2010, about 4 years after low-rate initial production has begun. The
program office has developed plans to resolve performance challenges, and
believes they will succeed. However, until the plans are actually
implemented successfully, the EFV's design and production maturity will
not be demonstrated and the potential for additional cost and schedule
increases remains while production units are being made.

We are making recommendations in this report to the Secretary of Defense
that (1) the EFV program delay Milestone C until design maturity and other
conditions are achieved and (2) draw lessons from the EFV experience that
can be applied to other acquisition programs. After a review of a draft of
this report, DOD concurred with our recommendations and provided some
technical comments that were incorporated, as appropriate.

  Background

The EFV is the Corps' number-one priority ground system acquisition
program and is the successor to the Marine Corps' existing amphibious
assault vehicle. It is designed to transport troops from ships offshore to
their inland destinations at higher speeds and from farther distances, and
to be more mobile, lethal, reliable, and effective in all weather
conditions. It will have two variants-a troop carrier for 17
combat-equipped Marines and a crew of 3 and a command vehicle to manage
combat operations in the field. The Marine Corps' total EFV program
requirement is for 1,025 vehicles. Figure 1 depicts the EFV system.

Figure 1: Current EFV under Development

Source: General Dynamics Land Systems.

The EFV's total acquisition cost is currently estimated to be about $12.6
billion. In addition, the EFV accounts for a substantial portion of the
Marine Corps' total acquisition budget for fiscal years 2006 through 2011,
as figure 2 shows.

Figure 2: Comparison of EFV Acquisition Cost to the Marine Corps' Total
Acquisition Cost for Fiscal Years 2006-2011 (Then-year dollars)

Marine Corps' total acquisition costs($14.1 billion)

EFV acquisition cost ($3.6 billion)

All other Marine Corps' systems ($10.5 billion) Source: GAO analysis of
EFV program office data.

The EFV program began its program definition and risk reduction phase in
1995, and was originally referred to as the Advanced Assault Amphibious
Vehicle. The Marine Corps' existing assault amphibious vehicle was
originally fielded in 1972 and will be over 30 years old when the EFV is
fielded. Several Marine Corps studies identified deficiencies in the
existing vehicle, including the lack of necessary lethality to defeat
projected emerging threats. Despite efforts to extend the service life of
the existing vehicle, Marine Corps officials stated that serious
warfighting deficiencies remained. The studies concluded that the existing
vehicle was unable to perform the type of combat missions envisioned by
the Marine Corps' emerging combat doctrine and that a new vehicle was
needed.6

In September 2003, DOD officially changed the name of the new vehicle to
the EFV, which was in keeping with the Marine Corps' cultural shift from
the 20th century force defined by amphibious operations to a 21st century
force focusing on a broadened range of employment concepts and
possibilities across a spectrum of conflict. The new vehicle is a self-

In 2003, GAO also reported that the existing amphibious assault vehicle
needed attention due to aged equipment that needed upgrading. Military
Readiness: DOD Needs to Reassess Program Strategy, Funding Priorities, and
Risks for Selected Equipment, GAO-04-112 (Washington, D.C.: December
2003).

Page 6 GAO-06-349 Defense Acquisitions

  Cost, Schedule, and Other Problems Have Reduced EFV Buying Power

deploying, high water-speed, amphibious, armored, tracked vehicle, and is
to provide essential command, control, communications, computers, and
intelligence functions for embarked personnel and EFV units. These
functions are to be interoperable with other Marine Corps systems as well
as with Army, Air Force, Navy, and NATO systems. The EFV transitioned to
SDD in December 2000. The use of a knowledge-based acquisition approach
was evident at the onset of the EFV program. Early in the program at the
start of program definition and risk reduction, the Marine Corps ensured
that four of the five critical program technologies were mature. Although
the fifth technology (the moving map navigation technology, which provides
situational awareness) was not mature at this same time, it was
sufficiently matured after the program transitioned to SDD. Furthermore,
the EFV design showed evidence of being stable by the completion and
release of design drawings. At critical design review, 84 percent of the
drawings were completed and released. The program now has 100 percent of
the EFV drawings completed. Program officials expect that only about 12
percent of the design drawings are likely to be changed in the future as a
result of planned reliability testing.

Since entering SDD in December, 2000, the EFV program's total cost has
grown by about $3.9 billion, or 45 percent. 7 Production quantities have
been reduced by about 55 percent over fiscal years 2006-2011, thereby
reducing the capabilities provided to the warfighter during this period.
Cost per vehicle has increased from $8.5 million to $12.3 million.
However, total quantities remain unchanged. During the same period, the
EFV's development schedule has grown by about 4 years, or 35 percent.
Furthermore, a key requirement has been lowered. EFV reliability-a key
performance parameter-has been reduced from 70 hours of continuous
operation to 43.5 hours. Thus, overall EFV buying power has been reduced,
for it will now take substantially more money than was estimated at the
start of SDD to acquire the same number of vehicles later and more slowly,
and with a reduced operational reliability requirement.

7

In constant 2006 dollars, the December 2000 cost is $9.6 billion, for an
increase of $3.1 billion, or 32 percent.

Page 7 GAO-06-349 Defense Acquisitions

EFV Costs and Schedule Since entering SDD in December 2000 and holding the
SDD critical design Have Grown Significantly review in January, 2001, the
EFV program's total acquisition cost has Since Entering SDD grown by about
$3.9 billion, or 45 percent, to $12.6 billion. Figure 3 shows

                        how costs have grown over time.

Figure 3: EFV Acquisition Cost Growth Since the Start of System
Development and Demonstration

Dollars in billions

2000 2001 2002 20032004 2005

Source: GAO analysis of program office data.

While total quantities have not changed, production quantities over fiscal
years 2006-2011 were reduced by about 55 percent, from 461 vehicles to
208. This means that the warfighter will get the capability the EFV
provides more slowly.

The EFV program has been rebaselined three times since SDD began, as shown
in table 1.8

8

A program's baseline is derived from its performance and schedule needs
and the estimates of total program cost consistent with projected funding,
and reflects the program's estimated total acquisition cost and schedule
at the time the baseline is derived. Under certain circumstances, DOD will
"rebaseline" a program--i.e., change its estimated cost and schedule so
that goals more realistically reflect the program's current status.
Rebaselining is useful and appropriate in many situations.

Page 8 GAO-06-349 Defense Acquisitions

Table 1: Program Office Rationales for Rebaselining the EFV Program Since
Entering SDD

Impact on program Date of rebaseline Rationale for rebaselines schedule

November 2002 Prototypes were not delivered as 12-month increase
anticipated; additional time was needed for reliability testing prior to
the Milestone C decision.

March 2003 DOD's Director, Operational Test 12-month increase and
Evaluation directed more time be added for more robust operational testing
prior to Milestone C.

March 2005 Rebaseline was implemented to 24-month increase incorporate the
program changes as a result of DOD's Program Budget Decision 753.

Source: GAO analysis of EFV program office data.

Because the rebaselines have occurred incrementally over time, the EFV
program has not previously been required to submit a unit cost increase
report to Congress. Congress in 1982 enacted the unit cost reporting
statute, now codified in 10 USC 2433, which is commonly referred to as
Nunn-McCurdy, after the congressional leaders responsible for the
requirement. The statute required the Secretary of Defense to certify a
program to Congress when the unit cost growth in constant dollars reaches
25 percent above the most recent rebaseline cost estimate and report to
Congress when it reaches 15 percent. The National Defense Authorization
Act9 for fiscal year 2006 made changes to Nunn-McCurdy. The primary change
that affects the EFV program was the additional requirement to report 30
percent unit cost growth above the original baseline estimate approved at
SDD. The EFV program recently reported an increase in the EFV's program
average unit cost increase of at least 30 percent above its original
baseline estimate at SDD. Although the EFV program acquisition unit costs
have increased by about at least 30 percent since SDD began, no single
increase between rebaselines has reached the 15 percent reporting
threshold.

Overall, the program schedule has grown by 48 months or 35 percent from
December 2000 at the start of SDD to the most recent rebaselining in

9Public Law 109-163.

    Reliability Requirement Reduced

March 2005. This schedule growth has delayed the occurrence of key events.
For example, the EFV program was originally scheduled to provide the
Marine Corps with its initial operational capability vehicles in September
2006, but is now scheduled to provide this capability in September 2010.
Details of key event schedule changes are shown in table 2.

Table 2: Comparison of Key Events Timing
Baseline SDD key                                           Current SDD key 
events (12/2000)   Key events                              events (3/2005) 
December 2000      Milestone B                             December 2000   
                                    (System Development and 
                        Demonstration)                      
January 2001              Critical Design Review           January 2001    
October 2003a             Milestone C                      December 2006   
                             (Low-rate initial Production)  
Start-August 2007           Initial Operational Test &    Start-May 2009   
End-April 2008            Evaluation                      End-January 2010 
August 2008               Full-Rate Production              August 2010    
May 2010                  Deliveries start               May 2012          
September 2006            Initial Operational Capability    September 2010 

Source: GAO analysis of EFV program office data.

a

In 1999, the program office accelerated Milestone C from July 2005 to
October 2003.

In 2005, the Marine Corps received approval to lower the EFV's reliability
requirement from 70 hours before maintenance is needed to 43.5 hours
before maintenance is needed.10 This decision was based on a revised
analysis of the EFV's mission profile and the vehicle's demonstrated
reliability. At the start of SDD, the EFV's operational reliability
requirement was 70 hours of operation before maintenance is needed.
Program officials told us this 70-hour requirement was based on the EFV's
mission profile at the time, which called for a "do-all" mission for one

24.3 hour period of operation. The original reliability growth plan
anticipated that this requirement would be met after initial operational
test and evaluation, which was then planned for August 2007.

10

As measured by mean time (hours) between operational mission failures.

  Difficulty of Demonstrating Design Maturity Was Underestimated

In 2002, the Marine Corps' Combat Development Command performed an
independent analysis of the original 70-hour reliability requirement and
determined that it was likely that it would be very difficult to achieve.
Additionally, the analysis determined that this requirement was
excessively high when compared to similar types of vehicles. In fiscal
year 2004, DOD's Director of Operational Test and Evaluation (DOT&E)
office reported that overall EFV reliability remained a significant
challenge because of the system's comparative complexity and harsh
operating environment. In 2004, The Marine Corps' Combat Development
Command reviewed the 70-hour requirement and recommended that it be
reduced to

43.5 hours. According to program officials, the primary reason for the
reduction to 43.5 hours was to more accurately depict the Marine Corps'
current mission profile for the EFV, which calls for a 12.5-hour mission
day. The Joint Requirements Oversight Council approved the reliability
reduction to 43.5 hours in January 2005.

The program's development schedule did not allow enough time to
demonstrate maturity of the EFV design during SDD. The critical design
review was held almost immediately after SDD began. Testing of early
prototypes continued for 3 years after the decision to begin building the
SDD prototypes. Test schedules for demonstrating design maturity in the
integrated, full-system SDD prototypes proved optimistic and
successoriented, and were extended twice. After the schedules were
extended, major problems were discovered in testing the prototypes.

Best Practices for Conceptually, as figure 4 illustrates, SDD has two
phases: a system Demonstrating Design integration phase to stabilize the
product's design and a system Maturity demonstration phase to demonstrate
the product can be manufactured

                        affordably and work reliably.11

11

GAO, Best Practices: Capturing Design and Manufacturing Knowledge Early
Improves Acquisition Outcomes, GAO-02-701 (Washington, D.C.: July 15,
2002).

Page 11 GAO-06-349 Defense Acquisitions

           Figure 4: Best Practices for Demonstrating Design Maturity

System development           Critical design             Milestone C       
and demonstration                 review           
(Program start)            (Knowledge point 2)         (Knowledge point 3) 
Source: DOD and GAO.                               

The system integration phase is used to stabilize the overall system
design by integrating components and subsystems into a product and by
showing that the design can meet product requirements. When this knowledge
is captured, knowledge point 2 has been achieved. Leading commercial
companies use several criteria to determine that this point has been
achieved, including completion of 90 percent of engineering drawings and
prototype or variant testing to demonstrate that the design meets the
requirements. When knowledge point 2 is reached, a decision review-or
critical design review-is conducted to ensure that the program is ready to
move into system demonstration. This review represents the commitment to
building full-scale SDD prototypes that are representative of the
production vehicle. The system demonstration phase is then used to
demonstrate that the product will work as required and can be manufactured
within targets. When this knowledge is captured, knowledge point 3 has
been achieved. DOD uses this conceptualization of SDD for its acquisition
policy and guidance.12

The EFV program met most of the criteria for SDD critical design review,
which it held January 2001, about 1 month after entering SDD. In
particular, it had 84 percent of drawings completed and had conducted
early prototype testing during the last year of program definition and
risk reduction. However, this early prototype testing had not been fully
completed prior to critical design review. Testing of the early prototypes
continued for 3 years into SDD, well after the program office established
the SDD critical design decision to begin building the SDD prototypes.

Department of Defense Instruction 5000.2, Subject: Operation of the
Defense Acquisition System (May 12, 2003).

Page 12 GAO-06-349 Defense Acquisitions

    Initial SDD Test Schedules Were Optimistic and Success-Oriented

The program did not allow enough time to demonstrate maturity of the EFV
design during SDD. The original SDD schedule of about 3 years proved too
short to conduct all necessary planning and to incorporate the results of
tests into design changes. Specifically, the original schedule did not
allow adequate time for testing, evaluating the results, fixing the
problems, and retesting to make certain that problems are fixed before
moving forward. Testing is the main process used to gauge the progress
provided to the customer. Consequently, it is essential to build
sufficient testing and evaluation time into program development to
minimize or avoid schedule slippages and cost increases being made when an
idea or concept is translated into an actual product.13 Evaluation is the
process of analyzing and learning from a test. The ultimate goal of
testing and evaluation is to make sure the product works as intended
before it is provided to the customer. Consequently, it is essential to
build sufficient testing and evaluation time into program development to
minimize or avoid schedule slippages and cost increases.

Prior to entering SDD, during both the concept evaluation and the program
definition and risk reduction phases, the EFV program conducted a variety
of component and subsystem tests. This testing included an
engineering-model and prototype-testing program, as well as modeling and
simulation test programs. Early EFV testing also included early
operational assessment tests on the initial prototype developed during
program definition and risk reduction. During this phase, the EFV program
demonstrated key aspects of performance including the technological
maturity to achieve the high water speed and land mobility needed for the
EFV mission. In addition, a number of subsystem tests were conducted on
key components of the EFV, including the main engine; water jets;
propulsion drive train components; weapons; nuclear, biological and
chemical filters; track, suspension units; and nearly all of the vehicle
electronics.

Nevertheless, the SDD schedule was extended twice to ensure adequate
system-level testing time. In November 2002, the program office extended
the test schedule by 12 months for additional testing prior to low-rate
initial production. According to program officials, this extension was
necessary for several reasons. Lessons learned from testing the early
prototypes necessitated design changes in the SDD prototypes, which

13GAO, Best Practices: A More Constructive Test Approach Is Key to Better
Weapon System Outcomes, GAO/NSIAD-00-199 (Washington, D.C.: July 31,
2000).

Page 13 GAO-06-349 Defense Acquisitions

delayed delivery and testing of the SDD prototypes. In addition, testing
was taking longer than anticipated, additional time was needed for
reliability testing, and more training was required to qualify crews prior
to certain events. For example, the results of the early EFV firepower,
water operations, and amphibious ship testing revealed the need for more
testing. The schedule was delayed further to allow more time to
demonstrate the reliability of the EFV using the SDD prototypes. In March
2003, DOT&E directed that the EFV test schedule be extended for yet
another 12 months so that more developmental testing and more robust
operational testing could occur before initial production.

EFV Program Encountered After the two schedule adjustments, testing of SDD
prototypes revealed

Design Maturity Problems major problems in maturing the system's design.
Specifically, the program experienced problems with the HEU, bow flap,
system hydraulics, and reliability.

Hull Electronic Unit The HEU provides the computer processing for the
EFV's mobility, power, and auxiliary computer software configuration and
for the command and control software application. Figure 5 shows the HEU.

Figure 5: EFV Hull Electronics Unit

Source: EFV Program Office.

In November 2004, during integrated system-level testing on the SDD
prototypes, there were major problems with the HEU. For example, the
water-mode steering froze, causing the vehicle to be non-responsive to the
driver's steering inputs and both the HEU and the crew's display panel
shut down during EFV operation. Consequently, testing ceased until the
causes of the problems could be identified and corrections made. The
program office conducted a root-cause analysis and traced the problems to
both hardware and software sources. The program office made design changes
and modifications to correct the problems, and testing resumed in January
2005, after about a 2-month delay. According to program officials, these
changes and modifications were installed by May 2005, in the vehicles that
will be used to conduct the operational assessment tests. Again, according
to program officials, these problems have not recurred.

However, the HEU has experienced some new problems in testing since then.
For example, in June 2005, some status indicators on the crew's display
panel shut down during land operations and had to be rebooted. Program
officials commented that corrective actions for HEU problems have been
initiated and tested to ensure that the actions resolved the problems. We
did not independently verify program officials' statements about
initiation and testing of corrective actions.

Bow Flap The bow flap is a folding appendage on the front of the EFV that
is hydraulically extended forward during EFV water operations. The bow
flap provides additional surface area that is used to generate additional
hydrodynamic lift as the vehicle moves through the water. Figure 6 shows
the bow flap.

                             Figure 6: EFV Bow Flap

Source: EFV Program Office.

Prior to entering SDD, major problems occurred with an earlier version of
the bow flap in testing using early prototypes. Root-cause analysis traced
these problems to bow flap overloading. Consequently, the bow flap was
redesigned but was not retested on the early prototypes before the new
design was installed on the SDD prototypes.

Problems with the new bow flaps occurred during subsequent SDD prototype
testing. For example, in September and October 2004, two bow flaps
failed-one bent and one cracked. Again, the program office conducted a
root-cause analysis, which determined that loading-while no longer
excessive-was inappropriately distributed on the bow flaps. Following
corrective action, tests were conducted in Hawaii during July to August
2005 to validate the load capacity of the new bow flap. These tests
revealed that the design of the new bow flap needed some refinements in
order to meet the operational requirement that the EFV be capable of
operating in 3-foot significant wave heights.14 A program official
indicated that the test results will be used to refine the design of the
new bow flap. However, the refined bow flap design will not be tested in
the operationally required 3-foot significant wave heights until initial
operational testing and evaluation, well after the program enters low-rate
initial production.

Hydraulics Hydraulic systems are key components in the EFV. For example,
they control raising and lowering the bow flap, engine cooling systems,
marine steering, and troop ramps. Hydraulic system failures are one of the
top reliability drivers in the EFV program. If the reliability requirement
is to be achieved, the myriad hydraulic problems must be resolved. The EFV
has encountered hydraulic system problems on both early and SDD
prototypes. The top four hydraulic system problems are:

     o Leaks from all sources, particularly leaks due to the loosening of
       fittings and connectors because of vibration during EFV operations.
     o Various component mechanical failures experienced during EFV testing.
     o Hydraulic fluid pressure spikes, particularly in the EFV's
       transmission and pumps.
     o Hydraulic fluid contamination by air, water, and particulates.

Program officials said that the program office has instituted a
design/test/redesign process to identify deficiencies and implement
corrections to increase vehicle reliability. According to program
officials, this process brings together the program office, contractor,
various subcontractor vendors of hydraulic components, and experts from
industry and academia to address and correct hydraulic problems as they
occur. Corrective actions thus far include:

o  Leaks-better sealing of connections; installation of specialized,
self-locking devices at connections most susceptible to vibration leaks;
and replacement of rigid tubing with flexible hoses to absorb vibration.

14

Significant wave height is defined as the distance from the crest to the
trough of the biggest one-third of the waves.

Page 17 GAO-06-349 Defense Acquisitions

                               System Reliability

     o Component mechanical failures-redesigning, strengthening, and
       upgrading various parts.
     o Hydraulic fluid pressure spikes-reducing gear shifting during EFV
       operations and installing devices to control pressure.
     o Hydraulic fluid contamination-flushing hydraulic systems and
       instituting a variety of monitoring, maintenance, and inspection plans
       to maintain hydraulic fluid and component cleanliness requirements.

Program officials noted that corrective actions thus far have been tested
to ensure that they resolved the problems, and have been installed on the
SDD prototype vehicles. We did not independently verify this.

Based on lower demonstrated reliability and problems with early program
testing, the EFV's reliability has not grown as planned. Expectations for
reliability are now lower, as reflected in the recent reduction to the
reliability requirement. When SDD began, the EFV was expected to
demonstrate 48 hours between failures by September 2005. Actual growth
demonstrated 28 hours between failures in August 2005. At the time of the
low-rate initial production decision now planned for December 2006,
demonstrated reliability is projected to be 38 hours between failures. The
original and current reliability growth curves for the EFV are shown in
figures 7 and 8, respectively.

Figure 7: Original Reliability Growth Plan Mean time (hours) between
operational mission failures

Initial operational test,

and evaluation about 48 hours

Milestone C

80

60 hours9/2005 8/2007

70

Reliability goal-

70 hours

60 50 40 30 20 10

      0 1,000 12,000 28,000 48,000 Growth test hours

Source: GAO analysis of EFV program office data.

      Figure8: Current Reliability Growth Plan

Mean time (hours) between operational mission failures80

70

60

50

40

30

20

10

        0 1,000 13,000

        Growth test hours

Source: GAO analysis of EFV program office data.

In comparing the planned and actual reliability growth curves, it is clear
that the actual test hours accumulated have been significantly less than
planned. In fact, the original plan called for conducting 12,000 hours of
testing by the original September 2005 production decision; according to
the current plan, test hours will not reach this level until early 2008.
The reduction in test hours is due, in part, to the other problems that
occurred in testing. The accumulation of test hours is significant for
reliability. In general, reliability growth is the result of an iterative
design, build, test, analyze, and fix process. Initial prototypes for a
complex product with major technological advances have inherent
deficiencies. As the prototypes are tested, failures occur and, in fact,
are desired so that the product's design can be made more reliable.
Reliability improves over time with design changes or manufacturing
process improvements.

The program office acknowledges that even with the changes in mission
profile and reduction in the operational requirement, reliability for the
EFV remains challenging. In addition, the most recent DOT&E annual report
found that the EFV system's reliability is the area of highest risk in the
program.15 DOT&E has reviewed the EFV's current reliability growth

Director of Operational Test and Evaluation's Fiscal Year 2005 Annual
Report, December 2005.

Page 20 GAO-06-349 Defense Acquisitions

  Risks Remain for Demonstrating Design and Production Maturity

plan and believes that it is realistic but can only be validated during
initial operational testing and evaluation in 2010.

According to the program manager, an additional 15 months would have been
needed for more robust reliability testing, production qualification
testing, and training, after the program entered low-rate initial
production in September 2005, as originally planned. The March 25, 2005,
rebaselining extended the schedule by 24 months and postponed low-rate
initial production until September 2006, which has now been extended to
December 2006. While DOD's December 2004, Program Budget Decision 753
served as the catalyst for this rebaselining, the program manager stated
that he probably would have asked for a schedule extension of 15 months
after entering low-rate initial production in September 2005, even if the
budget decision had not occurred. DOD and Marine Corps officials verified
that, although the program manager did not officially request this
15-month extension, he had been discussing an extension with them before
the budget decision was issued. However, to the extent that the extra 9
months resulting from the budget decision prove unneeded for program
management reasons, they will be an added cause for schedule and cost
growth.

Three areas of risk remain for demonstrating design and production
maturity, which have potential cost and schedule consequences-risks to the
EFV business case. First, while the EFV program has taken steps and made
plans to reduce risk in the production phase, production risk remains in
the program. Current plans are to enter low-rate initial production
without requiring the contractor to ensure that all key EFV manufacturing
processes are under control. Second, the EFV program will transition to
initial production without the knowledge that software capabilities are
mature. Third, two key performance parameters- reliability and
interoperability-are not scheduled to be demonstrated until the initial
operational test and evaluation phase in fiscal year 2010, about 4 years
after low-rate initial production has begun. The program office has
developed plans to resolve performance challenges and believes it will
succeed. However, until the plans are actually implemented successfully,
the EFV's design and production maturity will not be demonstrated and the
potential for additional cost and schedule increases remains.

    Manufacturing Process Maturity Problems

While the EFV program has taken steps and made plans to reduce risk in the
production phase, production maturity risk remains in the program. Current
EFV program plans are to enter low-rate initial production without
requiring the contractor to ensure that all key EFV manufacturing
processes are under control, i.e., repeatable, sustainable, and capable of
consistently producing parts within the product's tolerance and standards.
Establishing such control is critical to ensuring that the EFV can be
produced reliably and without unexpected production problems. In addition,
DOD's system acquisition policy provides that there be no significant
manufacturing risks prior to entering low-rate initial production and that
manufacturing processes be under statistical process control prior to
starting full-rate production.16

Leading commercial firms rely on statistical process control to ensure
that all key manufacturing processes are under control before they enter
production.17 Statistical process control is a technique that focuses on
reducing variations in manufactured parts, which in turn reduces the risk
of entering production with unknown production capability problems.
Reducing and controlling variability lowers the incidence of defective
parts and thereby products, which may have degraded performance and lower
reliability. Defects can also delay delivery and increase support and
production costs by requiring reworking or scrapping. Consequently, prior
to entering production, leading commercial firms collect and analyze
statistical process control data. Leading commercial firms also use a
measure of process control called the process capability index to measure
both the consistency and the quality of output of a process. DOD's
acquisition policy applies a lower standard. It provides that there be no
significant manufacturing risks prior to entering low-rate initial
production and that manufacturing processes be under statistical process
control prior to starting full-rate production.18

The EFV program is working toward the DOD standard. EFV program officials
said that statistical process control will not be used to ensure that all
key EFV manufacturing processes are under control prior to entering

16

Department of Defense Instruction 5000.2, Subject: Operation of the
Defense Acquisition System (May 12, 2003).

17

GAO, DOD Acquisition Outcomes: A Case for Change GAO-06-257T (Washington,
D.C.: November 15, 2005).

18

Department of Defense Instruction 5000.2, Subject: Operation of the
Defense Acquisition System (May 12, 2003).

Page 22 GAO-06-349 Defense Acquisitions

    Software Development Capability Maturity Problems

low-rate initial production. They stated that they have taken actions to
enhance EFV production readiness. For example, they noted that one of the
most important risk mitigating actions taken was ensuring that SDD
prototypes were built using production-representative tooling and
processes. Program officials also believe that production process maturity
will be demonstrated by achieving repetitive schedule and quality
performance during low-rate initial production. In addition, the program
plans to collect statistical process control data during low-rate initial
production to track equipment and machine performance and detect
statistical shifts. The program believes that using statistical process
control data in this manner will result in earlier detection of machine
malfunctions. Program officials told us that once sufficient quantities of
the EFV are produced and baseline statistical process control data
collected, the results of the analyses of this data will be implemented
for any production measurements that demonstrate process stability. The
program office believes that this approach will allow for use of
statistical process control for implementation of stable manufacturing
processes during low-rate initial production. However, the program office
does not plan to set and achieve a process capability index for the EFV
production efforts.

The actions taken by the program may help to mitigate some production
risk. In fact, EFV's plan to collect and use statistical process control
data goes further than what we have found on most DOD weapon system
programs. However, these actions do not provide the same level of
confidence as having the manufacturing processes under statistical process
control before production. The EFV program's approach of foregoing such
control increases the risk of unexpected production problems during
manufacturing. This risk is compounded by the fact that plans call for
reliability and interoperability, along with resolution of other technical
problems, to be operationally tested and demonstrated during low-rate
initial production, not before.

Under current plans, the EFV program is at risk of entering low-rate
initial production before software development capabilities are mature.
Again, leading commercial firms ensure that software development
capabilities are mature before entering production in order to prevent or
minimize additional cost growth and schedule delays during this phase.19

19

GAO, Best Practices: A More Constructive Test Approach Is Key to Better
Weapons System Outcomes, GAO/NSIAD-00-199 (Washington, D. C.: July 31,
2000).

Page 23 GAO-06-349 Defense Acquisitions

Furthermore, DOD's weapon system acquisition policy calls for weapon
systems to have mature software development capabilities before they enter
low-rate initial production.20

In assessing software capability maturity, commercial firms, DOD, and GAO
consider the software capability maturity model developed by Carnegie
Mellon University's Software Engineering Institute to be an industry
standard. 21 This model focuses on improving, standardizing, and
certifying software development processes, including key process areas
that must be established in the software developer's organization. The
model is essentially an evolutionary path organized into five maturity
levels:

     o Level 1, Initial-the software process is ad hoc and occasionally
       chaotic. Few processes are defined, and success depends on individual
       effort.
     o Level 2, Repeatable---basic project management processes are
       established to track cost, schedule, and functionality. The necessary
       process discipline is in place to repeat earlier successes on projects
       with similar applications.
     o Level 3, Defined-the software process for both management and
       engineering activities is documented, standardized, and integrated
       into a standard process for the organization. All projects use an
       approved, tailored version of the organization's standard process for
       developing and maintaining software.
     o Level 4, Managed-Detailed measures of the software process and product
       quality are collected. Both the software development process and
       products are quantitatively understood and controlled.
     o Level 5, Optimizing-Continuous process improvement is enabled by
       quantitative feedback from the process and from plotting innovative
       ideas and technologies.

The EFV program has had problems with maturing its software development
capabilities. The EFV's prime contractor, General Dynamics Land Systems
(GDLS), which at the time had a level 3 maturity software

20

DOD Instructions 5000.2, Subject: Operation of the Defense Acquisition
System (May 12, 2003).

21

GAO, Defense Acquisitions: Stronger Management Practices Are Needed to
Improve DOD's Software-Intensive Weapon Acquisitions, GAO-04-393
(Washington, D.C.: March 1, 2004).

Page 24 GAO-06-349 Defense Acquisitions

capability, developed all software for the early EFV program.22 According
to the program office, when the program entered SDD, responsibility for
EFV's software development was transferred to GDLS' amphibious development
division, General Dynamics Amphibious Systems (GDAMS). GDAMS has a level 1
maturity software capability. Consequently, the SDD contract required GDLS
to achieve a software development capability maturity level 3 for all EFV
software contractors and subcontractors within 1 year of the contract
award date, July 2001. In January 2002, the program extended this
requirement by 1 year, until July 2003. Nevertheless, while GDAMS twice
attempted to achieve level 3 software development capability maturity, it
did not succeed.

Program officials considered GDAMS's inability to achieve an acceptable
level of software development capability maturity a risk to the program.
To mitigate this risk, in January 2004, the program manager began
developing a risk mitigation plan. As part of this plan, representatives
from the EFV program office, GDAMS, and Ogden Air Logistics Center's 309th
Software Maintenance Group-a certified level 5 maturity software
development organization-formed a Software Partnership Working Group to
address software development capability maturity issues. As of February
2006, EFV program officials were in the process of negotiating a
memorandum of agreement with the 309th Software Partnership Working Group
to develop the EFV's low-rate initial production software. The 309th will
work in partnership with GDAMS as specified by the terms of the memorandum
of agreement. Its involvement is to ensure that the EFV's software
development capability will be at the desired maturity level.

However, the 309th Software Maintenance Group will not complete the
software development for the EFV's low-rate initial production version
until September 2006. Furthermore, GDAMS does not plan to insert this
software into the EFV vehicles until fiscal year 2008, well after low-rate
initial production has begun. This means that the low-rate initial
production decision will be made without the integration of mature
software. Furthermore, the software itself will not be demonstrated in the
vehicle until well into low-rate initial production. While the program
office believes that the level of software risk is an acceptable level
risk, we have

22GDLS now has level 5 certification.

                 Performance Challenges Not Yet Fully Resolved

found that technology-including software-is mature when it has been
demonstrated in its intended environment.23

While involving the 309th Software Maintenance Group helps to mitigate the
risk of immature software development capability in the EFV program, it
increases certain other risks. The memorandum of agreement distributes the
responsibility for software development between the three participants.
However, much of the responsibility for developing a working software
package in an acceptably mature environment shifts from the prime
contractor to the Marine Corps. The software will now become
government-furnished equipment or information. In essence, the Marine
Corps has now assumed much of the risk in the software development effort.
If the software does not work according to the requirements, it will be
incumbent upon the Marine Corps-not the prime contractor, GDLS-to correct
the problems. Furthermore, if the integration of the government-furnished
software into the vehicles creates additional problems, the Marine Corps
could be responsible for corrections. Both of these situations could lead
to cost and schedule growth, and thus increase risks to the program.

Several EFV performance challenges are not yet fully resolved.
Specifically, a key performance parameter-interoperability-cannot be
properly demonstrated until initial operational testing and evaluation in
fiscal year 2010, well after low-rate initial production has begun.
Interoperability means that the EFV communication system must provide
essential command, control, communications, and intelligence functions for
embarked personnel and EFV units. In addition, the EFV communication
system must be compatible-able to communicate-with other Marine Corps
systems as well as with Army, Navy, Air Force, and North Atlantic Treaty
Organization systems. In order to demonstrate interoperability, the EFV
must participate in operational tests that involve these joint forces.
Another key performance parameter-reliability-has been problematic and
still presents a significant challenge.24 It also is not scheduled to be
demonstrated until initial operational testing and evaluation.
Furthermore, the bow flap has been problematic and, while

23

GAO Missile Defense: Knowledge-Based Practices Are Being Adopted, but
Risks Remains, GAO-03-441 (Washington, D.C.: April 30, 2003).

24

Director, Operational Test and Evaluation's FY 2005 Annual Report,
December 2005.

Page 26 GAO-06-349 Defense Acquisitions

                                  Conclusions

improved, still requires some design refinement and has not yet been
successfully tested at its operational performance level.

Program officials commented that they have developed plans to resolve
remaining EFV performance challenges and are optimistic that these plans
will be implemented effectively and testing successfully completed.
However, there are no guarantees that this will actually happen.
Consequently, the performance challenges remain risks to the program until
they are fully resolved with effective solutions actually demonstrated.

The EFV has encountered risks to its business case because of problems
encountered in full-system testing, coupled with an SDD schedule that did
not allow enough time for conducting the testing and learning from it.
Using the lens of a knowledge-based business case, the start of SDD was
sound on requirements and technology maturity (knowledge point 1). While
design stability was judged to be attained at the critical design review
(knowledge point 2) immediately after entering SDD, it appears that
holding critical design review so soon was premature. The acquisition
strategy did not provide the resources (time and money) necessary to
demonstrate design maturity and production maturity (knowledge point 3).
However, we do note that the EFV program is planning to do more with
statistical process control than most other programs we have reviewed.

In retrospect, the EFV program would have been more executable had the SDD
phase allowed for completion of early prototype testing before holding the
SDD critical design review and committing to building the SDD prototypes.
Another lesson learned is that while it is necessary to demonstrate one
knowledge point before a subsequent one can be demonstrated, this alone is
not sufficient. Attaining one knowledge point does not guarantee the
attainment of the next one. Rather, the acquisition strategy for any
program must adequately provide for the attainment of each knowledge point
even in programs, such as the EFV, which were in a favorable position at
the start of SDD.

The EFV program has put into place a number of corrective actions and
plans to overcome and mitigate weaknesses in acquisition strategy.
Nevertheless, design, production, and software development capability
maturity have not yet been fully demonstrated and technical problems fully
corrected. It is important for the business case for the EFV to remain
valid in light of these changes and that the remainder of SDD adequately
provide for the demonstration of design, production, and software
development capability maturity before committing to production.

While these problems must be acknowledged and addressed, the fact that the
EFV program has had a number of sound features should not be overlooked.
In this vein, the program can still be the source of lessons that DOD can
apply to other programs. In particular, it is important that all of the
elements of a sound business case be present at the start of SDD. While it
is generally recognized that missing an early knowledge point will
jeopardize the remaining ones, it must also be recognized that later
knowledge points are not guaranteed even if early ones are achieved. If
the acquisition strategy does not adequately provide for the attainment of
all knowledge points, the estimates for cost and schedule will not have a
sound basis.

We are recommending that the Secretary of Defense ensure that:

     o EFV design, production, and mature software development capabilities
       are demonstrated before Milestone C;
     o adequate resources are available to cover such demonstration and
       provide for risks; and
     o the business case for EFV (including cost and expected capability),
       after including the above, still warrants continued investment.

We also recommend that the Secretary of Defense draw lessons learned from
EFV and apply them to the Defense Acquisition University's curriculum for
instructing program executives, managers, and their staffs. Such lessons
might include understanding that attaining one knowledge point does not
guarantee the attainment of the next one; the importance of having a sound
business case for each phase of development; the right time to hold a
critical design review; and the importance of allowing sufficient time to
learn from testing.

  Recommendations for Executive Actions

                       Agency Comments and Our Evaluation

In commenting on a draft of our report, DOD's Acting Director for Defense
Systems concurred with our recommendations. In doing so, DOD stated that
the Department currently plans to assess the readiness of the EFV program
for a low-rate initial production decision within a year. This assessment
will review the maturity of the EFV design, including software, its
production readiness for low-rate initial production, and its demonstrated
capability, as well as program costs and risks. Continued investment in
EFV will be based on that information. The full text of the department's
response is in appendix II.

The Department notes that our best practices construct for production
readiness is difficult to reconcile with its current acquisition
production decision points. World class companies we have visited do, in
fact, often have a limited production run that they use to manufacture a
small number of production representative assets; however, they do not
make a decision to invest in the tooling necessary to ramp up to full
production until after those assets have been tested by the customer and
their critical manufacturing processes are in control. DOD's low-rate
initial production decision reflects the decision to invest in all of the
resources needed to achieve full-rate production. We believe this is too
soon and that DOD would benefit from this lesson by focusing low-rate
initial production on demonstrating the product and process and waiting to
invest in more resources, such as tooling, to ramp up until the full-rate
production decision has been made.

We are sending copies of this report to the Secretary of Defense,
Secretary of the Navy, and other interested parties. We will also provide
copies to others on request. In addition, the report will be available at
no charge on the GAO Web site at http://www.gao.gov.

If you or your staff have any questions about this report, please contact
me on (202) 512-4841. Contact points for our Offices of Congressional
Relations and Public Affairs may be found on the last page of this report.
GAO staff who made major contributions to this report are listed in
appendix III.

Paul L. Francis Director, Acquisition and Sourcing Management.

                       Appendix I: Scope and Methodology

To assess the current status of the EFV (particularly the status of the
production decision), the factors that contributed to the current status,
and future risks in the program, we interviewed key officials from DOD's
Director, Operational Test and Evaluation, the Office of the Secretary of
Defense's Program Analysis and Evaluation office, the U.S. Marine Corps,
Isothermal Systems Research, Inc., in Washington, D.C., and the 309th
Software Maintenance Group, in Ogden, Utah. We also interviewed the Direct
Reporting Program Manager for the EFV and the prime contractor, General
Dynamics Land Systems, in Woodbridge Virginia. We examined and analyzed
pertinent program documentation, including the Selected Acquisition
Reports; Test and Evaluation Master Plan; Developmental Testing Schedule;
Budget Justification documents, Program Management Plan; Acquisition
Strategy Plan; DOD's Operational Testing, and Evaluation reports;
Operational Requirement Documents, and the Software Development Plan. We
relied on previous GAO work as a framework for knowledge-based
acquisition.

Appendix II: Comments from the Department of Defense

Appendix II: Comments from the Department of Defense

Appendix II: Comments from the Department of Defense

Appendix III: GAO Contact and Staff Acknowledgement

Paul Francis (202) 512-4841

  GAO Contact

In addition to the contact named above, D. Catherine Baltzell, Assistant
Director; Leon S. Gill; Danny Owens; Steven Stern; Martin G. Campbell; and
John Krump made key contributions to this report.

(120447)

  GAO's Mission

The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site (www.gao.gov). Each weekday, GAO posts GAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To

have GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

                             Order by Mail or Phone

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

  To Report Fraud, Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm

  E-mail: [email protected]

Federal Programs Automated answering system: (800) 424-5454 or (202)
512-7470

Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Relations
Washington, D.C. 20548

Paul Anderson, Managing Director, [email protected] (202) 512-4800

  Public Affairs

U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

    PRINTED ON

RECYCLED PAPER
*** End of document. ***