Defense Acquisitions
Cost to Deliver Zumwalt-Class Destroyers Likely to Exceed Budget Gao ID: GAO-08-804 July 31, 2008In October 2008, the U.S. Navy will begin construction of the first of two lead DDG 1000 Zumwalt-class destroyers--at an expected cost of $6.3 billion. Given the history of cost growth on shipbuilding programs, as well as the Navy's request for approval of a third ship, GAO was asked to assess the progress of the program. GAO examined (1) whether key systems can be delivered on time and work as intended (2) design maturity (3) shipyard readiness and (4) whether lead and follow-on DDG 1000 ships can be built within budget. To accomplish this, our work included analysis of schedules, ship progress reviews and cost estimates; interviews with Navy and other officials; and our own past work.
From the outset, DDG 1000 has faced a steep challenge framed by technical sophistication, demanding mission requirements, and a somewhat unforgiving schedule. The Navy conceived a thoughtful strategy to meet these demands and has had success with several technologies and its design approach. Yet, the program did not proceed as planned and the Navy recently realigned the construction schedule for the first two ships to provide more time to finish key systems and software. Still, the Navy will produce--and in some cases install--key systems on the ship before fully demonstrating and testing them. Software development has proven challenging; the Navy certified the most recent software release before it met about half of its requirements. At this point--the first year of a 6-year construction schedule for the lead ships--the Navy may have exhausted its options for solving future problems without adding money and time. The Navy expects to achieve a greater degree of design maturity before starting construction than has been the case on previous surface combatant programs. To meet this goal, the Navy will be pressed to complete a large amount of design work by October 2008 when construction will begin. From August 2007 through May 2008, the shipbuilders finished work on 16 of the 100 design "zones" that make up the ship, leaving 84 zones to finish the final design phases in the 5 months leading up to the start of construction. Both shipyards that will build DDG 1000 are preparing for construction through facility enhancements and production improvements. However, uncertainty remains. The ship's deckhouse will be built primarily from composite rather than steel and the shipbuilder is still refining the process for large scale composite manufacturing and assembly. Workforce instability could also prevent shipbuilders from fully realizing expected efficiencies. The full costs of constructing the two lead ships have not been entirely recognized or funded. The risk of cost growth is high in part because of the potential for late delivery of key systems and software and difficulties in constructing and integrating sections of the ship, like the deckhouse. Remaining funds may not be sufficient to buy key components and pay for other work not yet under contract. The Navy has already requested funding for a third ship and plans to contract for this ship with options for four more ships in fiscal year 2009. The Navy will not have enough data then on the actual costs of the lead ships to develop realistic prices for follow-on ships. As currently planned, all ships will be under contract and all but one under construction before the Department of Defense holds the production milestone review in 2013.
RecommendationsOur recommendations from this work are listed below with a Contact for more information. Status will change from "In process" to "Open," "Closed - implemented," or "Closed - not implemented" based on our follow up work.
Director: Team: Phone:GAO-08-804, Defense Acquisitions: Cost to Deliver Zumwalt-Class Destroyers Likely to Exceed Budget
This is the accessible text file for GAO report number GAO-08-804
entitled 'Defense Acquisitions: Cost to Deliver Zumwalt-Class
Destroyers Likely to Exceed Budget' which was released on July 31, 2008.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
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. 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.
Report to the Subcommittee on Seapower, Committee on Armed Services,
U.S. Senate:
United States Government Accountability Office:
GAO:
July 2008:
Defense Acquisitions:
Cost to Deliver Zumwalt-Class Destroyers Likely to Exceed Budget:
GAO-08-804:
GAO Highlights:
Highlights of GAO-08-804, a report to the Subcommittee on Seapower,
Committee on Armed Services, U.S. Senate.
Why GAO Did This Study:
In October 2008, the U.S. Navy will begin construction of the first of
two lead DDG 1000 Zumwalt-class destroyers”at an expected cost of $6.3
billion. Given the history of cost growth on shipbuilding programs, as
well as the Navy‘s request for approval of a third ship, GAO was asked
to assess the progress of the program. GAO examined (1) whether key
systems can be delivered on time and work as intended (2) design
maturity (3) shipyard readiness and (4) whether lead and follow-on DDG
1000 ships can be built within budget. To accomplish this, GAO‘s work
includes analysis of schedules, ship progress reviews and cost
estimates; and interviews with Navy and other officials.
What GAO Found:
From the outset, DDG 1000 has faced a steep challenge framed by
technical sophistication, demanding mission requirements, and a
somewhat unforgiving schedule. The Navy conceived a thoughtful strategy
to meet these demands and has had success with several technologies and
its design approach. Yet the program did not proceed as planned and the
Navy recently realigned the construction schedule for the first two
ships to provide more time to finish key systems and software. Still,
the Navy will produce”and in some cases install”key systems on the ship
before fully demonstrating and testing them. Software development has
proven challenging; the Navy certified the most recent software release
before it met about half of its requirements. At this point”the first
year of a 6-year construction schedule for the lead ships”the Navy may
have exhausted its options for solving future problems without adding
money and time.
The Navy expects to achieve a greater degree of design maturity before
starting construction than has been the case on previous surface
combatant programs. To meet this goal, the Navy will be pressed to
complete a large amount of design work by October 2008 when
construction will begin. From August 2007 through May 2008, the
shipbuilders finished work on 16 of the 100 design ’zones“ that make up
the ship, leaving 84 zones to finish the final design phases in the 5
months leading up to the start of construction.
Both shipyards that will build DDG 1000 are preparing for construction
through facility enhancements and production improvements. However,
uncertainty remains. The ship‘s deckhouse will be built primarily from
composite material rather than steel and the shipbuilder is still
refining the process for large-scale composite manufacturing and
assembly. Workforce instability could also prevent shipbuilders from
fully realizing expected efficiencies.
The full costs of constructing the two lead ships have not been
entirely recognized or funded. The risk of cost growth is high in part
because of the potential for late delivery of key systems and software
and difficulties in constructing and integrating sections of the ship,
like the deckhouse. Remaining funds may not be sufficient to buy key
components and pay for other work not yet under contract. The Navy has
already requested funding for a third ship and plans to contract for
this ship with options for four more ships in fiscal year 2009. The
Navy will not have enough data then on the actual costs of the lead
ships to develop realistic prices for follow-on ships. As currently
planned, all ships will be under contract and all but one under
construction before the Department of Defense holds the production
milestone review in 2014.
What GAO Recommends:
GAO is recommending that:
* the design be completed before starting ship construction;
* award of contracts for the third ship be delayed until costs of the
lead ship are better understood, and;
* the production milestone review be held before contracts for the
third ship are awarded.
The Department of Defense concurred with the first and partially
concurred with the third recommendation. It did not agree to delay the
third ship, citing potential cost and other impacts.
This report also contains a matter for congressional consideration
aimed at providing additional insight into the program.
To view the full product, including the scope and methodology, click on
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-08-804]. For more
information, contact Paul Francis at (202) 512-4841 or
francisp@gao.gov.
[End of section]
Contents:
Letter:
Results in Brief:
Background:
Delays in Technology and Software Development Pose Risk to Successful
Program Execution:
The Navy Is Emphasizing a Mature Design on DDG 1000, but Timely
Completion Will Be Challenging:
Despite Planned Improvements, Shipyard Facility, Labor, and
Manufacturing Issues Could Affect Construction of the Lead Ships:
Actual Costs of Ships Will Likely Exceed Budget:
Conclusions:
Recommendations for Executive Action:
Matter for Congressional Consideration:
Agency Comments and Our Evaluation:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: Comments from the Department of Defense:
Appendix III: Major Events in the Development of DDG 1000:
Appendix IV: GAO Contact and Staff Acknowledgments:
Tables:
Table 1: DDG 1000 Critical Technologies:
Table 2: Prototype Development and Results:
Table 3: Challenges Faced by Shipbuilders in Meeting Detail Design
Schedule:
Table 4: Design Activities at Stages of Zone Progress:
Table 5: Unfunded Lead Ship Expenses:
Figures:
Figure 1: DDG 1000 Destroyer Currently in Development:
Figure 2: DDG 1000 Lead Ship Schedule:
Figure 3: Integrated Power System Schedule:
Figure 4: Dual Band Radar Schedule:
Figure 5: Total Ship Computing Environment Schedule:
Figure 6: Total Ship Computing Environment Releases:
Figure 7: DDG 1000 Detail Design Process:
Figure 8: Detail Design Schedule Performance:
Figure 9: Zone Milestone Review Progress (as of March 2008):
Figure 10: Northrop Grumman Shipbuilding Productivity Constructing DDG
51 Class (as of January 2008):
Figure 11: DDG 1000 Oversight, Ship Contract Award, and Construction
Events after Milestone B:
Abbreviations:
DOD: Department of Defense:
RDT&E: Research, Development, Test, and Evaluation:
[End of section]
United States Government Accountability Office:
Washington, DC 20548:
July 31, 2008:
The Honorable Edward M. Kennedy:
Chairman:
The Honorable Mel Martinez:
Ranking Member:
Subcommittee on Seapower:
Committee on Armed Services:
United States Senate:
The U.S. Navy aims to begin construction of the first lead Zumwalt-
class destroyer (DDG 1000) this year--a major step in its shift to the
next generation of warships. DDG 1000 is a multimission surface ship
designed to provide advanced land attack capability and contribute to
military dominance in the shallow coastal waters known as the
littorals. To meet its objectives, DDG 1000 will employ a revolutionary
hull design (known as the tumble-home form) and an array of cutting-
edge technologies, including a missile launch system lined in the
ship's hull, an advanced gun system that fires long-range precision-
guided munitions, and highly capable sensors integrated into the sides
of a deckhouse made primarily of composite material--not steel. Many of
these innovations contribute to making DDG 1000 significantly less
detectable than current ships. The Navy plans to achieve these advances
while also reducing the crew size to less than half of the predecessor
Arleigh Burke-class destroyer (DDG 51) through extensive computer
automation. The Navy is investing significantly in the DDG 1000
program--almost $9 billion in research and development and almost $20
billion to design and deliver the seven ships of the class.
Given the challenges associated with developing, designing, and
constructing a ship as complex as DDG 1000--including the potential for
cost growth, you asked us to examine the program's progress,
particularly in light of the start of construction of the two lead
ships (DDG 1000 and DDG 1001) and the Navy's request for authorization
of the third ship (DDG 1002) in its fiscal year 2009 budget.
Specifically, we assessed (1) the Navy's ability to deliver key systems
when needed to support lead ship construction and whether these systems
will work as intended, (2) the level of design maturity and the
obstacles to completing the ship's design as planned, (3) the
shipyards' readiness to build DDG 1000, and, finally, (4) the
challenges of building the ships (lead and follow-ons) within budget.
To address the first objective, we analyzed key documents, including
test plans, test reports, and production schedules and supplemented our
analysis by visiting contractors and test sites where the ship's major
systems are being developed and tested. To assess design maturity and
obstacles to completing the ship's design, we examined the Navy's plans
and guidelines for design management and evaluated the program's design
progress. We analyzed design maturity metrics captured in the program's
integrated master schedule and assessed design performance by analyzing
shipbuilder cost performance reports. To examine shipyard readiness, we
reviewed key shipyard data, including past construction performance and
facility improvement and investment plans. Finally, we examined ship
costs by analyzing the budget, Navy and independent cost estimates, and
contractor costs. To address all of the above objectives, we held
discussions and attended briefings with the Office of the Secretary of
Defense, Navy, and DDG 1000 program officials; contractors for the
mission systems; and the shipbuilders. We conducted this performance
audit from September 2007 to July 2008 in accordance with generally
accepted government auditing standards. Those standards require that we
plan and perform the audit to obtain sufficient, appropriate evidence
to provide a reasonable basis for our findings and conclusions based on
our audit objectives. We believe that the evidence obtained provides a
reasonable basis for our findings and conclusions based on our audit
objectives. Appendix I further discusses our scope and methodology.
Results in Brief:
A recent decision to restructure and add 1 year to the program's
schedule will provide more time to produce and test key combat systems
that will be installed on the lead ships, but calls for doing more work
later in construction and after ship delivery. The Navy planned to
demonstrate the capabilities of production-representative units of the
ship's major systems before installation, an approach aimed at
minimizing cost and schedule risk on the lead ship. It has been able to
do so for a number of systems. The restructure does not delay the
completion of the hull, mechanical, and electrical systems but does
delay verifying the performance of the integrated power system, the
dual band radar, and software before producing--and in some cases
installing--them on the ship. Specifically:
* Land-based testing of the integrated power system will not finish
until 2011, over a year after units are installed on the first ship.
* Delays in land-based testing have slipped installation of the dual
band radar by at least 3 years. Moreover, the Navy will not demonstrate
the full-power output of the radar needed to meet requirements until
after testing of the first production unit sometime after 2010.
* Software development--crucial to the automation that enables reduced
manning of the ship--has proved challenging. For the most recent
software release, the contractor had difficulty coding the ship's
command and control component. The Navy certified the release without
it meeting about half of the software system requirements, moving work
to later releases. Yet the most challenging phases of software
development lie ahead. Delays would disrupt plans for activating the
ship's critical systems, as well as its combat systems.
The DDG 1000 program is positioned to achieve a greater degree of
design maturity before starting construction than has been the case on
previous surface combatant programs. However, the shipbuilders may face
challenges completing the later, more complex phases of design on
schedule. Since August 2007, the shipbuilders finished work for
approximately 16 of 100 zones (the individual units that make up the
ship's design). The shipbuilders believe that they will complete
another 84 design zones by October 2008, when construction of the first
ship begins. According to the shipbuilders, a revised schedule allows
them to complete design tasks more efficiently. However, delays in the
delivery of final technical information for the ship's key systems
could prevent the shipbuilders from completing the design according to
plan. If the shipbuilders cannot achieve their design goals according
to plan, the program may experience design-related cost growth as have
previous shipbuilding programs.
Although the shipbuilders are preparing for construction of the lead
ships by investing in facilities, uncertainty remains. Both shipyards
are preparing for DDG 1000 construction through facility enhancements
and production improvements that they believe will increase efficiency.
For example, one shipbuilder is building a facility to produce larger
steel panels. The other shipbuilder built a new facility that will
allow construction of DDG 1000 from fewer, larger units. However,
achieving the full benefit of these improvements will depend in part on
successful execution of the shipbuilders' optimal construction
strategy--which includes the right mix of facilities, workforce, and
manufacturing capability. Continued labor instability and potential
problems manufacturing the deckhouse could lead to increases in the
amount of work involved in lead ship construction. Because DDG 1000
requires a new manufacturing method, projected efficiencies may not be
fully realized on the lead ships. In particular, the facility that will
manufacture the composite deckhouse was damaged by Hurricane Katrina,
and the shipbuilder is still refining the process for large-scale
composite manufacturing and assembly. Problems discovered during
construction would likely disrupt the ship's construction sequence.
Costs of the DDG 1000 ships are likely to exceed current budgets
because the true costs necessary to deliver fully operational ships
have not been fully recognized or funded.[Footnote 1] The lead ships
are at risk of cost growth for a number of reasons--including late
delivery or problems with the performance of key systems or issues
constructing or integrating sections of the ship. For example, if
composite manufacturing for the deckhouse takes longer than expected,
ship construction would be delayed and costs associated with
manufacturing as well as integrating the deckhouse would likely rise.
Since there is little margin in the budget, Navy officials have said
that funds reserved for buying components of the dual band radar and
certain communications antennas in later fiscal years may be needed to
support ship construction. If this occurs, additional funds would need
to be approved in fiscal year 2010 or later. The Navy has requested
funding for the third ship and plans to contract for this ship with
options for follow-on ships during fiscal year 2009. By this time, the
Navy will have limited data on actual costs for the lead ships and
their combat systems. Finally, the program's production decision is
slated for 2014--after all but one ship is under construction. Thus,
its timing does not allow for insight that could better inform decision
makers about follow-on ship costs.
We are making recommendations to the Secretary of Defense aimed at
reducing risk in the DDG 1000 program by (1) requiring that the ship's
detail design is completed as planned before the start of construction
and (2) deferring contract award for follow-on ships until the costs of
the lead ship are better understood. We are also recommending that the
program hold a Milestone C review in advance of awarding a contract for
the third ship. The Department of Defense (DOD) agreed with our first
recommendation, reiterating the importance of achieving design
maturity. DOD did not agree to hold a Milestone C review, stating that
such a review would not provide any additional benefit for the program.
We believe that a milestone review provides a consistent framework and
documentation that will help decision makers recognize and make any
needed adjustments to the program.
DOD also did not agree with our recommendation to defer the contract
award for the follow-on ships, stating that such as delay will impact
warfighting capability gaps, cost, and the shipbuilding industrial
base. We note that the Navy and DOD are currently considering canceling
the third and follow-on ships. This notwithstanding, we believe that
the risks of prematurely awarding follow-on ship contracts are
considerably higher than any potential cost increase or industrial base
effects that could occur as a result of delaying funding for a follow-
on ship.
Given the risk of cost growth in the program, this report also contains
a matter for congressional consideration to provide more insight into
potential cost increases as the DDG 1000 program progresses.
Background:
The Zumwalt-class destroyers are designed as multimission surface
combatant ships to provide advanced land attack capability to support
forces ashore and contribute to military dominance in shallow coastal
water environments. Along with nine new ship classes over the next 5
years, the Zumwalt-class destroyers are part of the Navy's planned
shift to the next generation of warships intended to complete future
mission objectives. The Navy will take ownership of the first two
Zumwalt-class destroyers--DDG 1000 and DDG 1001--in 2013 and 2014,
respectively. Figure 1 depicts an artist's rendition of the Zumwalt-
class ship (commonly referred to as DDG 1000).
Figure 1: DDG 1000 Destroyer Currently in Development:
[See PDF for image]
This figure is an artist's rendition of the DDG 1000 Destroyer
currently in development.
Source: Navy.
[End of figure]
The DDG 1000 program has been framed by challenging multimission
requirements, resultant numerous technologies, and a tight construction
schedule driven by shipyard workloads. The Navy's acquisition strategy
for DDG 1000 has changed significantly since the initial concept for
the ship in 1997. Over its development cycle, the cost of the DDG 1000
program has increased considerably, while the quantities have been
reduced. Currently, the Navy plans for a class of seven ships. The
Congress has limited the procurement cost of the fifth ship to $2.3
billion, plus adjustments for inflation and other factors.[Footnote 2]
For a more complete accounting of DDG 1000 program events see appendix
III.
Technology Development Approach:
Detail design and construction of DDG 1000 was preceded by a technology
development phase that continues today. The Navy sees DDG 1000 as the
technology driver for the fleet and a bridge to future ship capability.
The Navy plans to incorporate 12 new technologies to meet the ship's
demanding requirements. To reduce program risk and demonstrate the
ship's technologies, the Navy planned to build and test 10 prototypes
(referred to as engineering development models) representing the ship's
most critical subsystems. Table 1 describes the prototypes.
Table 1: DDG 1000 Critical Technologies:
System: Advanced gun system;
Description: Will provide long-range fire support for forces ashore
through the use of unmanned operations and the long-range land attack
projectile.
System: Integrated deckhouse and apertures;
Description: A composite structure that integrates apertures of radar
and communications systems.
System: Dual band radar;
Description: Horizon and volume search improved for performance in
adverse environments.
System: Integrated power system;
Description: Power system that integrates power generation, propulsion,
and power distribution and management.
System: Total ship computing environment;
Description: Provides single computing environment for all ship systems
to speed command while reducing manning.
System: Peripheral vertical launch system;
Description: Multipurpose missile launch system located on the
periphery of the ship to reduce damage to ship systems.
System: Integrated undersea warfare system;
Description: System for mine avoidance and submarine warfare with
automated software to reduce workload.
System: Infrared mock-up;
Description: Seeks to reduce ship's heat signature in multiple areas.
System: Hull form; Description:
Designed to significantly reduce radar cross section.
System: Autonomic fire suppression system;
Description: Intended to reduce crew size by providing a fully
automated response to fires.
Source: Navy data.
[End of table]
Prototype development and testing helped to give the Navy confidence in
a technology's ability to operate as intended. For example, the Navy
successfully demonstrated the advanced gun system through initial
guided flight and testing on land. In other cases, such as for the
integrated power system, tests brought to light technical problems that
the Navy was able to address by going to an alternate technology.
However, not all prototypes were completed as planned and development
continues even as the Navy prepares for construction of the lead ships.
Table 2 provides a status of DDG 1000 prototype development efforts.
Table 2: Prototype Development and Results:
Technology: Advanced gun system;
Results: Verified gun and magazine response time, rate of fire, range,
and pallet unloading rate, and demonstrated complete firing sequence
during 2004 through 2005. Separate flight tests demonstrated the long-
range land attack projectile munitions to be used by the gun. The gun
and munitions were not tested together because of test facility
limitations.
Technology: Autonomic fire suppression system;
Results: Successfully tested on two Navy test ships during 2004 through
2005. Demonstrated system's ability to detect damage and control fires
in specific ship environments. With these tests, the Navy designated
this technology as fully mature at critical design review.
Technology: Integrated deckhouse and apertures;
Results: An integrated deckhouse test article was tested in 2005 for
radar cross section and to verify that signals from apertures do not
interfere with one another. Fire and shock testing of a large-scale
test article to verify deckhouse strength was postponed because of
uncertainties about the composite material's strength and safety as
well as facility delays. This test is currently scheduled to be
completed by late 2008, and, if it is successful, the deckhouse will be
considered fully mature following analysis of the test.
Technology: Dual band radar;
Results: Multifunction radar proved clutter rejection and firm track
range--key functions required for demonstration--during land-based and
at-sea tests from 2004 through 2006; the radar was designated as fully
mature following these tests. Volume search radar encountered
difficulties with transmit-receive unit; "string" tests demonstrated
radar at lower power output than required in 2007 after 1-year delay.
Tests of the fully assembled dual band radar were delayed almost 2
years and are currently planned to begin in 2009.
Technology: Integrated power system;
Results: Turbine generators were factory tested in 2004 and 2005.
Permanent magnet motor failed 2005 tests and was replaced by its
backup, the advanced induction motor, which was then successfully
tested at land-based facility in 2005, allowing the program to pass
Milestone B.
Technology: Total ship computing environment;
Results: First three releases--24 percent of total code--were
successfully developed during 2003 through 2005. These releases
included critical computing infrastructure functionality needed for the
total ship computing environment, as well as limited ship mission
functionality.
Technology: Peripheral vertical launch system/MK57 vertical launch
system;
Results: The 2004 test resulted in destruction of the test article,
requiring redesign. A second test in 2005 replicated the same
conditions with the new design and materials and was successful.
Restrained firing tests--proving that the enclosure and launcher can
survive an explosion--were postponed and successfully performed in
2006.
Technology: Integrated undersea warfare system (acoustic sensor suite
element);
Results: Array interference tests were conducted in 2004 at the Navy's
Seneca Lake test facility. In 2005, the Navy verified mine avoidance
capability during at-sea tests and automated submarine warfare
detection and tracking in lab tests, and designated the technology as
fully mature by late 2005.
Technology: Infrared signature mock-ups;
Results: At-sea materials and panel tests followed by design tests were
conducted in 2004. Design trade-offs for performance, weight, and cost
led to changes in infrared materials used to reduce ship's heat
signature. A sheeting water system for the hull was replaced with an
alternate system.
Technology: Hull form;
Results: Scale models were tested during 2004 thorough 2005 for factors
like resistance, propeller efficiency, and capsize probability. Scale
model tests in 2006 led to design changes to reinforce the hull and
deckhouse. Tests continued into 2007 and 2008 and further development
is planned through 2015 to develop a "safe operating envelope" to help
safely operate the ship in high sea states.
Source: GAO analysis of Navy and contractor data.
[End of table]
Detail Design and Construction:
The Navy awarded contracts for detail design in August 2006 and
negotiated contract modifications for construction of two lead ships in
February 2008. Construction of the first ship is scheduled to begin in
October 2008, followed by the second ship in September 2009.
To build a ship efficiently and stay within budget, the lead ships need
to be constructed following an optimal construction sequence. This cost-
efficient approach calls for designing and building the ship in
modules, maximizing the amount of construction, test, and outfitting
completed in shipyard shops and in the dry dock while minimizing work
to be performed once the ship is in the water, which tends to be
costlier than tasks on land. This sequence is outlined in the
shipbuilders' integrated master schedule, which links all of the
detailed construction tasks based on key event dates. Shipbuilders try
to install the ship's systems in order to take advantage of
construction efficiencies. If equipment is not ready in time for
installation, the shipbuilder will have to work around the missing
equipment. Once units are installed, access to internal ship
compartments becomes more difficult. Additional labor hours may be
needed because spaces will be less accessible and equipment may require
more time for installation.
Contracting Strategy:
The Navy's approach to the DDG 1000 program is complex. Since 2006 the
Navy has been managing the DDG 1000 program through contracts with four
different prime contractors--two shipbuilders are jointly designing and
constructing the ships and two system developers are providing combat,
communications, and weapon systems as well as the software that will
operate and integrate these systems:
* Bath Iron Works is designing the fore and aft of the ship, equaling
about 50 percent of the individual design units of the ship, known as
design zones. This includes the distributive systems--such as cables
and pipes--that pass through the zones. Bath Iron Works will build the
first lead ship (DDG 1000) and will build the forward middle section to
be integrated on the second lead ship (DDG 1001).
* Northrop Grumman Shipbuilding is designing the deckhouse and ship
center (50 percent of the design zones), including the distributive
systems that pass through the zones. Northrop Grumman Shipbuilding will
build the deckhouse (at its composite facility in Gulfport,
Mississippi), hangar, and aft launch system sections to be integrated
on DDG 1000 and all ships in the class. Northrop Grumman Shipbuilding
will also build the second lead ship (DDG 1001).
* Raytheon is providing most of the ships' electronic systems (combat
and communications), including development of the radars and antennas.
Raytheon is also responsible for developing the ship's software system
(known as the total ship computing environment)--including defining
requirements, design, code, test, and integration.
* BAE is responsible for development, design, testing, and production
of the advanced gun system, including gun, automated magazine, and long-
range land attack munitions.
Under this contracting arrangement, the Navy is essentially acting as
the overall program integrator. Since the shipyards are sharing the
design of the ship equally, the Navy manages the overall ship
configuration and ensures that the ship's baseline design is aligned
across all of the contracts. All of the products produced by the
program's prime contractors (including the major systems, as well as
sections of the ship) are government-furnished equipment--meaning
equipment purchased by the Navy for installation and integration on the
ship. As such, the Navy is responsible for the receipt of sections of
the ship (such as the deckhouse or mid-fore body) from one shipbuilder
and the delivery to the other shipbuilder for integration with the rest
of the ship. As a result, the Navy is also responsible for managing the
thousands of interface points between the ship's sections to ensure
that they are correctly defined.
Delays in Technology and Software Development Pose Risk to Successful
Program Execution:
Although recent restructuring of the program allows more time for
producing, installing, and testing key combat systems and the software
that supports these systems, it calls for doing more work later in
construction and after ship delivery. While additional time is
beneficial, unexpected difficulties in producing systems or getting
them to operate as expected would likely result in additional delays
and increased costs. Some key systems, including the dual band radar,
have not been fully demonstrated as planned. The integrated power
system will be produced and installed before land-based testing is
complete. This approach increases the risk that problems discovered
during testing will require expensive rework to incorporate fixes later
on. The most challenging phases of the software development--crucial to
the automation that enables reduced manning of the ship--lie ahead,
presaged by problems encountered in testing the most recent software
release. Delays in software delivery would disrupt plans for activating
the ship's critical systems as well as its combat systems and delay
delivery of the ship and its combat systems to the Navy.
Recent Adjustments Provide Additional Time, but Many Events Need to
Come Together for Success:
Until recently, Navy plans called for delivery of the first ship in
2012, with initial fielding of the ship (known as initial operational
capability) in 2014. According to this schedule, the shipbuilders would
begin construction of the first ship in 2008. According to the Navy,
this was an important step in managing shipyard workloads as starting
later would have caused shipyard workload to drop too low. In order to
reduce the risk of cost growth and schedule delay during construction,
the Navy planned to demonstrate mature systems before installation on
the lead ship. To meet this schedule the Navy laid out a two-stage
approach to maturing the key systems. First, the Navy planned to build
and test almost all of its 10 prototypes of the key systems before
beginning detail design of the ship and, second, planned to transition
from development of prototypes to systems integration test and
qualification of production units after the program's critical design
review in 2005. While this approach was successful on several key
systems, the Navy has not been able to develop all key systems as
planned--some prototypes still have not completed testing. The Navy
chose to move forward into detail design of the ship before
demonstrating all technologies. In fact, the Navy continues to shift
work--in the form of testing, developing software, integrating
subsystems, and actual fabrication--until later stages of design and
even ship construction. In these stages, the cost of work and delays is
much higher and the schedule less forgiving than in earlier stages of
the program.
Following negotiations of the contract modifications for constructing
the lead ships, the Navy realigned the program's schedule. While the
timing of the start of ship construction remains roughly the same, the
Navy changed the definition of what constitutes ship delivery. Rather
than delivering a fully mission-capable ship, the Navy stated that it
will take ownership of just the vessel and its mechanical and
electrical systems--including the power system--in April 2013.[Footnote
3] At that point, the Navy will have activated and tested the power,
mechanical, and electrical systems aboard ship. Under the restructured
schedule, the Navy will delay work that typically occurs during
construction until after the ship is initially delivered. Specifically,
the Navy will light off--that is activate and shipboard test--the
combat systems, including the radars, antennas, guns, and the missile
launch systems, in May 2013--more than 2 years later than originally
planned (see fig. 2). Following combat systems light-off, the Navy will
begin acceptance trials of the combat systems--inspecting these systems
and noting any deficiencies that should be corrected. Only at the
conclusion of combat system acceptance trials in February 2014 will all
of the ship's systems be fully integrated and final ship delivery
complete.
Figure 2: DDG 1000 Lead Ship Schedule:
[See PDF for image]
This figure is an illustration of the DDG 1000 Lead Ship Schedule, as
follows:
Construction start:
Schedule as of 2005 program start: Early, 2008;
Schedule as of May 2008: Late, 2008.
Ship light-off:
Schedule as of 2005 program start: Early to mid-2011;
Schedule as of May 2008: Early to mid-2011.
Combat systems light-off:
Schedule as of 2005 program start: Early to mid-2011;
Schedule as of May 2008: Mid-2013.
Final acceptance trials:
Schedule as of 2005 program start: Mid-2012;
Schedule as of May 2008: Early 2014.
Final delivery:
Schedule as of 2005 program start: Mid-2012;
Schedule as of May 2008: Early 2014.
Initial operating capability:
Schedule as of 2005 program start: Early 2014;
Schedule as of May 2008: Early 2015.
Source: GAO analysis of Navy data.
[End of figure]
According to the Navy, conducting light-off in phases allows the
program to test and verify the ship's major systems, in particular the
integrated power system, in isolation before conducting simultaneous
tests with the combat and communication systems. Navy officials stated
that this approach will help to reduce the risk that problems with the
ship's systems, such as power load or cooling, could damage the combat
systems. By delaying these events until after delivery of the vessel,
the Navy is using this time to test, produce, and install key combat
systems, as well as the software that supports these systems.
Although this approach gives the Navy additional time to mature the
ship's combat systems before installation, it may limit its ability to
address any problems that may occur as a result of light-off and
shipboard testing. Since the Navy will only test and inspect the hull
prior to taking ownership of the vessel, it will not have a full
understanding of how the ship operates as a complete and integrated
system until after final shipboard testing of the combat systems in
2014. Moreover, final delivery of all the ship's software will now
occur right before final acceptance trials--compressing the time to
address any integration problems prior to initial operation of the
ship.
The Navy Will Not Demonstrate All Key Systems Prior to Producing and in
Some Cases Installing These Systems:
Because of delays in the development of several ship systems,
particularly the integrated power system, the dual band radar and the
electronic surveillance system, the Navy will not demonstrate
integrated systems before installing them on the lead DDG 1000 ship,
increasing the risk that problems discovered during testing will
require expensive and/or time-consuming rework to incorporate fixes
later on.
Integrated Power System:
The integrated power system will centrally generate and distribute
power for all ship functions, allowing greater flexibility in power
use. While the Navy has demonstrated the basic capability of the power
generation and distribution systems, tests of a complete integrated
power system with the control system will not occur until 2011--nearly
3 years later than planned (see fig. 3).
Figure 3: Integrated Power System Schedule:
[See PDF for image]
This figure is an illustration of the Integrated Power System Schedule,
as follows:
Construction start:
Schedule as of 2005 program start: Early, 2008;
Schedule as of May 2008: Late, 2008.
Integrated power system land-based test complete:
Schedule as of 2005 program start: Early 2008;
Schedule as of May 2008: Early to mid-2011.
Install integrated power system:
Schedule as of 2005 program start: Mid-2009;
Schedule as of May 2008: Late 2009.
Ship light-off:
Schedule as of 2005 program start: Mid-2011;
Schedule as of May 2008: Mid 2011.
Combat systems light-off:
Schedule as of 2005 program start: Mid-2011;
Schedule as of May 2008: Late 2013.
Final acceptance trials:
Schedule as of 2005 program start: Mid-2012;
Schedule as of May 2008: Early 2014.
Final delivery:
Schedule as of 2005 program start: Mid-2012;
Schedule as of May 2008: Early 2014.
Source: GAO analysis of Navy data.
[End of figure]
The Navy initially planned to verify power system performance through
integration tests with the first production unit at a land-based test
facility in 2008 before installing the power system on DDG 1000.
However, delays in developing the ship's software led the Navy to
postpone integration tests by over a year. To deliver the power system
on time to meet the shipyard's schedule, the Navy will buy a power
system intended for the third ship and use this unit in upcoming tests.
As a result, the integrated power system will not be demonstrated until
the completion of land-based tests that will occur over a year after
the power systems have been produced and installed on the two lead
ships.
Dual Band Radar:
The dual band radar is made up of two major radar systems, the
multifunction radar and the volume search radar. Development and
testing of the multifunction radar have progressed further than for the
volume search radar. Through testing on land and at sea, the Navy
demonstrated the multifunction radar's key functions. In contrast, the
volume search radar experienced difficulties during development that
still continue. Land-based demonstrations of the volume search radar
prototype and the integrated dual band radar originally planned to be
done before starting ship construction will not be completed until
2010--almost 2 years later (see fig. 4).
Figure 4: Dual Band Radar Schedule:
[See PDF for image]
This figure is an illustration of the Dual Band Radar Schedule, as
follows:
Construction start:
Schedule as of 2005 program start: Early, 2008;
Schedule as of May 2008: Late, 2008.
Volume search radar land-based test complete:
Schedule as of 2005 program start: Mid-2007;
Schedule as of May 2008: Mid-2009.
Dual band radar land-based test complete:
Schedule as of 2005 program start: Early 2008;
Schedule as of May 2008: Mid-2010.
Install dual band radar:
Schedule as of 2005 program start: Early 2010;
Schedule as of May 2008: Mid-2013.
Install dual band radar: Install multifunction radar:
Schedule as of 2005 program start: NA;
Schedule as of May 2008: Early 2010.
Install dual band radar: Install volume search radar (TBD)
Schedule as of 2005 program start: NA;
Schedule as of May 2008: TBD.
Ship light-off:
Schedule as of 2005 program start: Mid-2011;
Schedule as of May 2008: Mid-2011.
Combat systems light-off:
Schedule as of 2005 program start: Mid-2011;
Schedule as of May 2008: Mid-2013.
Source: GAO analysis of Navy data.
[End of figure]
Development difficulties center on the radar's radome and transmit-
receive units. The contractor has not yet successfully manufactured the
volume search radar's radome--a composite shield of exceptional size
and manufacturing complexity--that is necessary to meet radar stealth
requirements. The first radome constructed for the volume search radar
prototype experienced cracking and delaminating problems stemming from
a faulty design. After conducting analysis to determine the root cause,
the contractor's second attempt in late 2007 appeared to have solved
the initial problems, but new internal cracks were noted, apparently
caused by machining difficulties. The Navy is now funding a parallel
effort with a second manufacturing source--one that supplied the radome
for the much smaller multifunction radar--and will make a decision on
the path forward after a June 2008 review. Upcoming land-based tests
will be conducted without a radome, and a replacement will not be
available until the first production unit of the volume search radar is
tested at the factory.
The transmit-receive units--the individual radiating elements that are
the essence of the volume search radar--experienced failures during
component testing, never reliably operating at the voltage needed to
meet range requirements despite multiple design iterations and
different suppliers. According to the contractor, the latest test
results from January 2008 met performance requirements, but subsequent
tests revealed a new issue concerning the heat generated by a component
of the transmit-receive unit. The Navy believes that the voltage
problem has been resolved and that the fix has been validated in recent
tests. However, land-based tests of the volume search radar prototype
and the integrated dual band radar will not incorporate the redesigned
transmit-receive unit and will not demonstrate the higher-voltage
output necessary to meet ship requirements. The contractor does not
anticipate testing the volume search radar at the necessary voltage
until it produces and tests the first unit at the factory.
Problems with the volume search radar and in constructing the land-
based test facility delayed and compressed the production schedule for
the dual band radar. Land-based testing of the volume search radar
started late and is scheduled to be complete in May 2009. Following the
conclusion of these tests, the volume search radar and multifunction
radar prototypes will be integrated and tested with their enabling
software at the same land-based facility in 2010. However, the
multifunction radar will have been produced and delivered to the lead
shipyard before the integrated tests are completed. This concurrency
introduces additional risk if problems are discovered during testing
and components require rework or replacement. Moreover, delays have led
officials to plan for installation of the volume search radar when the
first ship is already afloat, a more expensive approach than the
original plan of installing the radar as the deckhouse is being
completed at the Gulfport facility. Meeting this schedule will also be
a challenge; the Navy recently decided to postpone contracting for the
major components of the volume search radar until 2010 or later.
Electronic Surveillance System:
The Navy has not decided on the electronic surveillance system. This
system is part of the deckhouse and will now be installed separately
after the first lead ship is initially delivered. The electronic
surveillance system provides passive threat detection for ship self-
defense and is essential for ship deployment. Existing systems cannot
be used because they do not meet the ship's radar cross section
requirements. The Navy is hoping to utilize one of two ongoing Navy
research and development activities, the multifunction electronic
warfare system or the surface electronic warfare improvement plan, but
does not currently know when these activities will be completed.
Meanwhile, in order to complete deckhouse design, the Navy has
allocated space for the electronic surveillance system. The final
deckhouse will be delivered with placeholders until a final system is
purchased and installed. Continued uncertainty about the electronic
surveillance system could ultimately delay ship deployment.
Software Development Faces Challenges and May Not Keep Pace with Ship
Integration and Test Schedule:
Achieving a high degree of computer automation is crucial to realizing
DDG 1000's required manning reductions. Given the risks associated with
the ship's software system, referred to as the total ship computing
environment, the Navy initially planned to develop and demonstrate all
software functionality over 1-year prior to ship light-off, when the
ship's systems are turned on for the first time. As a result of changes
in the software development schedule, the Navy eliminated this margin.
Until recently, the Navy was able to keep pace with its development
schedule, successfully completing the first three software releases.
However, the Navy is now entering the complex phases of software
development, which have proven challenging.
The magnitude and complexity of DDG 1000 software development is
unprecedented in Navy ships. Rather than having numerous stand-alone
systems, each operated by several sailors, the total ship computing
environment will automate and integrate many ship functions. Program
officials estimate that the total ship computing environment will
consist of 14 million to 16 million lines of code--including about 7
million lines of new and modified code developed in incremental
software blocks, currently described as six releases and one spiral. It
will be the single most important enabler of the ship's reduced manning
requirement. If the ship's software does not work as intended, crew
size would need to be increased to make up for any lack of automation.
With release 4, initial functionality of the major ship systems was
introduced. While the contractor successfully developed and delivered
the first three software releases, it encountered significant problems
developing release 4. The contractor delivered release 4 without
meeting all requirements and deferred work to release 5, mainly because
of issues coding the ship's command and control component--the heart of
the ship's decision-making suite. Problems discovered in this release,
coupled with the deferred work, may signify larger software issues that
could disrupt the development of releases 5 and 6 and prevent the
timely delivery of software to meet the ship's schedule.
In fact, the Navy certified release 4 without having met about half of
the release's software system requirements. Specifically, approximately
32 percent failed acceptance tests and another 15 percent were not
tested in this release.[Footnote 4] According to the contractor, the
main problem area was the command and control component, which
represents over one-third of release 4 code. This component repeatedly
slipped internal development milestones because of an inability to meet
recovery requirements, essentially, the need for software to
automatically restart and resume mission activities in the event of a
failure. The Navy believes that the causes of the recovery failures are
well understood and that fixes to these problems will be incorporated
early in release 5 development. However, officials from the Defense
Contract Management Agency and the Navy's Operational Test and
Evaluation Force believe that the problem is more complicated--the
recovery failures were largely driven by problems achieving timely data
communication through the JavaTM programming language[Footnote 5] used
for much of the ship's software,[Footnote 6] and were only discovered
late in release 4 integration tests when the command and control
software was integrated with the rest of release 4 code. The contractor
has also recognized that Java performance problems contributed to the
recovery failures. The program has had data communication difficulties
with this programming language since 2003.
Initial comments from the Navy's certification panel[Footnote 7] for
release 4 gave the program high marks in several areas, including
identifying errors early and building a strong validation and
verification capability. At the same time, the certification panel
commented that the functionality in the next release will present
challenges.
In order to mitigate program risk, the Navy's initial plan from 2005
envisioned developing all ship software by 2010--approximately 1 year
before ship light-off. This schedule would give developers time to
conduct integration tests to verify that the software would work with
the ship's mechanical and electrical systems as well as combat systems
before light-off. If any software problems were discovered during
testing, the program would have had time to incorporate software fixes
before and after ship light-off--but well before delivery. The Navy
still plans to conduct integration tests with the ship's software at
land-based test facilities. However, as a result of five software
schedule revisions over the last 2-1/2 years, the Navy now plans to
complete delivery of the software later in the ship's build sequence,
as shown in figure 5.
Figure 5: Total Ship Computing Environment Schedule:
[See PDF for image]
This figure illustrates the Total Ship Computing Environment Schedule,
as follows:
Program start, 2005:
Release 4: completion in late 2007;
Release 5: completion in late 2008;
Construction start: Mid-2008;
Release 6: completion in late 2009;
Spiral: begin mid-2007, completion mid-2010;
Ship and combat systems light-off: mid-2011;
Final acceptance trials: mid-2012;
Final delivery: late-2012;
Initial operating capability: early 2014.
Today, 2008:
Release 4: completion in Mid-2008;
Construction start: Mid-2008;
Release 5: completion in Mid-2010;
Ship light-off: Mid-2011;
Release 6: completion in Mid-2012;
Spiral (in draft): begin mid-2009, completion mid-2013;
Combat systems light-off: mid-2013;
Final acceptance trials: early 2014;
Final delivery: early 2014;
Initial operating capability: early 2015.
Source: GAO analysis of Navy and contractor data.
[End of figure]
Release 6, which contains critical engineering control system software,
has been delayed by 28 months and will now arrive in 2012, nearly 1
year after ship light-off in 2011 when the ship's power and electrical
components are turned on and tested on board the ship. While the
engineering control system software will be tested with the power
system at a land-based test facility in the months prior to ship light-
off, the software will not yet be integrated with the rest of release 6
code at this time. Instead, the Navy plans to conduct ship light-off
using interim engineering control system software while concurrently
integrating the code into release 6. If problems are discovered in the
engineering control system software during release 6 testing, not only
could ship light-off be delayed but ultimately delivery of the ship
could be delayed as well. Moreover, as early as the program's 2005
critical design review, the Navy recognized that all software code
needed to meet requirements would not be completed within the six
releases. Instead, the Navy added another block, or "spiral," of
software development. According to the Navy, approximately 25 percent
of the total software effort will be developed outside of the six
software releases. As figure 6 illustrates, a number of important ship
functions will not be integrated with the total ship computing
environment within the six releases.
Figure 6: Total Ship Computing Environment Releases:
[See PDF for image]
This figure is a stacked horizontal bar graph depicting the following
data:
Capability: Support, Training, and Testing;
Thousand equivalent lines of code, Spiral: 1051 (68%).
Thousand equivalent lines of code, Releases 4-6: 498.
Capability: Command and Control;
Thousand equivalent lines of code, Spiral: 39 (4%);
Thousand equivalent lines of code, Releases 4-6: 1005.
Capability: Power and Ship Control System;
Thousand equivalent lines of code, Spiral: 78 (12%);
Thousand equivalent lines of code, Releases 4-6: 560.
Capability: Network and Infrastructure;
Thousand equivalent lines of code, Spiral: 22 (4%);
Thousand equivalent lines of code, Releases 4-6: 467.
Capability: External Communications Antennas;
Thousand equivalent lines of code, Spiral: 8 (2%);
Thousand equivalent lines of code, Releases 4-6: 336.
Capability: Dual Band Radar;
Thousand equivalent lines of code, Spiral: 0;
Thousand equivalent lines of code, Releases 4-6: 299.
Capability: Acoustic Sensors and Decoys;
Thousand equivalent lines of code, Spiral: 39 (13%);
Thousand equivalent lines of code, Releases 4-6: 250.
Capability: Guns, Missiles, and Weapons Control;
Thousand equivalent lines of code, Spiral: 84 (32%);
Thousand equivalent lines of code, Releases 4-6: 180.
Capability: Anti-Terror/Force Protection;
Thousand equivalent lines of code, Spiral: 0;
Thousand equivalent lines of code, Releases 4-6: 184.
Capability: Electronic Surveillance;
Thousand equivalent lines of code, Spiral: 0;
Thousand equivalent lines of code, Releases 4-6: 165.
Source: GAO analysis based on Navy and contractor data.
[End of figure]
For example, software that automates the electronic surveillance and
antiterror/force protection systems will be integrated into the ship's
software after the ship has already been initially delivered.[Footnote
8] This software, as well as the final software spiral will not be
installed until shortly before combat systems light-off. Further, in
2005, the Navy's Operational Test and Evaluation Force raised concerns
that the Navy would increase the risk that critical defects may go
undetected until after the ship is delivered by sacrificing software
integration testing for the sake of schedule--concerns that continue to
be raised in current assessments of the program. The contractor has
reduced software integration test runs by 20 percent in release 4
through consolidating test procedures in an effort to save costs. The
Navy, however, believes that this approach appropriately balances these
risks and states that these reductions avoid redundant testing.
The Navy Is Emphasizing a Mature Design on DDG 1000, but Timely
Completion Will Be Challenging:
The Navy recognizes the need to mature DDG 1000's design before
beginning ship construction and plans to have a more complete design
than previous surface combatant programs. The Navy finished the initial
phases of design development, and the shipbuilders are currently
developing the individual zones of the ship's design. However, the
program faces challenges completing the core of detail design: product
modeling of the ship's zones. In March 2008 the shipbuilders revised
the design schedule to better manage the design work remaining. While
the shipbuilders believe they can finish the design of most of the
ship's zones by the start of ship construction, there are a number of
challenges in meeting this schedule. In particular, delays in the
development of the ship's key systems could impede completion of the
design and eventually interfere with DDG 1000 construction. If the
delays continue, the shipbuilders may not finish all modeling prior to
the start of lead-ship construction, placing the program at greater
risk for costly rework and out-of-sequence work during construction.
The Navy Is Emphasizing Greater Design Maturity Prior to Starting
Construction Than in Previous Shipbuilding Programs:
The Navy aims to have 85 percent of all detail and manufacturing
drawings complete by the start of lead ship construction--significantly
more than in previous surface ship programs. In contrast, the DDG 51
program had only 20 percent of necessary design and manufacturing
products finished before the start of construction, and the first
Littoral Combat Ship (LCS 1) had less than 25 percent complete.
According to the Navy, the DDG 1000 program is modeling the Virginia-
class submarine program that had its design matured prior to the start
of construction, thereby reducing cost growth related to design
immaturity.
The DDG 1000 program has already completed most of the initial two
phases of the ship's design (see fig. 7).
Figure 7: DDG 1000 Detail Design Process:
[See PDF for image]
This figure is an illustration of the DDG 1000 Detail Design Process,
as follows:
Functional design:
Defines ship's initial engineering and design.
Transition design:
Arranges major equipment and ship systems in each zone.
Zone/nonzone detail design:
Finalizes the design with three-dimensional modeling.
Two-dimensional production extraction:
Converts the design to two-dimensional graphics for production.
Source: Navy and shipbuilder data.
[End of figure]
Since 2007, the shipbuilders have been using a computer-aided design
product model to work on the details of the 100 individual zones that
constitute the ship, the heart of detail design.[Footnote 9] The
product model is shared between both shipbuilders and provides
transparency in the designs at both shipyards, as well as access to the
Navy to conduct design reviews. The product model generates a detail
design, allowing engineers to visualize spaces and test the design.
This validates elements of the design prior to construction, thereby
avoiding potentially costly rework.
During product modeling, the designers finalize the interfaces between
zones, complete the design for shipwide cables and pipes, and add all
detail necessary to support manufacturing of the ship. The Navy
assesses the progress of each zone when it is 50, 70, and 90 percent
complete with product modeling. At these critical reviews, the Navy and
other stakeholders review the zone design as it progresses and provide
input to ensure that the design meets specifications. At the completion
of zone design, the design is converted to drawings used for
construction.[Footnote 10]
Following the construction contract negotiations, the Navy revised the
detail design schedule to better ensure that the design was largely
complete by the ship's final production readiness review in October
2008. According to the shipbuilders, the revised schedule moved design
work to fit a more achievable timeline and realigned work to recognize
that the first ship would be built at Bath Iron Works rather than
Northrop Grumman Shipbuilding as initially planned. The program aims to
have all but 15 design zones essentially finished with product modeling
by construction start in October 2008, which should help to mitigate
the risk of the design-related cost growth that has occurred in
shipbuilding programs.
Shipbuilders Experienced Design Delays Early in Product Modeling:
Thus far, the Navy has not been able to meet its initial detail design
schedule. The Navy aimed to have essentially completed 82 zones by
March 2008, but only finished 12. The Navy and shipbuilders have cited
a number of causes for these schedule delays, discussed in detail in
table 3.
Table 3: Challenges Faced by Shipbuilders in Meeting Detail Design
Schedule:
Challenge: Latency and stability issues with the design tool;
Description: The modeling software has not been able to efficiently
accommodate the complexity of the design because of the large volume of
information it generates. Problems with the modeling software have
delayed detail design. For at least 7 months, computer downtime at one
shipyard consistently resulted in the loss of at least an hour of work
per person per day because of system failures, inconsistent software
behaviors, and more than double the anticipated time taken for data to
be stored in the database. Since some of the problems are due to the
complexity in the design, the shipbuilder estimates that the
unfavorable performance with the design tool will likely continue to
negatively affect the design schedule as modeling complexity increases.
As the models grow in complexity and size they will continue to stress
the system. The shipbuilders believe that they are taking actions to
improve the latency in the design tool; these actions include software
upgrades, hardware upgrades, and process improvements.
Challenge: Understaffing at one shipyard;
Description: One shipyard experienced slow hiring during the beginning
of the detail design phase, and it had to outsource some engineering
tasks because of its inability to hire.
Challenge: Delays incorporating changes to the design specifications;
Description: The product modeling tool allowed shipbuilders to discover
design problems early in design development, but any changes that
altered the ship's specifications required formal Navy approval. In
addition, a number of interfaces for key systems were not available
during initial design phases, so the shipbuilders had to use surrogate
data where information did not exist. As the systems matured, any
space, weight, power, or cooling changes also required formal Navy
approval. The process to receive Navy approval and incorporate the
changes into the design took longer than anticipated--sometimes as long
as 4 to 6 months.
Challenge: Missing technical Information for library parts;
Description: Late technical diagrams and calculations delayed interface
specifications and space dimensions needed for the shipbuilders to
issue final library parts--the individual pieces of equipment and
material within the design zones. The shipbuilders had to create
temporary library parts when none existed in order to proceed with ship
design products.
Source: GAO analysis of shipbuilder data.
[End of table]
Cost performance reports from both shipyards indicate that until the
recent schedule revision, the shipbuilders have not been able to
complete the product modeling work as planned. Performance in product
modeling is a leading indicator for the overall detail design effort
for two reasons. First, product modeling in zone design accounts for
almost half of the total planned labor hours in the detail design
contract. Second, of the remaining labor hours in detail design, about
65 percent are planned as "level-of-effort" tasks, for which
performance equals scheduled tasks and no schedule variances can occur.
[Footnote 11]
Using data from cost performance reports, we analyzed the shipbuilders'
schedule performance in product modeling according to a schedule
performance index that measures the value of the work completed against
the work scheduled (see fig. 8). If a schedule performance index is
less than 1.00, then less than a dollar's worth of work has been
completed compared with a dollar's worth of the work that was
scheduled.[Footnote 12] Over a 7-month period, the Navy received an
average of less than 84 cents worth of scheduled work for every dollar
planned.
Figure 8: Detail Design Schedule Performance:
[See PDF for image]
This figure is a line graph depicting the following data:
Month: September;
Schedule performance index: 0.87.
Month: October;
Schedule performance index: 0.86.
Month: November;
Schedule performance index: 0.84.
Month: December;
Schedule performance index: 0.82.
Month: January;
Schedule performance index: 0.79.
Month: February;
Schedule performance index: 0.77.
Month: March;
Schedule performance index: 0.88.
Source: GAO analysis of shipbuilder data.
Note: The shipbuilders believe that performance is understated because
work done on changes had not been recognized in the baseline.
[End of figure]
The shipbuilders believe that the March 2008 revised schedule will
enable them to complete the design according the new schedule. While
the progress of the product design model itself has not changed, the
revised schedule has realigned the planned work more closely with
actual progress, which explains the improvement in the performance
index from February to March 2008. A revised schedule will only show a
temporary improvement unless it is backed up with sustained progress.
Thus, if the shipbuilders cannot complete design work according to the
new schedule, performance will again erode.
The Shipbuilders Will Face Challenges Completing the Design to Meet the
Construction Schedule:
The shipbuilders will face challenges completing the final complex
product modeling stages according to the Navy's plan. The program is
now entering the more difficult final product modeling phases--when the
details of the design must be finalized to support constructing the
ship. Most design zones have passed the 50 percent milestone review,
and shipbuilders are now primarily working on the design activities
that occur when the zone is from 50 to 90 percent complete. As shown in
table 4, the second half of zone product modeling contains more
detailed design tasks; in addition to completing all interfaces,
detailed system arrangements, and pipe and cable routing, the
shipbuilders must finalize and incorporate all outstanding data from
the key systems. When the zone is 90 percent complete it is considered
essentially finished with detail design--all prior review actions must
have been incorporated into the design and the zone must be completely
outfitted and considered ready for production.
Table 4: Design Activities at Stages of Zone Progress:
Percentage: complete: 50 percent;
Required design activities:
Major arrangements, primary structure, major distributive systems;
* Passageways have been defined;
* Foundational design is complete;
* Major arrangements are defined.
Percentage: complete: 70 percent;
Required design activities: Remaining distributive systems;
* Distributive systems are routed;
* Large-size pipe work and cableways are identified and modeled;
* Hull items are modeled;
* Removal routes are known;
* Penetration requirements for system interfaces are updated.
Percentage: complete: 90 percent;
Required design activities: Essentially complete model content;
* Any remaining distributive systems are fully completed;
* Design zone is completely outfitted;
* One hundred percent of all critical system information is delivered;
* All interfaces are defined;
* Formal handoff for final quality assurance check before production.
Source: Navy and shipbuilder data.
[End of table]
The shipbuilders have had difficulty completing these tasks. In some
cases, they did not anticipate the substantial amount of work required
after 70 percent of the zone was complete. As shown in figure 9, since
starting zone design in August 2007, 70 fewer zones than initially
planned had passed the 90 percent review by March 2008.
Figure 9: Zone Milestone Review Progress (as of March 2008):
[See PDF for image]
This figure is a stacked horizontal bar graph depicting the following
data:
50% review:
Number of zones complete, actual reviews complete: 75;
Number of zones complete, remaining reviews: 18;
Number of zones complete, Originals plan: 7.
70% review:
Number of zones complete, actual reviews complete: 36;
Number of zones complete, remaining reviews: 53;
Number of zones complete, Original plan: 11.
90% review:
Number of zones complete, actual reviews complete: 12;
Number of zones complete, remaining reviews: 70;
Number of zones complete, Original plan: 18.
Source: GAO analysis of shipbuilder and Navy data.
[End of figure]
Despite revising the design schedule, the shipbuilders still have a
significant amount of design work to complete in a short amount of
time. The shipbuilders now aim to complete product modeling in 84
additional design zones by October 2008 to meet the Navy's current
schedule.[Footnote 13] This could be a difficult goal to meet, since
the shipbuilders already delayed the same tasks for most design zones
by an average of more than 4 months and now have a compressed amount of
time in which to complete the final tasks needed in order to meet the
target dates for production.
Key System Delays Could Impede Final Design Phases:
The Navy is required to deliver technical information to the
shipbuilder, including cooling and power needs, space and weight
requirements, and electrical interface data of various technology-
dependent systems. Without this information, the shipbuilder cannot
complete the design of the ship. Because key systems are in development
at the same time as the shipbuilders are working on the detail design,
the technical information needed for design has not been available when
needed and the design schedule may continue to slip if there are delays
in the delivery of technical information. Since some of the ship's key
systems were still immature when detail design began, much specific
technical information was not yet known and only general parameters,
such as space and weight requirements, could be established. In order
to meet the overall ship schedule, the Navy developed a system of
incremental "drops" of technology information, in which the
shipbuilders and contractors exchange data requirements of increasing
complexity at specific times to support design development. However,
the development schedule for some key systems has already slipped,
resulting in late drops of technical information to the shipbuilders
and delays to the design schedule. For example, most design zones have
been affected by missing interface data for cables that are attached to
key systems. Further delays in completing tests for critical systems,
such as the dual band radar, could continue to delay the delivery of
technical information about these systems to the shipbuilders. If
delays occur, the shipbuilders may not be able to complete the design
zones as planned.
Some of DDG 1000's key systems, such as the integrated power system and
dual band radar have not completed testing. Problems discovered in
testing may lead to redesign, which could result in changes to sections
of the ship. For example, testing using models revealed weaknesses in
the ship's hull form in extreme sea conditions. According to Navy
officials, this led to design changes with additional stiffeners added
to the ship for increased stability. In addition, since the Navy has
not defined the electronic warfare system, its final interfaces with
the ship are not yet known. While the shipbuilders have space reserved
for the system, its specifications may change once its physical
characteristics are defined. As the technologies continue to mature,
these changes and subsequent delays could continue.
Design Maturity Is Essential to Achieving Program's Cost Goals:
If the shipbuilders cannot achieve their design goals according to
plan, the program may experience the same design-related cost growth
that has been common in other shipbuilding programs.[Footnote 14]
Programs that did design work concurrently with construction were later
forced to conduct work out of sequence, devote additional labor hours
responding to design changes, or perform rework at the shipyard. For
example, CVN 77, the final aircraft carrier of the Nimitz class,
required concurrent design, planning, and production activities.
Additional labor hours were spent responding to design changes, which
ultimately affected cost and schedule. Similarly, when construction of
DDG 91 and DDG 92 began, critical technologies were not mature and
significant details of the design had not been captured. As a result,
workers were required to make significant structural changes to 26 of
the ship's 90 design zones. In contrast, the DDG 1000 program is
planning to essentially complete detail design before beginning
construction, but faces much work ahead to achieve this goal. Moreover,
development of key systems such as the dual band radar and integrated
power system will continue well into construction, creating the risk of
costly design changes later.
Despite Planned Improvements, Shipyard Facility, Labor, and
Manufacturing Issues Could Affect Construction of the Lead Ships:
Although the Navy and the shipbuilders are attempting to reduce the
risks associated with lead ship construction, considerable uncertainty
remains. Both shipyards are preparing for DDG 1000 construction through
facility enhancements and production improvements that they believe
will increase efficiency. In order to balance the risk associated with
constructing the lead ships, the shipyards are building test units
designed by each other's yard. While these efforts will certainly build
confidence, executing the plan will be challenging. In order to deliver
DDG 1000 with its promised capabilities and within planned costs, the
shipbuilders need to successfully execute the optimum construction
sequence--which includes the right mix of facilities, workforce, and
manufacturing capability. Continued labor instability and potential
problems manufacturing the deckhouse could lead to increases in the
amount of work involved in lead ship construction.
Shipbuilders Expect Investments in Facilities and New Design Features
to Improve Efficiency during Construction of DDG 1000:
In order to balance the risk associated with constructing the lead
ships, the Navy and the shipbuilders are taking steps to prepare for
construction. Investments in upgraded facilities, production pilots,
and new design features are expected to make DDG 1000 easier to
construct than previous ships.
Both shipbuilders have invested in new facilities that the companies
believe will improve efficiency during DDG 1000 construction. At
Northrop Grumman Shipbuilding, many of the shipyard's facilities were
damaged or destroyed by Hurricane Katrina and ships under construction
in the yard and labor productivity were affected, including
construction of the DDG 51 class (see fig. 10). While estimates
indicate that performance has not yet returned to pre-Katrina levels
for all DDG 51s currently under construction, the shipbuilder believes
that reductions in labor hours will be achieved with construction of
its final DDG 51-class ship (see fig. 10).[Footnote 15]
Figure 10: Northrop Grumman Shipbuilding Productivity Constructing DDG
51 Class (as of January 2008):
[See PDF for image]
This figure is a line graph depicting the following data:
DDG 51 class: 91;
Labor hours: 3,850,000.
DDG 51 class: 93;
Labor hours: 3,650,000.
DDG 51 class: 95;
Labor hours: 3,500,000.
DDG 51 class: 97;
Labor hours: 3,450,000.
[Hurricane Katrina from 98 to 105]
DDG 51 class: 98;
Labor hours: 3,400,000.
DDG 51 class: 100;
Labor hours: 3,650,000.
DDG 51 class: 103;
Labor hours: 3,850,000.
DDG 51 class: 105;
Labor hours: 3,700,000.
DDG 51 class: 107;
Labor hours: 3,500,000.
DDG 51 class: 110;
Labor hours: 3,400,000.
Source: Navy data.
Note: Actual labor hours are not computed because data may be
proprietary.
[End of figure]
According to Navy officials, the shipbuilder is taking advantage of
opportunities to improve its facilities as it rebuilds and repairs the
shipyard. Expansion of the facility where the shipbuilder manufactures
steel panels and other upgrades is underway. The shipbuilder expects to
first use these facilities to construct a DDG 1000 test unit in June
2008.
At Bath Iron Works, a new facility known as the Ultra Hall and a new
transportation process will allow the shipbuilder to assemble larger
manufacturing units than its prior processes. The facility was finished
in February 2008, and the shipbuilder believes that efficiencies will
be achieved on construction of its last three DDG 51-class ships. These
ships are expected to be erected from just 11 units compared with the
43 units that were erected to build the first DDG 51. Bath Iron Works
plans to erect DDG 1000 from 9 units--2 fewer units than the
shipbuilder's final DDG 51 ship--which is significantly smaller than
DDG 1000.
The Navy and the shipbuilders have recognized that building DDG 1000
using a design shared between shipbuilders carries some risk, and
initiated measures to build confidence in preparation for ship
construction. The shipbuilders will construct units to test production
processes and demonstrate capability to build the other shipyard's
design. The test units--called advanced machinery blocks--are intended
to demonstrate detail design and production processes, material costs,
and craft labor actual costs. While these demonstrations are important
tests of DDG 1000's build strategy, the shipbuilders are not scheduled
to complete all units until November and December 2008, after they plan
to start lead ship construction.
In addition, the shipbuilders and the Navy expect a number of new
design features will improve the producibility of DDG 1000 compared to
that of previous ships built:
* Increased rafted machinery. Machinery will be assembled into a single
module (raft) and the raft will be installed on board the ship, instead
of attaching individual pieces of machinery directly inside the ship.
This allows pre-outfitting and test and supports more efficient
installation; a 9,500 labor hour reduction is expected.
* Increased deck height. Three-meter deck height allows for less
congested outfitting with straighter pipe runs and fewer holes for
distributed systems, as well as an easier prefabrication process.
* Bundled distributed systems. Use of cableways allows work to be done
in the shops instead of on board the ship. In conjunction with
increased deck height, the bundling reduces interferences with ship
structure; a 46,000 labor hour reduction is expected.
Shipyard Labor Instability and Difficulty in Manufacturing the
Composite Deckhouse Could Disrupt the Ship's Optimum Construction
Sequence:
While capital investments and new design features may result in
improved construction efficiency, achieving the full benefit of
shipyard improvements will depend on successful execution of the
optimal construction sequence, which must be supported by facilities
that can manage construction capacity and new manufacturing methods.
Labor instability at the shipyards and uncertainties associated with
manufacturing the new composite deckhouse may affect execution of an
optimum construction sequence.
Labor Instability:
Shipbuilders emphasize that an experienced and stable workforce leads
to better quality, higher productivity, and ultimately lower production
costs. However, if current workforce trends continue, cost and schedule
targets for the construction of DDG 1000 may be affected.
Bath Iron Works is a smaller yard, highly dependent on the DDG 51
class. Therefore, the company has less flexibility to absorb workload
shifts before needing to initiate layoffs. A number of its most
experienced workers are nearing retirement and, thus, if laid off would
be less likely to return to the shipyard when business conditions
improve.
Northrop Grumman Shipbuilding has made progress in recovering from the
effects of Hurricane Katrina, but the shipyard continues to struggle
with worker attendance and attrition rates--resulting in a less
experienced and less efficient workforce. Trends in attendance and
attrition at the Northrop Grumman shipyard have moved in a positive
direction; average attendance is up from 34 to 36 hours per week and
average monthly attrition rate has improved from 1.7 to 1.0 since 2006.
Nevertheless, the workforce is still working at less than 40 hours a
week on average and the shipbuilder continues to compete for critical
craft labor with rebuilding activities in the region.[Footnote 16] Less
experienced workers require more time to complete tasks with additional
labor hours spent on rework. If workforce instability continues, DDG
1000 construction performance will likely be affected, ultimately
leading to higher labor hours and increased costs.
Furthermore, Northrop Grumman Shipbuilding may not have enough workers
to meet its demand. Including DDG 1000, the shipbuilder will have at
least nine hulls (from five different programs) in its shipyard when
construction of DDG 1001 is scheduled to begin in September 2009. The
choices the shipbuilder will have to make to balance its workload could
have negative affects on DDG 1000 construction. In order to manage its
construction schedule, Northrop Grumman may choose to make greater use
of overtime hours, though in the past, programs that have made
significant use of overtime faced additional cost increases because
overtime rates are higher than normal hourly wages. The shipbuilder
could also decide to outsource work in order to meet demand. Last fall,
the shipbuilder (with approval from the Navy) moved construction of
portions of the LHA 6 amphibious assault ship from Pascagoula to
Northrop Grumman Newport News, illustrating a challenge to managing
existing work in the yard.
Composite Deckhouse Manufacturing:
The DDG 1000 composite deckhouse design requires an expanded
manufacturing capacity and a new manufacturing method that is not yet
fully mature. While the Gulfport facility has manufactured composite
for several Navy ship classes, the combination of materials that will
be used on DDG 1000 is unique.[Footnote 17] Moreover, the shipbuilder
has never built a composite structure to the scale of the deckhouse
before. Finalizing deckhouse manufacturing and assembly processes and
ensuring that rigorous quality controls are in place are essential to
constructing and delivering the deckhouse as planned.
The Gulfport facility was severely affected by Hurricane Katrina and
repairs have been extensive. In addition, facility upgrades are
necessary to meet DDG 1000 demand. Because of the size of the
deckhouse, a large new bay is needed for assembly. Facility and
machinery upgrades necessary to construct and assemble the deckhouse
are not all scheduled to be complete until March 2010--over a year
after the start of construction of the first deckhouse. Northrop
Grumman Shipbuilding expects to complete efforts by the time the
upgrades are needed in the deckhouse construction schedule. However, if
difficulties with composite manufacturing occur, the deckhouses may not
be delivered to the shipyards on time, disrupting the construction
sequence of the ships.
Northrop Grumman Shipbuilding has yet to finalize deckhouse
manufacturing and assembly processes. Research and development efforts
were undertaken to assess the feasibility of using composite material
in the deckhouse. A number of test articles were built, including a
section of the deckhouse that was full scale and one face (side) of the
deckhouse that represented the most complex panel. Development efforts
continue as the shipbuilder has been changing the manufacturing
processes for deckhouse production and manufacturing test articles to
validate them. These include:
* changes to the formula of resin that binds the composite material to
improve flow;
* reduction in panel thickness to meet weight specifications, and:
* determination of processes to join composite panels together and join
them with the steel foundation of the deckhouse.[Footnote 18]
Process changes are being validated through production and inspection
of a series of test units, culminating with the construction test, and
inspection of a large-scale prototype. According to the shipbuilder,
this prototype is being manufactured to the same thickness and other
specifications of the deckhouse. The Navy will not have final
validation of the manufacturing processes for deckhouse construction
until after construction, inspection, and shock testing of the large-
scale prototype. However, test and inspection activities are not
scheduled for completion until after the deckhouse production readiness
review in September 2008. Problems discovered during testing and
inspection may require additional changes to manufacturing methods.
Further, the shipbuilder has not finalized the quality assurance
standards for production of composite panels manufactured for DDG 1000.
The shipbuilder is adapting the quality assurance processes from those
established for the manufacturing and assembling of composite masts for
the San Antonio-class ships. However, since the composite materials
used differ, the shipbuilder may need to develop alternative tests or
inspection methods to ensure that quality standards are rigorous enough
for DDG 1000.[Footnote 19] For example, the shipbuilder recently traced
apparent panel defects back to errors in calibrating ultrasonic testing
equipment, rather than the panels themselves. Results from testing the
large-scale prototype will provide the shipbuilder a final opportunity
to refine its quality assurance process prior to actually starting to
manufacture panels for the DDG 1000 deckhouse.
The deckhouse represents the Navy's first experience with composite
material at this large of a scale, and the Navy's quality assurance
procedures are still being developed. The Supervisor of Shipbuilding--
the Navy's technical oversight located at the shipyard--recently hired
additional staff with composite experience, and quality assurance
training has just gotten underway.
The Navy recognizes the challenges of deckhouse manufacturing and the
importance of the deckhouse to achieving overall ship construction.
Importantly, the contractor is able to earn a special incentive fee for
deckhouse construction and delivery, but the criteria and metrics for
awarding the fee were not included in the contract and have yet to be
established.
Actual Costs of Ships Will Likely Exceed Budget:
Costs of the DDG 1000 ships are likely to exceed the Navy's fiscal year
2009 budget. In particular, the total cost to deliver a fully
operational ship has not been fully recognized or funded. There is
little margin in the budget to fund any cost increases that could occur
during construction of the lead ships. However, cost growth during lead
ship construction appears likely given technology, design, and
construction risks. Independent estimates are higher than the Navy's
budget. The Navy does not have adequate funding reserves for these
contingencies. Funding problems will not be known until well after the
shipyards begin construction of the lead ships. This is particularly
true under the restructured program that has moved key events later in
the schedule. However, the Navy has already requested funding for the
third ship in this fiscal year, with plans to compete and award a
contract as early as January 2009, including options to build the
remaining four ships. At that time, the Navy will not have enough data
on actual lead ship construction performance to confidently price the
costs of the follow-on ships. By negotiating prices for the follow-on
ships prior to understanding lead ship construction performance, the
Navy risks cost growth that will require offsets in the budget in
future fiscal years. With the program's production decision milestone
scheduled for 2014--after the award of contracts for all of the follow-
on ships--decision makers will miss an important opportunity to
reevaluate whether DDG 1000 still meets the Navy's needs.
Construction Funds Leave Little Margin for Cost Growth and Do Not
Include All Work Necessary for Fully Operational Ships:
Based on the contracted costs for construction of the lead ships, the
remaining budget is not sufficient to pay for cost growth that will
likely occur. The Navy estimates a total shipbuilding budget of $6.3
billion, including almost $2.4 billion for the ships' combat systems,
$2.6 billion for building the lead ships, $910 million for detail
design, $216 million for other costs (including acquisition support),
and $258 million for change orders--that is, funding to compensate for
unknowns during design and construction. The Navy already recognizes
that there is no additional funding in the budget to pay for any
unanticipated costs.
Lead Ship Costs Appear Likely to Increase Given Risks:
Construction of the lead ships will likely be challenging and the risk
of cost growth is high. If key systems are delivered late or do not
function as planned, additional labor hours will be needed to work
around missing systems or will be spent reworking defective components.
Further, the shipbuilders may have difficulties manufacturing unique
ship components. For example, if composite manufacturing for the
deckhouse takes longer than expected, ship construction could be
delayed and costs associated with manufacturing will likely rise. Added
to this are the challenges of integrating sections of the ship built in
separate shipyards. Since the Navy is responsible for the delivery of
government-furnished equipment, the Navy will bear the costs if there
are schedule delays that affect another contractor. However, these
problems will not surface until well after the shipyards have begun
construction of the lead ships.
Our prior work has shown that cost growth for recent lead ships has
been on the order of 27 percent--partly because the total effort needed
to build new designs and incorporate new technologies is not yet
understood. With ships as expensive as DDG 1000, even a small
percentage of cost growth could lead to the need for hundreds of
millions of dollars in additional funding.
Some Costs Will Not Be Budgeted until Later in the Construction Phase:
As of June 2008, the Navy has approximately $363 million in unobligated
funds to cover its outstanding costs and manage any cost growth. Even
without cost growth, this appears inadequate to fund work necessary to
complete the two lead ships. Table 5 summarizes costs that still need
to be funded.
Table 5: Unfunded Lead Ship Expenses:
Estimated value:
Expense: Deferred ship construction scope;
Status: Work removed from scope of construction contract to stay within
construction budget. Since this work is necessary to meet ship
specifications, the Navy plans to perform and fund work sometime after
the lead ships are delivered. Includes the following:
* Windows and enclosures for certain sensors;
* Special hull treatment;
* Deck coverings that comply with the ship's radar cross section
requirements;
* Secondary hull sheathing;
* Anchor handling system;
Estimated value: $85 million.
Expense: Contract price adjustments;
Status: Construction contracts structured to allow price adjustments
based on future events that were considered largely outside of the
shipbuilders' control. Adjustments reduced the shipbuilders' risk
premium allowing a lower initial contract price. Includes the
following:
* Shifts in future workload;
* Escalations in future rates;
* Changes in the price of raw materials, such as steel and copper;
Estimated value: Not available.
Expense: Deferred procurement of select combat systems;
Status: Purchase and installation are not yet under contract for the
following systems:
* Volume search radar aperture and other components;
* Vertical launch system electronics, cell adapters, uptakes, and
junction boxes;
* Thirty-four external communications antennas and apertures per ship;
The contractor estimate of these costs is approximately $763 million;
the Navy estimates approximately $200 million for both ships;
Estimated value: $264 million to $767 million.
Expense: Deferred activation of combat systems;
Status: Funds also not obligated toward light-off and final shipboard
testing of the combat systems. The Navy estimates as much as $64
million for both ships, including about $4 million in costs for
activation to be provided to the shipbuilders. Contractor and
shipbuilder estimates may be higher.
Source: GAO analysis of Navy and contractor data.
[End of table]
Partly to ensure that there was enough funding available in the budget
to cover the costs of building the lead ships, the Navy negotiated
contracts with the shipbuilders that shifted costs or removed planned
work from the scope of lead ship construction and reduced the risk
contingency in the shipbuilders' initial proposals. For example, the
Navy stated that it shifted in excess of $100 million associated with
fabrication of the peripheral vertical launch system from the scope of
ship construction and funded this work separately using research and
development funding.[Footnote 20]
The Navy has not contracted or funded work associated with activating-
-and in some cases purchasing and installing--many of the ship's key
systems. The Navy plans to leave funding uncommitted in order to create
a cash reserve to pay for any cost increases that may occur during
construction of the lead ships. Officials stated that by not committing
all available funding, the Navy will be able to more quickly divert
resources if problems occur during construction--thus avoiding
disruptions to construction while it obtains additional funding. Even
before ship construction is under way, the Navy does not have adequate
reserves to pay for work that has been deferred, is not under contract,
or is only partially funded. The Navy has $363 million in funding
remaining for the two lead ships. However, the Navy's known obligations
for the lead ships range from $349 million to $852 million. The main
discrepancy is the current estimated cost of the combat systems. The
Navy has estimated the cost of the combat system to be around $200
million, while the contractor's estimate is over $760 million. If the
agreed-on cost approaches the contractor's estimate, the Navy will not
have enough in its remaining funds to cover the cost. To the extent
that the lead ships experience cost growth beyond what is already
known, more funding will be needed to produce operational ships. Funds
could be provided through additional funding to the ships' budget line
or, once the ship is initially delivered, using outfitting and post-
delivery funding. After delivery, the Navy could also use other types
of funds such as support equipment and materials funding that are
accounted for separately--outside of the ship's end cost.[Footnote 21]
Estimate for the Third Ship Will Not Reflect Knowledge Gained from
Construction of the Lead Ships:
Cost uncertainty is high for new classes of ships--including early
follow-ons.[Footnote 22] Cost estimates become more certain only as
actual costs begin to replace earlier estimates. While the budgets for
DDG 1000 ships benefit from a better understanding of class design,
lengthy ship construction timelines have led the Navy to request
authorization for follow-on ships without sufficient understanding to
provide confidence in the lead ships' true cost.
Since the lead ships are not yet under construction, the Navy derives
the budget for construction of the follow-on ships from cost estimates
of the two lead ships, not actual costs. The Navy initially developed
these estimates in 2005 in support of the program's Milestone B
decision,[Footnote 23] updating its current budget submissions to
account for inflation changes and residual effects of Hurricane
Katrina. The Navy based its construction estimate largely on the
experiences of the DDG 51 class--the most closely related ship class,
adjusting these costs to account for systems unique to DDG 1000
construction and adding a premium for building the lead ships.[Footnote
24] Independent cost analysts within DOD estimate that another $878
million will be needed to build the lead ships. The cost of follow-on
ship construction is based on the lead ship estimate adjusted downward
to account for assumed efficiency gains through building each
successive ship at the shipyard. As a result, achieving the budget for
the follow-on ships depends on achieving lead ship costs--which appears
unlikely.
Navy analyses indicate the potential for cost growth on follow-on
ships. The Navy conducted risk and uncertainty analyses to measure the
probability of cost growth and established a 50 percent confidence
level for the budgets of the follow-on ships, recognizing that there is
only a 50 percent chance that these ships will achieve their estimated
costs. Independent cost analysts within DOD also estimate higher ship
construction costs. According to this estimate, an additional $2.3
billion in total construction costs across all five follow-on ships may
be needed.
The Navy plans to contract for the follow-on ships before little if any
knowledge of actual cost is gained on construction of the lead ships.
The Navy intends to compete the award for construction of the next five
ships between Bath Iron Works and Northrop Grumman Shipbuilding, with
the winning shipyard receiving the larger quantity of ships. Under this
approach, the Navy would award fixed-price-type contracts for the third
ship (DDG 1002), with priced options for the construction of the
remaining four ships. Based on the current schedule, the Navy would
award follow-on ship contracts as early as January 2009. At that time,
Bath Iron Works will have demonstrated less than 3 months of
construction experience and Northrop Grumman Shipbuilding's lead ship
will not have even begun construction. If the schedule for construction
of the lead ships slips further into fiscal year 2009 and fiscal year
2010, neither shipyard will have demonstrated construction experience
prior to contract award. The Navy will not have production cost trends
from either shipbuilder until late fiscal year 2009, at the earliest.
Consequently, the Navy will have little information on both shipyards'
construction performance on which to base not only the cost of the
third ship, but that of the remaining follow-ons. We have previously
reported that negotiating prices for follow-on ships prior to gaining
knowledge of lead ship construction costs contributes to cost growth on
follow-on ships.[Footnote 25]
Timing of Production Decision Milestone Does Not Allow for Meaningful
Oversight of Follow-on Ship Costs:
A Milestone C decision review is intended to represent DOD's commitment
to producing a system. However, in the DDG 1000 program, this review is
currently scheduled for 2014 after all ship contracts are awarded and
follow-on ship construction is well under way (see fig. 11). Thus, it
will have no decision to consider.
Figure 11: DDG 1000 Oversight, Ship Contract Award, and Construction
Events after Milestone B:
[See PDF for image]
This figure is an illustration of DDG 1000 Oversight, Ship Contract
Award, and Construction Events after Milestone B, as follows:
Milestone B: late 2005;
Program review: late 2006;
Program review: late 2007;
Ship contract award: 2008;
Ship start fabrication: mid-2008;
Ship contract award: 2009;
Ship start fabrication: mid-2009;
Ship contract award: 2010;
Ship start fabrication: mid-2010;
Ship contract award: 2011;
Ship start fabrication: mid-2011;
Ship contract award: 2012;
Ship start fabrication: mid-2012;
Ship contract award: 2013;
Ship start fabrication: mid-2013;
Milestone C: 2015.
Source: Navy data.
[End of figure]
In most weapon system programs, a Milestone C decision review occurs in
advance of awarding contracts to begin production.[Footnote 26] At this
review, the Under Secretary of Defense (Acquisition, Technology &
Logistics) authorizes the program to begin producing an initial
quantity of units (known as low rate initial production). In the DDG
1000 program, the Milestone B decision essentially authorized detail
design as well as production. While the Navy anticipates an annual
program review with decision makers within DOD, these reviews do not
have the same rigor as that associated with a milestone review.
Typically, entrance into Milestone C requires that programs prove that
they have appropriate knowledge to begin production, which includes
demonstrating that:
* manufacturing processes are under control,
* software capability is mature;
* the system is affordable throughout its life cycle, and:
* the system is realistically funded.
In preparation for the milestone, programs submit documents for well
over 10 information requirements, including documentation of readiness
for production and updated independent cost estimates. Such documents
are intended to help demonstrate that the product (in this case a ship)
can be produced within cost and according to requirements. As a result
of the program's schedule for DDG 1000, decision makers will not have
the advantage of the knowledge gained through these analyses at the
time the next five ships are authorized, lessening decision makers'
ability to recognize and make any needed adjustments in the remainder
of the program. Officials from DOD point out that Milestone C for ships
is normally scheduled after the lead ship passes operational testing.
In the case of the DDG 1000, the point occurs in fiscal year 2014 after
all key acquisition commitments for the class will be made. DOD
acquisition guidance (DOD 5000.2) does not prohibit shipbuilding
programs from holding Milestone C earlier in the production phase.
Conclusions:
The DDG 1000 program has from the outset faced a steep challenge framed
by demanding mission requirements, stealth requirements, less than half
the manning levels of predecessor destroyers, and a split construction
schedule planned around managing the shipbuilders' workload needs.
These requirements translated into significant technical and design
challenges. To meet these multiple and somewhat conflicting demands,
the Navy conceived a thoughtful acquisition strategy to (1) develop key
systems, (2) mature the design before starting to build the ship, and
(3) facilitate advance preparation at the shipyards for construction of
the lead ships. As the program progressed, the Navy achieved
developmental successes and made a number of adroit decisions, such as
switching to a more mature power system and changing its contracting
strategy. Several of the ship's key systems that depended on new
technologies were successfully demonstrated. The design approach, while
not yet completed, has already produced more design knowledge than on
the LPD 17 and LCS-class ships. Several construction innovations have
been sponsored at each of the shipyards.
Yet at this point--the first year of a 6-year construction schedule for
lead ships--the Navy may have exhausted its options for solving DDG
1000's potential problems without adding time and money. Recently, the
Navy deferred some key efforts, such as demonstrating the combat
systems, to later in the construction schedule. While the rescheduling
provides benefits with regard to testing the ship systems separately
from the mission systems, it was also a necessity given that the power
generation system, the dual band radar, and the ship's software are
each taking longer than planned. While these may be pragmatic decisions
to solve problems while minimizing schedule delays, they are also
compromises to original plans that called for systems to be tested
before installation. It is not the problems themselves that are of
concern, as they are to be expected. Rather, the fact that these
compromises are already being made before construction has begun
suggests that there is very little margin for solving future problems.
Given DDG 1000's complexity and unique features, along with the design
work, testing, and actual construction experience yet to come, the
discovery of future problems is likely.
Similarly, the $6.3 billion the Navy has budgeted for the two lead
ships provides little margin for error and is likely to be inadequate
to cover the cost of the first two ships. Cost growth during
construction for lead ships has historically been about 27 percent, and
an independent estimate already projects the cost of the two lead ships
to be $878 million higher than the Navy's budget. Moreover, the Navy's
redefinition of ship delivery, including the deferral of combat system
work until later in the construction schedule, will delay realization
of actual costs and may make it difficult to reconcile the $6.3 billion
budget with the actual cost of the ships.
The risks associated with the lead DDG 1000 ships should not carry over
to the follow-on ships. While a legitimate debate can be had on how
much knowledge should be gained on lead ships before contracting for
follow-on ships, in our view, it is not debatable that the current plan
to contract for the five remaining DDG 1000s in January 2009 is too
soon.[Footnote 27] Very little construction experience will have been
gained at that point to inform the cost estimates for the follow-on
ships. At the very least, a review by the Office of the Secretary of
Defense, with the same quality of information normally available at a
Milestone C decision, is warranted before the commitment and the
pricing are set for follow-on ships.
Recommendations for Executive Action:
We recommend that the Secretary of Defense take the following three
actions:
* Require the Navy to complete product modeling of the ship's design to
the level currently planned before the start of construction.
* Defer contract award for follow-on ships until the Navy has completed
a substantial amount of construction on the lead ships.
* Hold the Milestone C review in advance of awarding a contract for the
third ship.
Matter for Congressional Consideration:
To provide insight into the potential for cost growth as the program
progresses, the Congress may want to consider requiring the Navy to
report on (1) the current production and testing schedule for systems
necessary to meet ship light-off, ship delivery, and combat system
activation; (2) any changes to this schedule, particularly with the
dual band radar and the total ship computing environment; and (3) the
cost impact of these changes if the schedule is maintained and if the
schedule is stretched out.
Agency Comments and Our Evaluation:
Regarding our recommendation to require the Navy to complete product
modeling of the ship's design to the level currently planned before the
start of ship construction, DOD reiterated that the plan is to be
approximately 85 percent complete and noted that this is a higher
percentage of design completion than any other surface warship. We
agree that the Navy's plan is good, but a plan does not mean actual
attainment. The Department should be prepared to react should the
program's actual performance lag behind planned performance.
Regarding our recommendation that the Department hold the Milestone C
review in advance of awarding a contract for the third ship, DOD stated
that a review by the Defense Acquisition Executive is necessary prior
to awarding a contract for construction of the third ship, but does not
agree that a Milestone C review provides any additional benefit for the
program. According to DOD, the Milestone C review will only occur after
the lead ships complete operational testing and only if additional
ships above the current seven ships are added. While DOD acquisition
policy provides some flexibility for shipbuilding programs, the policy
applying to most major defense acquisitions calls for the Milestone C
review to authorize production of low rate initial quantities. Annual
program reviews will certainly enhance program knowledge, but they do
not provide the consistent framework and documentation that a Milestone
C review would provide, such as an updated independent cost estimate.
Nonetheless, it is the quality of the review rather than its title that
is important. Thus, the review that precedes awarding the contract for
the third ship should provide the framework and documentation of the
caliber that a milestone carries.
DOD did not agree with our recommendation to defer contract award for
follow-on ships until a substantial amount of construction on the lead
ships has been completed. DOD stated that delaying contract award and
subsequent procurement of the third and follow-on ships will impact
warfighting capability gaps, cost, and the shipbuilding industrial
base. We note that the Navy and DOD are currently considering canceling
the third and follow-on ships. This notwithstanding, we believe that
the risks of prematurely awarding follow-on ship contracts are
considerably higher than any potential cost increase that could occur
as a result of delaying funding for a follow-on ship. Industrial base
effects have likely been offset by the later-than-planned construction
start of the lead ships. Moreover, the Navy does not plan to start
construction of the third ship before fiscal year 2010 and advanced
procurement for its long-lead items has already been funded. The
program is moving toward lead ship construction with significant
uncertainties. In the past, it has proven costly to award contracts for
construction of follow-on ships with little or no actual performance
data on the lead ship--especially on a ship that includes so many
"firsts" in shipbuilding. With such limited construction planned by
contract award, the shipbuilders cannot incorporate lessons learned
from the first ship to inform--and reduce the risk premium in--their
estimates of follow-on ship costs.
DOD's written comments are reprinted in appendix II. The Department
also provided technical comments, which were incorporated into the
report as appropriate.
We are sending copies of this report to the Secretary of Defense, the
Secretary of the Navy, and interested congressional committees. Copies
will also be made available to others on request. In addition, the
report will be available at no charge on GAO's Web site at [hyperlink,
http://www.gao.gov].
If you or your staff have any questions concerning this report, please
contact me at (202) 512-4841 or francisp@gao.gov. 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 IV.
Signed by:
Paul L. Francis:
Director Acquisition and Sourcing Management:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
This report assesses:
1. the Navy's ability to deliver key systems when needed to support
lead ship construction and whether these systems will work as intended,
2. the level of design maturity and the obstacles to completing the
ship's design as planned,
3. the shipyards' readiness to build DDG 1000, and:
4. the challenges of building the ships (lead and follow-ons) within
budget.
To address the first objective, we reviewed technology development
plans, scope and schedules and compared these against similar
documentation for 2005 when the Navy held its Critical Design Review
and milestone review. We also analyzed test plans and schedules as well
as test reports to determine progress in demonstrating key systems.
We also assessed the degree of testing and hardware and software
integration planned prior to installation and identified areas of
testing and installation concurrency. We analyzed, among other
documents, DDG 1000 Cost Analysis Requirements Description, DDG 1000
Program Review briefings, technology test plans and reports, software
development schedules, critical design review findings, risk matrixes,
Defense Contract Management Agency reports, and contractor cost
performance reports for key technologies. To supplement our analysis,
we visited contractors and test sites where the ship's major
technologies are being developed and tested.
To address the second objective, we reviewed the Navy's plans and
guidelines for design management and completion, evaluated the design
progress and schedule, and assessed impacts of the product model on
that progress, including schedule delinquencies and key dates. We
applied established earned value management techniques to data captured
in the shipbuilders' cost performance reports in order to evaluate
design performance to date. Further, we analyzed variance reports that
accompany the cost performance reports in order to understand the
drivers of schedule delays. We compared DDG 1000's design process with
our knowledge-based acquisition methodology and past work on
shipbuilding programs. We analyzed and compared DDG 1000 design metrics
with the experience of previous shipbuilding programs, particularly the
DDG 51 class and the Virginia-class submarine. In conducting our
analysis, we examined key documents, including the Integrated Master
Schedule, Configuration Management Plan, and Supervisor of
Shipbuilding, Conversion and Repair's reports.
To address the third objective, we reviewed key shipyard performance
data. We examined recent cost performance reports for the DDG 51 class
under construction at both shipyards. We reviewed key indicators of
labor productivity, including data on Northrop Grumman Shipbuilding's
progress in recovering its labor and facilities from the effects of
Hurricane Katrina. We examined both shipyards' plans and progress
improving facilities to meet the schedule and manufacturing capacity
required to construct DDG 1000. Among other documents, we reviewed
facility improvement schedules compared with construction schedules,
labor metrics, shipbuilder cost performance reports, contract
documentation, and Supervisor of Shipbuilding, Conversion and Repair
reports. To supplement our analysis we visited both shipyards and
toured their facilities.
To address the fourth objective, we examined the Navy's budget request
and cost estimates for the ships in the class. We analyzed the Navy's
cost estimates by examining the Program Life Cycle Cost Estimate and
updates to that estimate since the Milestone B decision in 2005. We
compared the Navy's cost estimates with estimates from the Department
of Defense's independent cost analysts. We assessed the potential for
lead ship cost growth by analyzing the program's current budget
obligations by budget element and contract for each of DDG 1000's prime
contractors. We compared total contract estimates of the total cost to
complete the ship with the program's total remaining funding. Further,
we analyzed the shipbuilders' contracts and the Navy's business
clearance memorandums to understand the scope and cost of lead ship
construction. We evaluated follow-on ship costs by analyzing the
program's acquisition strategy. We compared DDG 1000's strategy with
DOD acquisition policy and guidance and our past work on shipbuilding
cost growth.
In conducting our analysis, we held discussions and attended briefings
with the shipbuilders and contractors responsible for DDG 1000
development; as well as officials from the Naval Sea Systems Command,
including the DDG 1000 Program Office; Program Executive Office,
Integrated Warfare Systems; Program Executive Office for Ships; Ship
Design Integration and Engineering Division; Cost Engineering and
Industrial Analysis Division; Contracts Division; Naval Surface Warfare
Center, Carderock in Pennsylvania and Maryland; Supervisor of
Shipbuilding, Conversion and Repair, Gulf Coast; and Supervisor of
Shipbuilding, Conversion and Repair, Bath. In addition, we interviewed
officials from the Navy's Commander of Operational Test and Evaluation
Force and DOD's Cost Analysis Improvement Group; Defense Contract Audit
Agency in Bath, Maine and Pascagoula, Mississippi; and Defense Contract
Management Agency in Tewksbury, Massachusetts, and Fort Snelling,
Minnesota.
We conducted this performance audit from September 2007 to July 2008 in
accordance with generally accepted government auditing standards. Those
standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.
[End of section]
Appendix II: Comments from the Department of Defense:
Note: Page numbers in the draft report may differ from those in this
report.
Office Of The Under Secretary Of Defense:
Acquisition Technology And Logistics:
3000 Defense Pentagon:
Washington, DC 20301-8000:
July 28, 2008:
Mr. Paul L. Francis:
Director, Acquisition and Sourcing Management:
U.S. Government Accountability Office:
441 G Street, N.W.
Washington, DC 20548:
Dear Mr. Francis:
This is the Department of Defense (DoD) response to the GAO draft
report, GAO-08-804, "Defense Acquisitions: Cost to Deliver Zumwalt-
Class Destroyers Likely to Exceed Budget," dated June 23, 2008 (GAO
Code 120672).
The DoD concurs with recommendation 1, non-concurs with recommendation
2, and partially concurs with recommendation 3. The Department's
comments on the recommendations are enclosed.
We appreciate the opportunity to comment on the draft report. Technical
comments were provided separately. For further questions concerning
this report, please contact Ms. Darlene Costello, Deputy Director,
Naval Warfare, at 703-697-2205 or Darlene.Costello@osd.mil.
Sincerely,
Signed by:
David G. Ahern:
Director:
Portfolio Systems Acquisition:
Enclosure: As stated:
GAO Draft Report Dated June 23, 2008:
GAO-08-804 (GAO CODE 120672):
"Defense Acquisitions: Cost To Deliver Zumwalt-Class Destroyers Likely
To Exceed Budget"
Department Of Defense Comments To The GAO Recommendations:
Recommendation 1: The GAO recommended the Secretary of Defense require
the Navy to complete product modeling of the ship's design to the level
currently planned before the start of construction. (p. 51/GAO Draft
Report)
DOD Response: Concur. DDG 1000 detail design is currently planned to be
approximately 85 percent complete prior to the start of fabrication.
The ship's detail design effort will be more complete at the start of
construction next year than any other previous surface warship. Detail
Design of the zones that will begin fabrication on the start
fabrication date will be 100 percent complete and will have undergone 2-
dimensional drawing extraction which produces the drawings that are
used for actual construction of the ship. All other zones, as they
reach their individual start fabrication dates, also will be 100
percent complete with Detail Design.
Design of the Mission Systems now is nearly 100 percent complete and a
Production Readiness Review for each system is planned prior to the
ship start fabrication date.
In the particular case of DDG 1000, the ship class has been in design
and development for almost six years. The Navy successfully built and
tested the 10 critical technologies that provide the capabilities that
future ships need on cost and schedule.
The DDG 1000 shipbuilders, General Dynamics Bath Iron Works (BIW) and
Northrop Grumman Shipbuilding (NGSB), have been involved in the
development of the ship design since preliminary design began in 2002.
This has contributed to the efficient completion of detail design. In
addition, the CATIA V5/ENOVIA Computer Aided Design tool in use at both
shipyards has been proven to be mature, complete, and stable and the
American Bureau of Shipping Naval Vessel Rules are fully incorporated
into the DDG 1000 detail design.
BIW and NGSB have been using a proven set of common Computer-Aided
Design (CAD) tools, incorporating design, simulation, visualization and
material sourcing capabilities. All vendor furnished information (VFI),
a common source of design disruption and rework, already has been
delivered. By comparison, DDG 51 design was much less mature at the
start of fabrication, much of the VFI was preliminary or not
available, and the CAD tool failed early in detail design, forcing the
shipbuilder to design by hand.
Start Fabrication is a risk-based assessment of adequate design
maturity versus consequences of delaying fabrication. The Government
uses design and construction contracts to facilitate the most cost
effective design-build sequence to maximize the efficient use of
shipyard resources. The consequences of delaying Start Fabrication
include:
* Increased cost due to schedule extension.
* Increased inflation escalation costs. Delaying procurement of
material generally leads to buying material at inflated price with no
benefit gained from delay.
* Disruptions in workforce. Staggered start of zones smoothes the
workload manning curves and reduces construction costs. Early start
zones are the least complex zones with low risk of cost growth.
Recommendation 2: The GAO recommended the Secretary of Defense defer
contract award for follow-on ships until the Navy has completed a
substantial amount of construction on the lead ships. (p. 51/GAO Draft
Report)
DOD Response: Non-concur. Deferring the contract award, and subsequent
procurement, of the third and follow DDG 1000 class ships will impact
warfighting capability gaps, cost, and the shipbuilding industrial
base. The cost of the first two ships, now under contract to General
Dynamics Bath Iron Works and Northrop Grumman Shipbuilding, will
increase significantly due to the lack of the shipbuilders' ability to
spread shipyard overhead cost among multiple ships. Additionally, the
cost of the mission systems equipment for the lead ships will increase
for a similar reason. The deferral of this workload also would likely
impact costs on other Navy contracts at these shipbuilders. Finally,
the lack of an FY 2009 ship will impact the shipbuilding industrial
base and workforce stability.
Each shipbuilder's lead-ship cost proposal was based on a seven-ship
program of record DDG 1000 workload. Deferring follow ship contract
awards would affect vendors that currently are under contract and
building class-specific systems and components, including systems such
as the Dual Band Radar (Raytheon/Lockheed Martin), Advanced Gun System
(BAE System), Integrated Power System (Converteam/DRS Technologies),
Advanced Vertical Launch System (Raytheon/BAE System) and Total Ship
Computing Environment Infrastructure (Raytheon). Hundreds of system and
component vendors employing thousands of people in 48 states also would
be impacted.
Recommendation 3: The GAO recommended the Secretary of Defense hold the
Milestone C review in advance of awarding a contract for the third
ship. (p. 51/GAO Draft Report)
DOD Response: Partial Concur. The Department agrees that a Defense
Acquisition Executive (DAE) review of the program is necessary prior to
the Navy awarding the construction contract for the third ship, DDG
1002. An Acquisition Decision Memorandum (ADM) signed on February 13,
2008, requires the Navy to return for a Defense Acquisition Board (DAB)
level program review prior to obligating any additional procurement
funding for construction of the DDG 1002 ship. The ADM also requires
the Navy to demonstrate that the negotiated cost for the DDG 1002 ship
is within budget and to propose exit criteria for the DAB reviews for
the DDG 1003 and DDG 1004 follow ships.
The Department does not agree that a Milestone C review, as defined in
the DoD Instruction 5000.2, provides any additional benefit for a
shipbuilding program such as the DDG 1000 program, when timed at the
first follow ship. Authorization to enter the production phase of the
program was granted by the DAE in an ADM signed on December 22, 2007.
This followed an extensive review of the readiness of the program to
start production of the lead ships. The program documentation updates
required at the Milestone C review will be directed as the DAE deems
appropriate during the production phase of the program. The Department
believes that conducting annual reviews of the program, prior to
awarding the contract for a ship in that year is a more practical
approach to oversight of the DDG 1000 program. The Department tailored
the Defense Acquisition System in the case of the DDG 1000 program and
will use the Milestone C event to authorize construction of any
additional ships that might be added above the current program of
record of seven ships. That tailored Milestone C review would occur
only after the lead ships complete Initial Operational Test and
Evaluation and the Director, Operational Test and Evaluation issues its
statutory Beyond Low Rate Initial Production report. The Milestone C
review would serve as the Full Rate Production Decision Review.
[End of section]
Appendix III: Major Events in the Development of DDG 1000:
Year: 1994;
Events:
* Navy identifies need for a new 21st century surface combatant (SC 21)
to provide naval surface fire support.
Year: 1997;
Events:
* Plans for DD 21 include 32 ships with an average unit cost not to
exceed $921million (FY 1996 dollars) and an initial operating
capability of fiscal year 2008.
Year: 2001;
Events:
* DD 21 program restructured to emphasize technology development and
affordability and renamed DD(X).
Year: 2002;
Events:
* DD(X) program starts Preliminary Design/Development;
* Navy completes its review of requirements for future surface
combatants and recommends a force structure of 16 DD(X) each with 2
advanced gun systems, 900 land attack munitions, and 96 missile cells;
* Navy recognizes the need for an additional $7.6 billion to complete
DD(X) technology development, testing, and evaluation efforts.
Year: 2004;
Events:
* Program restructured to include 8 ships, an average unit cost not to
exceed $2.6 billion (FY 1996 dollars) and an initial operating
capability of fiscal year 2013;
* DD(X) program completes its preliminary design review, initiating
system design.
Year: 2005;
Events:
* Navy approves a quantity of 10 ships, an average unit cost not to
exceed $3.1 billion (FY 1996 dollars) and an initial operating
capability of fiscal year 2014;
* The Congress prohibits the Navy from a "Winner Take All" competition
in Emergency Supplemental Appropriations Act;
* Milestone B review approves system development and demonstration and
approves Navy's strategy for purchasing dual lead ships.
Year: 2006;
Events:
* Navy outlines its 30-year shipbuilding plan and lists quantity of
seven DD(X) destroyers;
* Navy re-names DD(X) program DDG 1000-class destroyer and confirms
that the first ship in the class, DDG 1000, is to be named the Zumwalt,
in honor of Admiral Elmo R. Zumwalt, the Chief of Naval operations from
1970 to 1974;
* The Congress authorizes split funding of two lead ships but states
that procurement cannot exceed $6.53 billion;
* Navy awards dual lead detail design contracts to Northrop Grumman
Shipbuilding and Bath Iron Works.
Year: 2007;
Events:
* Navy resequences construction of the first lead ship from Northrop
Grumman Shipbuilding to Bath Iron Works.
Year: 2008;
Events:
* Navy negotiated construction contract modifications with shipbuilders
(February 14, 2008);
* DDG 1000 construction start (currently scheduled for October 2008);
* Contract modification with Raytheon for production of key systems for
lead ships (estimated definitization August 2008).
Year: 2009;
Events:
* Navy plans to compete and award construction contract award for the
ship with priced options for fiscal year 2010 through 2013 ships;
* DDG 1001 construction start (currently scheduled for September 2009).
Year: 2013;
Events:
* DDG 1000 initial delivery (April 2013).
Year: 2014;
Events:
* DDG 1000 final ship delivery;
* DDG 1001 delivery (May 2014).
Year: 2015;
Events:
* Initial operating capability (March 2015).
Source: Navy data.
[End of table]
[End of section]
Appendix IV: GAO Contact and Staff Acknowledgments:
GAO Contact:
Paul L. Francis, (202) 512-4841 or francisp@gao.gov:
Acknowledgements:
In addition to the contact named above, key contributors to this report
were Karen Zuckerstein, Assistant Director; Marie P. Ahearn; Erin
Carson; Raj Chitikila; Diana Moldafsky; and Gwyneth B. Woolwine.
[End of section]
Footnotes:
[1] Current budgets refer to the Navy's fiscal year 2009 President's
budget submission for the DDG 1000 program.
[2] National Defense Authorization Act for Fiscal Year 2006, Pub. L.
No. 109-163, Div. A § 123 (2006).
[3] The Navy describes the delivery in April 2013 as transfer of
ownership and hull delivery; for purposes of this report, we refer to
this as initial delivery. We use the term final delivery to indicate
when ship construction is complete--including the installation and
integration of the combat systems.
[4] The system requirements are made up of 978 verification items, of
which nearly 40 percent either failed acceptance tests or were not
tested in this release.
[5] JAVA is a trademark or registered trademark of Sun Microsystems,
Inc. in the United States and other countries.
[6] The Defense Contract Management Agency provides contract management
and surveillance. The Navy's Operational Test and Evaluation Force
provides independent evaluation of the operational effectiveness and
suitability of Navy programs. These agencies provide periodic,
independent evaluations of the DDG 1000 program's progress.
[7] The certification panel is made up of technical experts from within
the Navy.
[8] The antiterror/force protection requirement was added in 2005,
after the Congress required the Navy to protect ships from close-range
terror attacks such as the one on USS Cole in Yemen in 2000.
[9] The product model uses a shared server managed by Northrop Grumman
Shipbuilding with contract requirements for equal server availability
at both shipyards.
[10] The two-dimensional drawings include build strategies,
construction drawings, and final lists of needed construction
materials.
[11] These tasks are general or supportive in nature and do not produce
definite end products or results. Performance of these tasks is not an
accurate gauge of the overall design effort; since level-of-effort
tasks assume that staff will attain expected results, performance
equals the scheduled cost for those tasks and no schedule variances can
appear to occur. See Department of Defense (DOD) Earned Value
Management Implementation Guide § 2.2.3.4.3 (Oct. 2006).
[12] For example, if the scheduled work for a period of time included
the tasks leading up to a 90 percent review but not all of these tasks
were actually completed during this period, the scheduled performance
index would be less than one.
[13] As of May 2, 2008, the Navy had completed 16 out of the 100 design
zones.
[14] See GAO, Defense Acquisitions: Improved Management Practices Could
Help Minimize Cost Growth in Navy Shipbuilding Programs, [hyperlink,
http://www.gao.gov/cgi-bin/getrpt?GAO-05-183] (Washington, D.C.: Feb.
28, 2005) and Defense Acquisitions: Realistic Business Cases Needed to
Execute Navy Shipbuilding Programs, [hyperlink, http://www.gao.gov/cgi-
bin/getrpt?GAO-07-943T] (Washington, D.C.: July 24, 2007).
[15] According to the shipbuilder, DDG 51-class ships currently under
construction include 103, 105, 107, and 110.
[16] According to the shipbuilder, the composite facility--located in
Gulfport, Mississippi--can compete more effectively in the area economy
for workers because it is air-conditioned, providing a favorable
environment.
[17] The facility has previously manufactured the hulls of
decommissioned Osprey-class coastal mine hunters and the mast for a
Spruance-class destroyer, and is currently manufacturing the masts for
the San Antonio-class amphibious ships (LPD 17).
[18] The method for joining composite to steel is particularly
important. A joint that uses fasteners only carries risk because
properties of the carbon fiber and fasteners may lead to corrosion and
the fasteners are expensive.
[19] While both are composite, the DDG 1000 deckhouse contains carbon
fiber instead of fabric and a core of balsa wood throughout instead of
foam and other materials.
[20] By shifting these costs the Navy stated that it could use
research, development, test, and evaluation (RDT&E) funding instead of
procurement funding. However, this may lead to increases in the RDT&E
budget.
[21] This is known as Other Procurement, Navy.
[22] Recently, the Navy discussed canceling the remaining five ships of
the class with members of the Congress.
[23] A Milestone B decision represents DOD's commitment to design and
develop a system. The Milestone B estimate was also updated to reflect
the decision to procure two lead ships.
[24] Cost analysts adjusted the estimate to account for the integrated
power system, peripheral vertical launch system, and composite
deckhouse.
[25] See [hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-05-183], p.
23.
[26] In most shipbuilding programs, authorization for production of the
lead ship occurs at Milestone B.
[27] Recently, the Navy discussed canceling the remaining five ships of
the class with members of the Congress.
[End of section]
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 [hyperlink, http://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 [hyperlink, http://www.gao.gov]
and select "E-mail 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: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]:
E-mail: fraudnet@gao.gov:
Automated answering system: (800) 424-5454 or (202) 512-7470:
Congressional Relations:
Ralph Dawn, Managing Director, dawnr@gao.gov:
(202) 512-4400:
U.S. Government Accountability Office:
441 G Street NW, Room 7125:
Washington, D.C. 20548:
Public Affairs:
Chuck Young, Managing Director, youngc1@gao.gov:
(202) 512-4800:
U.S. Government Accountability Office:
441 G Street NW, Room 7149:
Washington, D.C. 20548:
