Defense Acquisitions: Realistic Business Cases Needed to Execute
Navy Shipbuilding Programs (24-JUL-07, GAO-07-943T).
The Navy is beset with long-standing problems that affect its
ability to accomplish ambitious goals for its shipbuilding
portfolio. Significant cost growth and long schedule delays are
persistent problems. Making headway on these problems is
essential in light of the serious budget pressures facing the
nation. This testimony focuses on (1) cost growth in
shipbuilding, (2) acquisition approaches in the LPD 17, Littoral
Combat Ship, DDG 1000 and CVN 78 programs and (3) steps the Navy
can take to improve its acquisition decision-making, particularly
the adoption of a knowledge-based framework.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-07-943T
ACCNO: A73210
TITLE: Defense Acquisitions: Realistic Business Cases Needed to
Execute Navy Shipbuilding Programs
DATE: 07/24/2007
SUBJECT: Best practices
Construction costs
Cost analysis
Defense cost control
Financial analysis
Fixed price contracts
Future budget projections
Military vessels
Naval procurement
Schedule slippages
Ships
Cost estimates
Cost growth
DDGX Class Destroyer
Littoral Combat Ship
Virginia Class Submarine
DDG-1000 Destroyer
******************************************************************
** 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-07-943T
* [1]Summary
* [2]Cost Growth Remains a Problem in Navy Shipbuilding Programs--
* [3]Shipbuilding Programs Often Have Unexecutable Business Cases
* [4]Elements of a Business Case
* [5]Business Cases Deteriorated with Construction of LPD 17 and
* [6]DDG 1000 and CVN 78 Have More Thoughtful Business Cases, but
* [7]DDG 1000
* [8]CVN 78
* [9]A Disciplined, Knowledge-Based Process Is Key to Better Outc
* [10]Aligning Knowledge and Decision Points Consistently across S
* [11]Establishing Executable Program Budgets through Improved Cos
* [12]Better Management of Costs through Fixed-Price Contracting a
* [13]Fixed-priced-Contracting for Construction
* [14]Cost Surveillance
* [15]Objectives, Scope, and Methodology
* [16]Contact and Staff Acknowledgments
* [17]Related GAO Products
* [18]GAO's Mission
* [19]Obtaining Copies of GAO Reports and Testimony
* [20]Order by Mail or Phone
* [21]To Report Fraud, Waste, and Abuse in Federal Programs
* [22]Congressional Relations
* [23]Public Affairs
Testimony
Before the Subcommittee on Seapower and Expeditionary Forces, Committee on
Armed Services, House of Representatives
United States Government Accountability Office
GAO
For Release on Delivery
Expected at 2:00 p.m. EDT
Tuesday, July 24, 2007
DEFENSE ACQUISITIONS
Realistic Business Cases Needed to Execute Navy Shipbuilding Programs
Statement of Paul L. Francis, Director
Acquisition and Sourcing Management Team
GAO-07-943T
Mr. Chairman and Members of the Subcommittee,
I am pleased to be here today to discuss the Department of the Navy's
shipbuilding programs, including its surface combatant programs. The Navy
has ambitious goals for its shipbuilding programs. The Navy expects to
build more--and often increasingly complex ships--and deliver them to meet
warfighter needs, while still achieving reduced acquisition and/or life
cycle costs. These are admirable goals, representing the Navy's desire to
provide the fleet with the most advanced ships to meet national defense
and military strategies. However, there is often tension among the Navy's
cost, schedule, industrial base, and capability goals. While this tension
is embedded at the beginning of shipbuilding programs, its effects are
realized later, during ship construction. Budgets set prior to beginning
construction are not realistically achievable and often include optimistic
dates for delivery to the fleet. The consequence is often that costs
increase after construction has begun, schedule targets slip, and contract
scope is reduced. The LPD 17--the lead ship of the San Antonio class
amphibious ships--is a case in point. The cost to construct the ship has
more than doubled, delivery was delayed by over 3 years, and ship quality
ultimately compromised.
Today, I would like to discuss (1) cost growth in shipbuilding programs,
(2) acquisition approaches in the LPD 17, Littoral Combat Ship (LCS), the
next-generation destroyer and aircraft carrier programs (DDG 1000 and CVN
78, respectively), and (3) steps the Navy can take to improve its
acquisition decision making, particularly the adoption of a
knowledge-based management framework.
Summary
Cost growth in shipbuilding programs remains a problem. Ships under
construction at the beginning of the fiscal year have experienced
cumulative cost growth almost $5 billion above their original budgets.
Cost growth displaces other ships contemplated in the Navy's 30-year
shipbuilding plan and reduces the buying power of the shipbuilding budget.
This cost growth illustrates the problems that arise when programs proceed
without a solid business case. A business case requires a balance between
the concept selected to satisfy warfighter needs and the
resources--technologies, design knowledge, funding, time, and management
capacity--needed to transform the concept into a product, in this case a
ship. Both the LPD 17 and the LCS programs illustrate the perils of
proceeding without a solid, executable business case. Both programs pushed
ahead without stable designs or realistic cost and schedule estimates,
resulting in higher costs, schedule delays, and quality problems.
A paradigm shift is needed for shipbuilding programs. Technology maturity
must be proved before a design can be considered stable, and production
outcomes cannot be guaranteed until a stable design is demonstrated. The
Navy also needs to
o define and align knowledge points more consistently across
programs to optimize resource allocation and improve performance;
o ensure initial shipbuilding budgets are realistic by improving
cost estimating, and
o improve cost management through increased use of fixed-price
contracting and comprehensive cost surveillance.
Cost Growth Remains a Problem in Navy Shipbuilding Programs--and May Threaten
Future Success
Cost growth is a persistent problem for shipbuilding programs as it is for
other weapon systems. Over 40 ships were under construction at the
beginning of this fiscal year. If the performance of future shipbuilding
programs continues at the same rate as current programs, the Navy will be
forced to fund cost growth in future budget years at the expense of other
ships in the Navy's shipbuilding plan.
Funding for the 41 ships under construction is over $56 billion, almost
$4.6 billion above initial budget requests.^a Congress has already
appropriated additional funds to cover most of these cost increases (see
table 1).
^aBased on the fiscal year 2008 President's budget request and over $513
million in fiscal year 2007 funding transfers from other Navy programs.
Table 1: Cost Growth in Program Budgets for Ships under Construction in
Fiscal Year 2007
Dollars in millions
Cost growth as
Initial Fiscal year 2008 or latest Total cost a percent of
Ship budget President's budget growth initial budget
CVN 77 $4,975 $5,822 $847 17%
DDG 100-112 14,309 14,679 370 3
LCS 1-LCS 2^a 472 1,075 603 128
LHD 8 1,893 2,196 303 16
LPD 18-23^b 6,194 7,742 1,548 25
SSN 775-783^c 20,744 21,678 934 5
T-AKE 1-9 3,354 3,386 32 1
Total: 41 $51,941 $56,578 $4,637
ships
Source: GAO analysis of Navy data.
aIncludes about $484 million in reprogrammed funding requested by the Navy
through June 2007. A small amount of these funds may be designated for
certain LCS research and development activities.
bIncludes $29.3 million in reprogrammed funding requested by the Navy in
2007 to complete LPD 18 and LPD 20.
cThe Navy has transferred $25.5 million in funding from SSN 776 to cover
the costs of completing SSN 775 after delivery--and believes that
additional unfunded shortfalls may still exist.
Note: For ships constructed on the Gulf Coast, cost growth can be
attributed in part to the effects of Hurricanes Katrina and Rita. The Navy
has already received over $1.1 billion in funding, and an additional $1.3
billion has also been appropriated for hurricane-related damages, but it
has not yet been allocated to individual programs.
Breaking these costs down further reveals the dynamics of shipbuilding
cost growth and the challenges it presents for the 30 year shipbuilding
plan. For example, cost growth in mature programs, like the Arleigh Burke
class destroyers (DDG 100-112) is low because most cost growth has already
been captured in earlier ships. Cost growth in the Virginia class
submarines (SSN 775-783) and the Lewis and Clark class auxiliary ships
(T-AKE 1-9) is also low because they include several ships early in
construction--before cost growth tends to occur. On the other hand, cost
growth is particularly high on lead ships of a new class (see fig. 1).
Figure 1: Cost Growth in Lead Ships and Significant Follow-ons (Dollars in
Millions)
^aLCS 1 and 2 include about $484 million in reprogrammed funding requested
by the Navy through June 2007. A small amount of these funds may be
designated for certain LCS research and development activities.
^bIncludes $20.6 million in reprogrammed funding requested by the Navy in
2007 to complete LPD 18.
^cSSN 774 and LPD 17 were delivered in October 2004 and January 2005,
respectively.
^dThe Navy has transferred $25.5 million in funding from SSN 776 to cover
the costs of completing SSN 775 after delivery--and believes that
additional unfunded shortfalls may still exist.
Cost growth for recent lead ships has been on the order of 27 percent.^b
The Navy is developing two lead ships in the LCS program--each with a
unique design. These ships have already experienced a 128 percent cost
growth. Cost increases are also significant if the second ship is
assembled at a different shipyard than the first ship. This is the case
with SSN 775, which has had cost growth of well over $500 million.
Although the ship has been delivered, the Navy continues to incur costs
for unfinished work.
^bBased on the initial and most recent President's budget request for all
lead ships authorized between fiscal year 1996 and fiscal year 2006,
including the second ship when the hull is constructed at a different
shipyard than the first ship of the class.
Follow-on ships in many cases are also experiencing significant cost
increases in construction. Although LPD 18 is the second ship of the
class, construction costs grew by over $500 million--a more than 90
percent increase over its initial budget for construction. LPD 18 has
recently been delivered, but the Navy requested an additional $20.6
million in reprogrammed funding in 2007 to complete the ship. Cost growth
is particularly prevalent where major changes were made to an existing
ship design. For example, CVN 77 is the final aircraft carrier of the
Nimitz class and is based on the design of previous carriers, but it
included over 3,500 design changes. CVN 77 has experienced cost growth in
construction of over $847 million--a 17 percent increase over the initial
budget.
Besides lead and mature ships, a number of ships under construction may
not have realized the full extent of cost growth--which tends to lag
behind the initial budget request by several years. In fact, the magnitude
of cost growth occurs in later phases of construction--after ships are 60
percent or more complete (see fig. 2).
Figure 2 Cost Growth in Ships by Percent of Construction Completed
The current budgets for many ships have already proven inadequate to cover
the likely costs to complete construction. Funding has been transferred
from other Navy programs, obtained through prior year completion requests
or shifted away from future build plans. The most prominent example is the
LCS program. The Navy has already reprogrammed almost $485 million to fund
cost increases for the first two LCS and deleted three Flight 0 ships from
its budgets. The Navy transferred about $62 million in funding from future
T-AKE ships to cover cost increases on the first two ships under
construction. In the Virginia-class program, the Navy estimates about $130
million shortfall and plans to cover the shortfall from transfers from
within the program. In December 2006, the Navy believed that about $67
million would be needed to complete LPD 20 and LPD 21. However, Navy
officials stated that these estimates are too conservative because they
represent cost growth against the current contract baseline for LPD ships
under construction. The Navy anticipates increasing the baseline for the
LPD 17 ships--resulting in even higher completion costs.
If current patterns of performance continue in the future, the Navy's
shipbuilding plan will be in jeopardy. The Navy outlined its strategy for
achieving a 313-ship force in its updated long-range shipbuilding plan.
Over the next 5 years, the Navy plans to significantly increase the rate
of construction and introduce nine new classes of ships, including the
Ford-class aircraft carrier (CVN 78) and the Zumwalt-class destroyer (DDG
1000). To support the plan, the Navy will require shipbuilding funding
significantly above current levels--on the order of $5 billion more by
fiscal year 2013. The Navy recognizes that the success of the plan will
depend on its ability to realistically estimate and control shipbuilding
costs. Over the next year the Navy will begin construction of CVN 78, DDG
1000, and LHA 6 amphibious assault ship. The Navy estimates that these
ships alone will require nearly $20 billion in construction funding,
representing the Navy's most costly lead ships. Even a small percentage of
cost growth on the big ships could lead to the need for hundreds of
millions of dollars in additional funding.
Shipbuilding Programs Often Have Unexecutable Business Cases
Navy shipbuilding programs are often framed around an unexecutable
business case, whereby ship designs seek to accommodate immature
technologies, design stability is not achieved until late in production,
and both cost and schedule estimates are unrealistically low. This
situation has recently been evidenced in the LPD 17 and LCS programs,
which have required costly out-of-sequence work during construction. The
DDG 1000 and CVN 78 programs are at risk because of lingering technology
immaturity, coupled with cost and schedule estimates with little margin
for error.
Elements of a Business Case
We have frequently reported on the wisdom of using a solid, executable
business case before committing resources to a new product development
effort. In its simplest form, a business case requires a balance between
the concept selected to satisfy warfighter needs and the
resources--technologies, design knowledge, funding, time, and management
capacity--needed to transform the concept into a product, in this case a
ship. At the heart of a business case is a knowledge-based approach that
requires that managers demonstrate high levels of knowledge as the program
proceeds from technology development to system development and, finally,
production. Adapting this approach to shipbuilding is a challenge, as I
will discuss later. Ideally, in such an approach, key technologies are
demonstrated before development begins. The design is stabilized before
the building of prototypes or, in the case of ships, construction begins.
At each decision point, the balance among time, money, and capacity is
validated. In essence, knowledge supplants risk over time.
A sound business case would establish and resource a knowledge-based
approach at the outset of a program. We would define such a business case
as firm requirements, mature technologies, and an acquisition strategy
that provides sufficient time and money for design activities before
construction start. The business case is the essential first step in any
acquisition program that sets the stage for the remaining stages of a
program, namely the business or contracting arrangements and actual
execution or performance. If the business case is not sound, the contract
will not correct the problem and execution will be subpar. This does not
mean that all potential problems can be eliminated and perfection
achieved, but rather that sound business cases can get the Navy better
shipbuilding outcomes and better return on investment. If any one element
of the business case is weak, problems can be expected in construction.
The need to meet schedule is one of the main reasons why programs cannot
execute their business cases. This pattern was clearly evident in both the
LPD 17 and LCS programs. In both cases, the program pushed ahead with
production even when design problems arose or key equipment was not
available when needed. Short cuts, such as doing technology development
concurrently with design and construction, are taken to meet schedule. In
the end, problems occur that cannot be resolved within compressed,
optimistic schedules. Ultimately, when a schedule is set that cannot
accommodate program scope, delivering an initial capability is delayed and
higher costs are incurred.
In shipbuilding programs, the consequences of moving forward with immature
technologies or an unstable design become clear once ship construction
begins. Ships are designed and constructed with an optimal sequence--that
is, the most cost-efficient sequence to construct the ship. This includes
designing and building the ship from the bottom up and maximizing the
units completed in shipyard shops and installed in the dry dock while
minimizing tasks performed when the ship is already in the water, which
tend to be costlier than tasks on land. Once units are installed access to
lower decks of the ship becomes more difficult. If equipment is not ready
in time for installation, the shipbuilder will have to work around the
missing equipment. Additional labor hours may be needed because spaces
will be less accessible and equipment may require more time for
installation.
Business Cases Deteriorated with Construction of LPD 17 and Littoral Combat
Ships
What happens when the elements of a solid business case are not present?
Unfortunately, the results have been all too visible in the LPD 17 and the
LCS. Ship construction in these programs has been hampered throughout by
design instability and program management challenges that can be traced
back to flawed business cases. The Navy moved forward with ambitious
schedules for constructing LPD 17 and LCS despite significant challenges
in stabilizing the designs for these ships. As a result, construction work
has been performed out of sequence and significant rework has been
required, disrupting the optimal construction sequence and application of
lessons learned for follow-on vessels in these programs.
In the LPD 17 program, the Navy's reliance on an immature design tool led
to problems that affected all aspects of the lead ship's design. Without a
stable design, work was often delayed from early in the building cycle to
later, during integration of the hull. Shipbuilders stated that doing the
work at this stage could cost up to five times the original cost. The lead
ship in the LPD class was delivered to the warfighter incomplete and with
numerous mechanical failures, resulting in a lower than promised level of
capability. These problems continue today--2 years after the Navy accepted
delivery of LPD 17. Recent sea trials of the ship revealed problems with
LPD 17's steering system, reverse osmosis units, shipwide area computing
network, and electrical system, among other deficiencies. Navy inspectors
noted that 138 of 943 ship spaces remained unfinished and identified a
number of safety concerns related to personnel, equipment, ammunition,
navigation, and flight activities. To date, the Navy has invested over
$1.75 billion constructing LPD 17.
In the LCS program, design instability resulted from a flawed business
case as well as changes to Navy requirements. From the outset, the Navy
sought to concurrently design and construct two lead ships in the LCS
program in an effort to rapidly meet pressing needs in the mine
countermeasures, antisubmarine warfare, and surface warfare mission areas.
The Navy believed it could manage this approach, even with little margin
for error, because it considered each LCS to be an adaptation of an
existing high-speed ferry design. It has since been realized that
transforming a high-speed ferry into a capable, networked, survivable
warship was quite a complex venture. Implementation of new Naval Vessel
Rules (design guidelines) further complicated the Navy's concurrent
design-build strategy for LCS. These rules required program officials to
redesign major elements of each LCS design to meet enhanced survivability
requirements, even after construction had begun on the first ship. While
these requirements changes improved the robustness of LCS designs, they
contributed to out of sequence work and rework on the lead ships. The Navy
failed to fully account for these changes when establishing its $220
million cost target and 2-year construction cycle for the lead ships.
Complicating LCS construction was a compressed and aggressive schedule.
When design standards were clarified with the issuance of Naval Vessel
Rules and major equipment deliveries were delayed (e.g., main reduction
gears), adjustments to the schedule were not made. Instead, with the first
LCS, the Navy and shipbuilder continued to focus on achieving the planned
schedule, accepting the higher costs associated with out of sequence work
and rework. This approach enabled the Navy to achieve its planned launch
date for the first Littoral Combat Ship, but required it to sacrifice its
desired level of outfitting. Program officials report that schedule
pressures also drove low outfitting levels on the second Littoral Combat
Ship design as well, although rework requirements have been less intensive
to date. However, because remaining work on the first two ships will now
have to be completed out-of-sequence, the initial schedule gains most
likely will be offset by increased labor hours to finish these ships.
The difficulties and costs discussed above relate to the LCS seaframe
only. This program is unique in that the ship's mission equipment is being
developed and funded separately from the seaframe. The Navy faces
additional challenges integrating mission packages with the ships, which
could further increase costs and delay delivery of new antisubmarine
warfare, mine countermeasures, and surface warfare capabilities to the
fleet. These mission packages are required to meet a weight requirement of
180 metric tons or less and require 35 personnel or less to operate
them.^c However, the Navy estimates that the mine countermeasures mission
package may require an additional 13 metric tons of weight and 7 more
operator personnel in order to deploy the full level of promised
capability. Because neither of the competing ship designs can accommodate
these increases, the Navy may be forced to reevaluate its planned
capabilities for LCS.
^cLCS mission packages include combat systems, support equipment,
computing environment, and mission crew. The mission package weight
requirement of 180 metric tons or less also includes aviation fuel, and
the manning requirement of 35 or less includes personnel comprising an
aviation detachment.
DDG 1000 and CVN 78 Have More Thoughtful Business Cases, but Significant
Technical Risks Remain
Elements of a successful business case are present in the Navy's
next-generation shipbuilding programs--CVN 78 and DDG 1000. The Navy's
plans for these programs call for a better understanding of the designs of
these ships prior to beginning construction, thereby reducing the risk of
costly design changes after steel has been bent and bulkheads built. Yet
some elements of their business cases put execution within budgeted
resources at risk. While the Navy has recognized the need to mature each
ship's design before beginning construction, CVN 78 and DDG 1000 remain at
risk of cost growth due to continuing efforts to mature technologies.
Success in these programs depends on on-time delivery and installation of
fully mature and operational technologies in order to manage construction
costs. Budgets and schedules for each ship leave little if any margin for
error.
DDG 1000
The DDG 1000 development has been framed by challenging multi-mission
requirements, resultant numerous technologies and a tight construction
schedule driven by industrial base needs. In the DDG 1000 program, the
Navy estimates that approximately 75 percent of detail design will be
completed prior to the start of lead ship construction in July 2008.
Successfully meeting this target, however, depends on maturing 12
technologies as planned. Currently, three of these technologies are fully
mature. Two DDG 1000 technologies--the volume search radar and total ship
computing environment--have only completed component-level demonstrations
and subsequently remain at lower levels of maturity. Schedule constraints
have also forced the Navy to modify its test plans for the integrated
power system and external communication systems.
The volume search radar, one of two radars in the dual band radar system,
will not have demonstrated the power output needed to meet requirements
even after integrated land-based testing of the prototype radar system is
completed in 2009. Production of the radars, however, is scheduled to
begin in 2008, introducing additional risk if problems are discovered
during testing. According to Navy officials, in the event the volume
search radar experiences delay in testing, it will not be integrated as
part of the dual band radar into the DDG 1000 deckhouse units that will be
delivered to the shipbuilders. Instead, the Navy will have to task the
shipbuilder with installing the volume search radar into the deckhouse,
which program officials report will require more labor hours than
currently allocated. The DDG 1000 program's experience with the dual band
radar has added significance as the same radar will be used on CVN 78.
In the case of the DDG 1000 total ship computing environment, the Navy is
developing hardware infrastructure and writing and releasing six blocks of
software code. Although development of the first three software blocks
progressed in line with cost and schedule estimates, the Navy has been
forced to defer some of the functionalities planned in software release
four to software blocks five and six due to changes in availability of key
subsystems developed external to the program, introduction of non
development items, and changes in program integration and test needs. The
Navy now plans to fully mature the integrated system following ship
construction start--an approach that increases program exposure to cost
and schedule risk in production.
The DDG 1000 program also faces challenges completing testing for its
integrated power system and external communications systems. Currently,
shipbuilder-required equipment delivery dates for these systems do not
permit time for system-level land-based integration testing prior to
delivery. As a result, the Navy has requested funds in fiscal year 2008
for the third shipset of this equipment so that testing can be completed
without interrupting the planned construction schedules of the first two
ships. However, in the event problems are discovered, DDG 1000
construction plans and costs could be at risk.
CVN 78
The Navy has completed the basic design of CVN 78, and the shipbuilder is
currently developing the carrier's more detailed design. According to the
shipbuilder, about 70 percent of CVN 78's design is complete, with almost
all of the very low decks of the ship completely designed. Progress in
designing CVN 78 is partially the result of a longer preparation period
that has enabled the shipbuilder to design more of the ship prior to
construction than was the case on previous carriers. However, the Navy may
face challenges in maintaining its design schedule because of delays in
the development of the ship's critical technologies. Such delays could
impede the completion of the ship's design and interfere with the
construction of the ship.
CVN 78 will feature an array of advanced technologies such as a new
nuclear propulsion and electric plant, an electromagnetic aircraft launch
system (EMALS) and an improved aircraft arresting system. These
technologies, along with an expanded and improved flight deck, are
designed to significantly improve performance that the Navy believes will
simultaneously reduce acquisition and life cycle costs compared to
previous carriers. The Navy has focused much attention on developing
technologies and has retired much risk. Yet risk remains. The schedule for
installing CVN 78's technologies takes advantage of construction
efficiencies. The shipbuilder has identified key dates when technologies
need to be delivered to the yard in order to meet its optimal construction
schedule. A number of CVN 78's technologies have an increased potential to
affect this schedule because they are (1) located low in the ship and
needed early in construction or (2) highly integrated or embedded in the
ship's design. The dual band radar is integrated into the design of the
carrier's island and is critical to the smaller island design. EMALS
crosses 48 of the ship's 423 zones (or separate units that make up the
ship's design). For example, problems with EMALS could have a cascading
effect on other areas of the ship.
While the Navy has mitigated the risk posed by some technologies, like the
nuclear propulsion and electric plant, key systems, in particular, EMALS,
the advanced arresting gear, and the dual band radar have encountered
difficulties during development that will likely prevent timely delivery
to the shipyard. Challenges include the following:
o EMALS encountered technical difficulties developing the
prototype generator and meeting detailed Navy requirements, which
led to increased program costs and an over 15-month schedule
delay. To meet shipyard dates for delivering equipment, the
contractor eliminated all schedule margin, normally reserved for
addressing unknown issues. Yet, significant challenges lay
ahead--the Navy will begin testing a production representative
system in 2008, and the shipboard system will be manufactured in a
new facility inexperienced with production. If problems occur in
testing or production, the contractor will not be able to meet its
delivery date to the shipyard, causing work to be done out of
sequence.
o The advanced arresting gear program faced difficulties
delivering drawings to the Navy, leading to program delays.
Schedule delays have slipped the production decision and delivery
to CVN 78 by 6 months. In an effort to maintain shipyard delivery
dates, the Navy has consolidated upcoming test events--increasing
test cycles and eliminating schedule margin. However, by
compressing test events, the Navy will have little time to address
any problems prior to production start. Late delivery of the
advanced arresting gear will require installation after the flight
deck has been laid. The shipbuilder will expend additional labor
to lower the system into place through a hole cut in the flight
deck.
o The dual band radar presents the most immediate risk to the DDG
1000 program, but delays in production could cascade down to CVN
78--affecting delivery to the shipyard. Moreover, upcoming
land-based testing will not include certain demonstrations of
carrier-specific performance. In particular, the Navy has not yet
scheduled tests to verify the radar's air traffic control
capability, but expects such demonstrations will occur by the end
of fiscal year 2012. This leaves little to no time to make any
necessary changes before the radar's 2012 in-yard date.
The CVN 78 business case also faces risks in the area of cost because the
estimate that underpins the budget is optimistic. For example, the Navy
estimates that fewer labor hours will be needed to construct CVN 78 than
the previous two carriers--even though it is a lead ship that includes
cutting edge technologies and a new design. Although the Navy is working
with the shipbuilder now to reduce costs prior to the award of a
construction contract (scheduled for early next year) through such
measures as subsidizing capital expenditures to gain greater shipyard
efficiency, costs will likely exceed budget if technologies or other
materials are delivered late or labor hour efficiencies are not realized.
A Disciplined, Knowledge-Based Process Is Key to Better Outcomes
How can the Navy achieve better outcomes in its shipbuilding programs? Our
work on best practices highlights the need for a disciplined,
knowledge-based approach so that programs proceed with a high probability
of success. This means technology maturity must be proven before a design
can be considered stable, and production outcomes cannot be guaranteed
until a stable design is demonstrated. The challenge in adapting such an
approach to shipbuilding is determining when these levels of knowledge
should be reached in shipbuilding programs and what standards should serve
as criteria for demonstrating this knowledge. It seems that no two
shipbuilding programs are run the same way. For example, it can be agreed
that key aspects of a ship's detail design must be completed before
construction begins. However, what those aspects are or how they should be
measured is not defined. What may be acceptable in one shipbuilding
program is not acceptable in another. In our reviews of ship programs, the
definition of phases, strategies, and decision points varies from program
to program. In addition to defining key junctures of knowledge, standards,
and corresponding decision points for shipbuilding programs, there are
other steps the Navy could take that would better inform its acquisition
decision making in shipbuilding programs. These include:
o ensuring that initial shipbuilding budgets are realistically
achievable by improving cost estimating, and
o improving cost management through increased use of fixed-price
contracting and comprehensive cost surveillance.
Aligning Knowledge and Decision Points Consistently across Shipbuilding Programs
Each shipbuilding program seems to embody its own strategy for making
decisions. In programs other than shipbuilding, the Milestone B decision
represents the commitment to design and develop a system, at which time
requirements should be firm and critical technologies mature. Milestone B
means different things in different shipbuilding programs. The CVN 21
program held its Milestone B review shortly before a preliminary design
review, and 3 years before the planned approval for the construction
contract. The Milestone B review for DDG 1000--called DD(X) at the
time--occurred over 1 year after the preliminary design review and shortly
after the critical design review--it was used to authorize negotiation of
a construction contract. The LCS program has received authorization for
construction for six ships--it has yet to hold a Milestone B review.
The need for a common understanding of what Milestone B represents is all
the more important given the requirements for certification at Milestone B
enacted by Congress in 2006.^d These provisions require the decision
authority to certify that, among other things,
o the technology has been demonstrated in a relevant environment,
o requirements have been approved by the Joint Requirements
Oversight Council,
o the program is affordable, and
o the program demonstrates a high likelihood of accomplishing its
intended mission.
We believe that the certification for all shipbuilding programs should
take place at the same point. The uniqueness of individual program
strategies leads to similar challenges in trying to establish what level
of knowledge is needed at subsequent critical junctures in shipbuilding
programs. Each shipbuilding program seems to have a different measure as
to how much of the design needs to be complete--and what constitutes
design readiness--prior to beginning ship construction. It seems to us
that there should be clear metrics for what the Navy expects at key
junctures across all shipbuilding programs. Further, there appears to be
few criteria across shipbuilding programs regarding how much
knowledge--such as the percentage of ship units built--is needed at
different decision points, including keel lay, fabrication start, and ship
launch. A clearer understanding of the key knowledge junctures and
corresponding criteria across shipbuilding programs would help establish a
better basis for cost and schedule estimates.
^dNational Defense Authorization Act for Fiscal Year 2006, Pub. L. No.
109-163, S 801; 10 U.S.C. S 2366a.
Establishing Executable Program Budgets through Improved Cost Estimating
As we have seen, the Navy's track record for achieving its initial budgets
for shipbuilding programs has not been good. If we expect programs to be
executed within budget, programs need to begin with realistic budgets.
Since ship construction is generally budgeted in 1 fiscal year--or in the
case of CVN 78 and DDG 1000--over 2 years, it is essential that the Navy
understand and plan for the likely costs of the ship when construction is
authorized. A ship's initial budget will, in large part, determine whether
and how much cost growth will occur and require funding in later years.
The foundation of an executable budget is a realistic cost estimate that
takes into account the true risk and uncertainty in a program. Realistic
cost estimates are important not only because they are used to establish
program budgets, but also because they help enable the Navy to determine
priorities, including whether to proceed with a program. Our past work has
shown that the Navy tends to underestimate the costs needed to construct
ships--resulting in unrealistic budgets and large cost increases after
ship construction has begun. Initial estimates of LPD 17 and LCS 1 assumed
significant savings based on efficiencies that did not materialize as
expected. Future ships like CVN 78 make similar assumptions.
One way to improve the cost-estimating process is to present a confidence
level for each estimate, based on risk and uncertainty analyses. By
conducting an uncertainty analysis that measures the probability of cost
growth, the Navy can identify a level of confidence for its estimates and
determine whether program costs are realistically achievable. Navy cost
analysts told us that they used quantitative risk analyses to test the
validity of cost estimates of CVN 78 and DDG 1000. We believe that the
Navy and the Department of Defense (DOD) should take this a step
further--requiring a high confidence level threshold when making program
commitments and budget requests. The Defense Acquisition Performance
Assessment Panel recommended an 80 percent confidence level, meaning that
a program has an 80 percent chance of achieving its estimated costs.^e
Whether this is the right level warrants thoughtful discussion, but it is
worth noting that analyses for CVN 78 and DDG 1000 were well below an 80
percent confidence level (in the case of DDG 1000 at around 45
percent)--increasing the likelihood that costs will grow above budget.
^eDefense Acquisition Performance Assessment Panel, Defense Acquisition
Performance Assessment Report (Washington, D.C., January 2006).
Timing is also an important element for achieving realism in budgets. In
the past, the Navy has generally requested approval for detail design and
construction of the lead ship at the same time. As a result, construction
budgets did not benefit from the knowledge gained in system design or
early stages of detail design. An alternative approach is to separate the
decision to fund detail design from the decision to fund construction. The
benefits of this approach are evident in the funding of DDG 1000. The Navy
first requested funding for detail design and construction of the lead
ship in its fiscal year 2005 budget request, estimating these costs to
total $2.7 billion. Congress did not fund construction of the lead ship,
but instead funded detail design and purchase of some materials in the
fiscal years 2005 and 2006 budgets. In March 2005, the Navy completed a
life-cycle cost estimate for the ship that placed the cost of DDG 1000 at
$3.3 billion. DOD independent cost analysts estimated even higher costs.
The budget request for fiscal years 2007 and 2008 included $3.3 billion
for each of the two lead ships, reflecting an improved understanding of
budget requirements compared to the initial fiscal year 2005 request.
Better Management of Costs through Fixed-Price Contracting and Comprehensive
Cost Surveillance
The Navy can take other steps to improve the outcomes of its shipbuilding
programs by strengthening its cost management capability, including
o greater use of fixed-price contracting and
o enhanced and comprehensive cost surveillance
Fixed-priced-Contracting for Construction
In an effort to improve cost management, the Navy is promoting
fixed-priced contracts for ship construction. In a fixed-price incentive
contract, costs above a target are shared with the contractor, up to a
ceiling price. Both the target cost and price ceiling are negotiated at
the outset. The contractor is responsible for costs above the ceiling
price, limiting the government's cost risk. In shipbuilding, lead ships
are commonly done under cost-plus-incentive-fee contracts as are some
follow-on ships. Under these contracts, the government is responsible for
paying allowable costs incurred and the fee will be adjusted according to
a negotiated formula. The first five LPD 17 ships use
cost-plus-incentive-fee contracts and the first two LCS are being built
under cost-type contracts. The Navy typically uses fixed-priced incentive
contracts for ships that are later in the class, including DDG 51 class
destroyers and CVN 77, and for all ships in the T-AKE class of auxiliary
ships, a less complex ship.
We are encouraged by the Navy's efforts to move to fixed-price contracts.
Fixed-price contracts limit the government's risk of cost growth while
encouraging realism in negotiating contract prices and careful cost
management. However, the move to fixed-price contracting is feasible only
if risks can be understood and managed. If the Navy is to use fixed-price
contracts for the second or third ship in the class--or even the lead
ship--it must supplant risk with knowledge. We are convinced that a move
to fixed-price contracting will only succeed if the Navy adopts a more
disciplined process, one that ensures that the elements of an executable
business case exist as the development effort begins. If technologies are
still being demonstrated, the delivery of critical systems when needed
cannot be assured. Nor can designs be finalized. Increased used of
fixed-price contracting requires that technologies be demonstrated early,
the design stabilized before construction begins, and realistic estimates
for cost and schedule made.
Cost Surveillance
Given the risk of cost growth in shipbuilding, it is equally important
that the Navy strengthen its oversight of shipyard cost performance. Our
work has shown that the Navy may not have adequate management tools
necessary to identify and react to early signs of cost growth. In
particular, in the CVN 78 program the Navy has not effectively used earned
value management data captured in cost performance reports submitted by
the contractor. Earned value management is a tool that provides the
government and contractors with insight into technical, cost, and schedule
progress on their contracts. Although the shipbuilder is designing much of
CVN 78 prior to the award of the construction contract, we found that
contractor cost performance reports do not provide an objective measure of
program schedule and costs incurred. While the Navy expects that future
cost performance reports will better reflect shipyard performance after
the construction contract is awarded and significant construction work is
under way, it has missed an opportunity to gain insight into current
costs--and gauge future shipyard performance. Moreover, the Navy may not
require the shipbuilder to submit monthly cost performance reports that
include variance analyses, which describe the reasons for cost and
schedule variances. Without monthly contractor performance reports that
include variance analyses, the Navy will miss timely information regarding
root causes for cost and schedule problems and mitigation efforts--making
it more difficult to identify risk and take corrective action.
But timely and complete cost performance reports are not enough. The Navy
must leverage this information to better manage shipbuilder performance.
In particular, the Navy's Supervisor of Shipbuilding (SUPSHIP) is not
engaged in evaluating shipbuilder cost performance for all shipbuilding
programs. SUPSHIP provides the Navy with unique insight into program
performance because it is located at the shipyard, providing on-site
program surveillance, including independent analysis of shipbuilder cost
and schedule performance. However, SUPSHIP does not currently have the
capability to conduct independent cost surveillance of the CVN 78 program.
We believe that this capability is necessary to effectively analyze
shipbuilder cost data and verify that the data depict actual conditions
and trends.
Cost surveillance at the shipyard is just one element of the management
capacity needed to plan and execute shipbuilding programs. There has also
been considerable discussion of the need to have a workforce with the
right skills in the right numbers. It has been more difficult to arrive at
a firm definition of the size and composition of the workforce needed, and
the appropriate balance between government and contractor personnel. Sharp
declines in the size of the acquisition workforce have occurred over the
last several years. The Navy's numbers are a case in point. The Navy
reports that staffing at Naval Sea Systems Command headquarters has
decreased by almost 50 percent since 1991, from 4,871 to 2,331 personnel.
Mr. Chairman, that concludes my statement. I would be pleased to answer
any questions.
Objectives, Scope, and Methodology
To develop information on the status of Navy shipbuilding programs and
practices that can improve the process for acquiring ships, we relied
largely on our prior reporting on shipbuilding programs and updates to
this work, as well as work under way for the committee on the CVN 78
program. In the course of this work, we analyzed program documents,
including program baselines, contractor performance reports, cost
estimates, budget documents and other program assessments, as well as
policy guidance. We also interviewed government, shipbuilding, and other
contractor officials associated with a number of shipbuilding programs,
including CVN 77 and 78, LPD 17, DDG 1000, LCS, and Virginia-class
Submarines.
Contact and Staff Acknowledgments
For future questions about this statement, please contact me at (202)
512-4841. Individuals making key contributions to this statement include
Lisa L. Berardi, Noah B. Bleicher, Gwyneth M. Blevins, Lily J. Chin, Todd
Dice, Christopher R. Durbin, Jennifer Echard, Diana Moldafsky, Moshe
Schwartz, and Karen Zuckerstein.
Related GAO Products
Defense Acquisitions: Progress and Challenges Facing the DD(X) Surface
Combatant Program. [24]GAO-05-924T , Washington, D.C.: July 19, 2005.
Defense Acquisitions: Assessments of Selected Weapon Programs.
[25]GAO-07-406SP , Washington, D.C.: March 30, 2007.
Defense Acquisitions: Challenges Associated with the Navy's Long-Range
Shipbuilding Plan. [26]GAO-06-587T , Washington, D.C.: March 30, 2006.
Defense Acquisitions: Challenges Remain in Developing Capabilities in
Naval Surface Fire Support. [27]GAO-07-115 , Washington, D.C.: November
30, 2006.
Defense Acquisitions: Improved Management Practices Could Help Minimize
Cost Growth in Navy Shipbuilding Programs. [28]GAO-05-183 , Washington,
D.C.: Feb. 28, 2005.
Defense Acquisitions: Plans Need to Allow Enough Time to Demonstrate
Capability of First Littoral Combat Ships. [29]GAO-05-255 , Washington,
D.C.: March 1, 2005.
Appendix I: Cost Growth for Individual Ships
Table 2: Cost Growth in Program Budgets
Dollars in millions (figures may not add due to rounding)
Cost
growth
as a
Fiscal year Cost growth percent
2008 Total Cost growth due to Navy- of
Initial President's cost due to furnished initial
Ship Budget budget^a growth construction equipment budget
CVN 77 $4,975 $5,822 $847 $771 $76 17%
DDG 100 938 1,066 128 142 (13) 14
DDG 101 935 984 49 62 (13) 5
DDG 102 1,016 1,097 80 126 (46) 8
DDG 103 1,107 1,117 10 56 (46) 1
DDG 104 1,062 1,113 51 97 (46) 5
DDG 105 1,184 1,207 23 42 (20) 2
DDG 106 1,233 1,240 7 27 (20) 1
DDG 107 1,089 1,093 4 21 (17) 0
DDG 108 1,102 1,103 1 18 (17) 0
DDG 109 1,138 1,142 4 21 (17) 0
DDG 3,505 3,517 12 29 (17) 0
110-112
LCS 1-2 472 1,075^b 603 NA NA 128
LHD 8 1,893 2,196 303 320 (17) 16
LPD 18 762 1,272 510 531 (22) 67
LPD 19 1,064 1,286 222 228 (6) 21
LPD 20 890 1,137 247 311 (64) 28
LPD 21 1,113 1,287 173 283 (110) 16
LPD 22 1,256 1,403 147 287 (140) 12
LPD 23 1,108 1,357 249 337 (88) 22
SSN 775 2,192 2,740 548 546 1 25
SSN 776 2,020 2,183 164 154 9 8
SSN 777 2,276 2,332 56 65 (9) 2
SSN 778 2,192 2,242 50 246 (196) 2
SSN 779 2,152 2,255 102 283 (180) 5
SSN 780 2,245 2,289 44 41 3 2
SSN 781 2,402 2,378 (24) (24) (0) -1
SSN 782 2,612 2,604 (7) (7) 0 0
SSN 783 2,654 2,654 - - - 0
T-AKE 1 489 538 49 44 6 10
T-AKE 2 358 370 12 9 3 3
T-AKE 3 361 335 (26) (25) (1) -7
T-AKE 4 370 342 (28) (32) 4 -8
T-AKE 5/6 683 702 19 20 (1) 3
T-AKE 7/8 713 712 (1) 4 (4) 0
T-AKE 9 380 386 6 9 (3) 2
Total $ 51,941 $ 56,578 $ 4,637 9%
Source: GAO analysis of Navy data.
aIncludes reprogramming actions and requests through June 2007.
bA small amount of these funds may be designated for certain LCS research
and development activities.
(120633)
This is a work of the U.S. government and is not subject to copyright
protection in the United States. The published product 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.
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.
Obtaining Copies of GAO Reports and Testimony
The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site ( [30]www.gao.gov ). Each weekday, GAO posts
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
[31]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, Waste, and Abuse in Federal Programs
Contact:
Web site: [32]www.gao.gov/fraudnet/fraudnet.htm
E-mail: [33][email protected]
Automated answering system: (800) 424-5454 or (202) 512-7470
Congressional Relations
Gloria Jarmon, Managing Director, [34][email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Washington,
D.C. 20548
Public Affairs
Paul Anderson, Managing Director, [35][email protected] (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
[36]www.gao.gov/cgi-bin/getrpt?GAO-07-943T .
To view the full product, including the scope
and methodology, click on the link above.
For more information, contact Paul L. Francis at (202) 512-4841 or
[email protected].
Highlights of [37]GAO-07-943T , a testimony before the Subcommittee on
Seapower and Expeditionary Forces, Committee on Armed Services, House of
Representatives
July 2007
DEFENSE ACQUISITIONS
Realistic Business Cases Needed to Execute Navy Shipbuilding Programs
The Navy is beset with long-standing problems that affect its ability to
accomplish ambitious goals for its shipbuilding portfolio. Significant
cost growth and long schedule delays are persistent problems. Making
headway on these problems is essential in light of the serious budget
pressures facing the nation.
This testimony focuses on (1) cost growth in shipbuilding, (2) acquisition
approaches in the LPD 17, Littoral Combat Ship, DDG 1000 and CVN 78
programs and (3) steps the Navy can take to improve its acquisition
decision-making, particularly the adoption of a knowledge-based framework.
[38]What GAO Recommends
While GAO is making no new recommendations in this testimony, GAO has made
numerous recommendations through the years to improve business cases for
Navy acquisitions as well as other Department of Defense weapon
acquisitions. The Department's acquisition policies largely incorporate
these recommendations, but they have not been implemented on actual
programs.
The Navy has exceeded its original budget by more than $4 billion for the
41 ships under construction at the beginning of this fiscal year. And more
cost growth is coming. Cost growth is not just a problem for lead ships of
a new class but also for follow-on ships. For example, costs for the first
two Littoral Combat Ships have more than doubled. Similarly, costs for the
first two San Antonio class (LPD 17 and LPD 18) amphibious ships have
increased by over $1.3 billion--almost a 77 percent increase above the
initial budgets. Cost growth of this magnitude leads to lost opportunities
for tomorrow's needs.
These types of problems point to the wisdom of using solid, executable
business cases to design and build ships. A business case requires a
balance between the concept selected to satisfy warfighter needs and the
resources--technologies, design knowledge, funding, time, and management
capacity--needed to transform the concept into a product, in this case a
ship. Neither LPD 17 nor the Littoral Combat Ship programs was framed
around an executable business case; rather, the programs pushed ahead
without a stable design and without realistic cost estimates, resulting in
higher costs, schedule delays, and quality problems. The Navy has a more
thoughtful business case for its next generation aircraft carrier and
destroyer programs (CVN 78 and DDG 1000, respectively) before
construction, but the programs remain at risk for cost growth partly
because of continuing efforts to mature technologies. GAO's work on best
practices highlights the need for a disciplined, knowledge-based approach
to help shipbuilding, and other defense acquisition programs achieve more
successful outcomes. This approach is predicated on certain essentials,
including:
o ensuring that technology maturity is proven before a design is
considered stable and understanding that production outcomes
cannot be guaranteed until a stable design is demonstrated;
o improving cost estimating to develop initial shipbuilding
budgets that are realistically achievable; and
o improving cost management through increased use of fixed-price
contracting and comprehensive cost surveillance.
A significant challenge to adapting a knowledge-based approach is the lack
of a common understanding across programs regarding the definition,
timing, and criteria for key knowledge junctures. For example, each
shipbuilding program seems to have a different measure as to how much of
the design needs to be complete before beginning ship construction.
Similarly, there appears to be little criteria across programs regarding
how much knowledge--such as the percent of ship units built--is needed at
different decision points, including keel lay, fabrication start, and ship
launch.
References
Visible links
24. http://www.gao.gov/cgi-bin/getrpt?GAO-05-924T
25. http://www.gao.gov/cgi-bin/getrpt?GAO-07-406SP
26. http://www.gao.gov/cgi-bin/getrpt?GAO-06-587T
27. http://www.gao.gov/cgi-bin/getrpt?GAO-07-115
28. http://www.gao.gov/cgi-bin/getrpt?GAO-05-183
29. http://www.gao.gov/cgi-bin/getrpt?GAO-05-255
30. http://www.gao.gov/
31. http://www.gao.gov/
32. http://www.gao.gov/fraudnet/fraudnet.htm
33. mailto:[email protected]
34. mailto:[email protected]
35. mailto:[email protected]
36. http://www.gao.gov/cgi-bin/getrpt?GAO-07-943T
37. http://www.gao.gov/cgi-bin/getrpt?GAO-07-943T
*** End of document. ***