Unmanned Aerial Vehicles: Changes in Global Hawk's Acquisition	 
Strategy Are Needed to Reduce Program Risks (05-NOV-04, 	 
GAO-05-6).							 
                                                                 
Global Hawk offers significant military capabilities to capture  
and quickly transmit high-quality images of targets and terrain, 
day or night, and in adverse weather--without risk to an onboard 
pilot. Global Hawk first flew in the late 1990s as a demonstrator
and supported recent combat operations in Afghanistan and Iraq.  
In 2001, the Air Force began an acquisition program to develop	 
and produce improved Global Hawks. In 2002, the Department of	 
Defense (DOD) restructured and accelerated the program to include
a new, larger and more capable air vehicle. GAO was asked to	 
review the program and discuss (1) the restructuring's effect on 
the Air Force's ability to deliver new capabilities to the	 
warfighter and (2) whether its current business case and	 
management approach is knowledge-based and can help forestall	 
future risks.							 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-6						        
    ACCNO:   A13312						        
  TITLE:     Unmanned Aerial Vehicles: Changes in Global Hawk's       
Acquisition Strategy Are Needed to Reduce Program Risks 	 
     DATE:   11/05/2004 
  SUBJECT:   Air Force procurement				 
	     Cost analysis					 
	     Defense capabilities				 
	     Future budget projections				 
	     Military aircraft					 
	     Military intelligence				 
	     Military research and development			 
	     Program management 				 
	     Research and development costs			 
	     Risk management					 
	     Program evaluation 				 
	     Concurrent development and production		 
	     Unmanned aerial vehicles				 
	     Global Hawk Unmanned Aerial Vehicle		 
	     RQ-4B Unmanned Aerial Vehicle			 
	     RQ-4A Unmanned Aerial Vehicle			 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-05-6

     

     * Report to the Chairman, Subcommittee on Tactical Air and Land Forces,
       Committee on Armed Services, House of Representatives
          * November 2004
     * UNMANNED AERIAL VEHICLES
          * Changes in Global Hawk's Acquisition Strategy Are Needed to
            Reduce Program Risks
     * Contents
          * Results in Brief
          * Background
               * Differences between the Two Global Hawk Models
          * Restructured Global Hawk Program Attempts to Do More in Less Time
               * Annual Funding Needs Are Higher under Restructured Program
               * Restructured Program Increased Concurrency between
                 Development and Production
               * Restructured Program Added Requirements and Increased Costs
               * Restructured Program Has Created Other New Challenges
               * Delays in Key Events Since Restructuring Can Impact Delivery
                 of New Capability
          * Global Hawk Program's Current Management Approach Sets Stage for
            Additional Risks
               * Restructured Acquisition Strategy Does Not Fully Capture
                 Product Knowledge at Key Decision Points
               * Joint Efforts Will Help Manage Risk in the Global Hawk
                 Program
          * Conclusions
          * Recommendations for Executive Action
          * Agency Comments and Our Evaluation
     * Scope and Methodology
     * Comments from the Department of Defense
     * Knowledge Gaps at Critical Knowledge Points
          * Technology Maturity Gap- Knowledge Point 1
          * Design Maturity Gap- Knowledge Point 2
          * Production Maturity Gap- Knowledge Point 3
     * Related GAO Products
     * http://www.gao.gov

                 United States Government Accountability Office

Report to the Chairman, Subcommittee

     on Tactical Air and Land Forces, Committee on Armed Services, House of
                                Representatives

November 2004

UNMANNED AERIAL VEHICLES

Changes in Global Hawk's Acquisition Strategy Are Needed to Reduce Program Risks

                                       a

UNMANNED AERIAL VEHICLES

Changes in Global Hawk's Acquisition Strategy Are Needed to Reduce Program
Risks

  What GAO Found

The restructuring of the Global Hawk program impacts the acquisition
program in multiple ways.

     o More and accelerated funding: Funding, which previously spanned 20
       years, now is compressed in about half the time. The restructured plan
       requires $6.3 billion through fiscal year 2012; the original plan
       would have needed $3.4 billion by that time. The budget request is now
       three times higher for some years (see figure below).
     o Immature technologies: Several critical technologies needed to provide
       the advanced capabilities are immature and will not be tested on the
       new air vehicle until late in the program, after which most of the air
       vehicles will already have been bought.
     o New requirements, new costs: DOD's desire to add additional Global
       Hawk capabilities tripled development costs. The program acquisition
       unit cost increased 44 percent since program start, yet fewer vehicles
       are to be produced than originally planned.
     o Challenges, trade-offs, and delays: The addition of new capabilities
       has led to space, weight, and power constraints for the advanced
       Global Hawk model. These limitations may result in deferring some
       capabilities. Some key events and activities-many related to testing
       issues-have been delayed.

Global Hawk's highly concurrent development and production strategy is
risky and runs counter in important ways to a knowledge-based approach and
to DOD's acquisition guidance. The restructuring caused gaps in product
knowledge, increasing the likelihood of unsuccessful cost, schedule,
quality, and performance outcomes. Because the restructured program is
dramatically different from the initial plan for the basic model, the
business case now seems out of sync with the realities of the acquisition
program.

Global Hawk's Annual Funding Requirements    
Funding, as of September 2001, before        Funding, as of December 2003, 
restructuring                                after                         
(in millions of then-year dollars)           restructuring (in millions of 
                                                           then-year dollars) 

                                                            800 600 400 200 0

    Years

2001

                         2003 2005 2007 2009 2011 2013

                                 2015 2017 2019

                       2001 2003 2005 2007 2009 2011 2013

                                 2015 2017 2019

                   Sources: Air Force (data); GAO (analysis).

                 United States Government Accountability Office

Contents

  Letter 1

Results in Brief 3 Background 4 Restructured Global Hawk Program Attempts
to Do More

in Less Time 7 Global Hawk Program's Current Management Approach Sets
Stage

for Additional Risks 16 Conclusions 23 Recommendations for Executive
Action 24 Agency Comments and Our Evaluation 24

  Appendixes

         Appendix I: Scope and Methodology 29 Appendix II: Comments from the
 Department of Defense 30 Appendix III: Knowledge Gaps at Critical Knowledge
                                                                   Points 35

  Related GAO Products

Table 1: Key Characteristics of Global Hawk RQ-4A and RQ-4B

  Tables

Models 6 Table 2: Global Hawk Program's Cost, Quantity, and Unit Costs 12
Table 3: How Global Hawk Product Knowledge Compares with

Three Critical Points in the Knowledge-Based

Approach 20

Figure 1: Global Hawk RQ-4A 5

  Figures

Figure 2: Global Hawk's Annual Funding Requirements 8 Figure 3:
Restructured 2004 Plan Is Highly Concurrent Compared with Original Global
Hawk Plan 11 Figure 4: Global Hawk Program Is Not Fully Aligned with DOD's
Acquisition Policy and Knowledge-Based Approach 19

Contents

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

A

United States Government Accountability Office Washington, D.C. 20548

November 5, 2004

The Honorable Curt Weldon Chairman Subcommittee on Tactical Air and Land
Forces Committee on Armed Services House of Representatives

The Air Force's Global Hawk is a high-flying unmanned aerial vehicle that
can capture detailed images of targets as well as wide swaths of terrain
and transmit those images on a near real-time basis to battlefield
commanders and intelligence centers. With an ability to see through
clouds, sandstorms, and other inclement weather conditions day or night
and to fly for more than a day, Global Hawk can significantly improve the
Department of Defense's (DOD) ability to gather intelligence,
surveillance, and reconnaissance1 information about potential adversaries
without risking an onboard pilot. An early model performed well in the
ongoing global war on terrorism, and the Secretary of Defense identified
Global Hawk as having the potential to transform military operations. The
Air Force began the acquisition program in 2001. With total program cost
estimated at $6.3 billion, Global Hawk is currently DOD's most expensive
unmanned aerial vehicle. The Navy is considering Global Hawk for a
maritime surveillance mission, and other federal agencies, several foreign
countries, and the North Atlantic Treaty Organization have all expressed
an interest in this aircraft system.

1 Intelligence is defined by DOD as the product resulting from the
collection, processing, integration, analysis, evaluation and
interpretation of available information concerning foreign countries or
areas. DOD also defines the term as information and knowledge about an
adversary obtained through observation, investigation, analysis, or
understanding. Surveillance is defined by DOD as the systematic
observation of places, persons, or things through visual and other means.
DOD defines reconnaissance as a mission undertaken to obtain, by visual
observation or other detection methods, information about activities and
resources of an enemy or potential enemy or to secure data characteristics
of a particular area.

In 2002, DOD restructured the program, accelerating development and
production, and, more importantly, changing the design significantly to a
new, larger, heavier, and more capable air vehicle to handle both imagery
and signals intelligence missions.2 Given Global Hawk's overall importance
to DOD, you asked us to review the program. This report discusses (1) the
restructuring's effect on the Air Force's ability to deliver new
capabilities to the warfighter in terms of funding, acquisition strategy,
cost, and other related issues and (2) whether the current business case
and management approach is knowledge-based and can help forestall future
risks.

To determine the effects of restructuring on the Global Hawk program's
cost, schedule, and performance goals, we reviewed the original
acquisition strategy, two major restructurings, and the current
acquisition strategy. We assessed changes to funding, annual budget
requests, the number of vehicles to be procured, vehicle capabilities, and
program cycles for developing, testing, and procuring Global Hawk. To
evaluate the program's likelihood of meeting its objectives, we compared
the current acquisition strategy with criteria established in DOD's
acquisition policy3 and with best practices and experiences of leading
commercial firms and successful government acquisitions. This methodology,
which we call a knowledge-based approach, enabled us to evaluate whether
the program achieved product knowledge at the right times in terms of
technology, design, and production maturity. We identified gaps in product
knowledge, reasons for those gaps, and risks associated with inadequate
knowledge. We performed our review from February 2004 to September 2004 in
accordance with generally accepted government auditing standards. More
details about our methodology are in appendix I.

2 Imagery intelligence is defined by DOD as being derived from the
exploitation of collection by visual photography, infrared sensors,
lasers, electro-optics, and radar sensors (such as synthetic aperture
radar sensors) wherein images of objects are reproduced optically or
electronically on film, electronic, display devices, or other media. DOD
defines signals intelligence as involving intelligence derived from
communications, electronic, and foreign instrumentation signals.

3 DOD Directive 5000.1 "The Defense Acquisition System" and DOD
Instruction 5000.2 "The Operation of the Defense Acquisition System," both
dated May 12, 2003.

  Results in Brief

The Global Hawk program has changed dramatically since the March 2001
decision to start both system development and low-rate production. The
program has been restructured twice to add the requirement for a totally
new and larger Global Hawk design to be developed and produced in less
time. Program funding, which previously had been stretched relatively
evenly across 20 years, is now compressed into roughly half the time,
tripling Global Hawk's budgetary requirements in some years. In fiscal
year 2006, for example, the program now plans to request about $750
million from Congress, three times what was planned originally. The
restructurings expanded the development period by 5 years and compressed
production by 9 years, creating significant concurrency between
development and production from fiscal year 2004 to 2010. Because of this
concurrency, the Air Force plans to invest in almost half of the total
fleet of the new larger Global Hawks before a production model is
flight-tested and completed to show that the air vehicle design works as
required. Likewise, full-rate production will begin before the airborne
signals intelligence and multiplatform radar (the two required
capabilities justifying the new, larger model) complete development and
are flight-tested to prove the integrated system will work as intended.
The primary reason for building the RQ-4B model was to integrate and carry
the advanced sensors to provide added capability to the warfighter. The
program's total cost estimates have increased by nearly $900 million,
driven by a threefold increase in development costs to pay for the
development of a new and larger air vehicle. As a result, the program
acquisition unit cost increased 44 percent since the program started.
Finally, in the past 2 years, the program has deferred some key
capabilities and experienced delays that can impact getting capabilities
to the warfighter.

The Air Force's restructured strategy does not fully follow the
evolutionary, knowledge-based approach espoused by best practices and
DOD's revised acquisition guidance. A knowledge-based approach encourages
managers to attain the necessary product knowledge at key points to
support investment decisions, ensuring, for example, that technologies are
mature before starting an acquisition program and that the design is
stable before beginning manufacturing. This approach includes incremental
or evolutionary development, which sets up a more manageable environment
for attaining and applying knowledge and is intended to increase the
chances of delivering a quality weapon system to the warfighter quickly
and cost effectively. While the original acquisition strategy more closely
adhered to this approach, the restructured strategy has caused gaps in
knowledge about technology, design, and manufacturing at major investment
decision points. These actions changed the underpinnings of the program's
original business case and increased the likelihood of future cost
increases and schedule delays in delivering the capabilities expected by
the warfighter. Air Force and contractor officials have established a wide
range of management controls to help mitigate risks. While some of these
controls may increase visibility into risks, the history of successful
product development programs has shown that risk mitigation plans do not
work optimally unless they are based on knowledge appropriate for
decisions that must be made at critical junctures.

We are recommending that the Secretary of Defense direct the Air Force to
revisit the decision to concurrently develop and produce the new Global
Hawk design. The Secretary should direct the Air Force to conduct and
present a new business case that defines the warfighter needs that can be
accommodated given current available resources of technology, engineering
capability, time, and money. To keep risks from increasing beyond current
levels in this program, we also recommend the Secretary delay further
procurement of the new Global Hawk, other than units needed for testing,
until a new business case is completed that reduces risk and justifies
further investments on a knowledge-based acquisition strategy. DOD
disagreed with both recommendations, stating that risks are being
effectively managed and that our approach would take more time and more
money to implement. We continue to believe that our recommendations would
improve congressional and DOD oversight, reduce program risks, and save
time and money over the life of the program through a more rigorous and
comprehensive application of knowledge-based practices.

The Global Hawk unmanned aerial vehicle system is designed to support

  Background

warfighting and peacekeeping missions by providing decision makers with
up-to-date information about potential adversaries' locations, resources,
and personnel. Operators on the ground can change Global Hawk's navigation
and direct the onboard sensors to survey a geographic area the size of
Illinois within a 24-hour cycle. As a high-altitude, long-endurance
aircraft, Global Hawk was originally designed to reach an altitude of
65,000 feet and fly for up to 35 hours.

Global Hawk began in 1994 as an acquisition concept technology
demonstration program, managed first by the Defense Advanced Research
Projects Agency and, since 1998, by the Air Force. Seven demonstrator
aircraft were eventually produced; three have since been destroyed in
mishaps. The demonstrator models logged several thousand-flight hours and
effectively supported combat operations in Afghanistan and Iraq. The
system passed a military usefulness assessment, completed several
demonstrations and other tests, and DOD judged it a success. However,
testing identified that significant improvements in reliability, sensor
performance, and communications were needed before producing operationally
effective and suitable systems.

In March 2001, DOD approved the Global Hawk for a combined start of system
development and low-rate initial production of six air vehicles based on
the successful demonstrations and operational deployments of demonstrator
aircraft. The Air Force planned to slowly develop more advanced
capabilities and acquire 63 air vehicles. This model, now called the
RQ-4A, is shown in figure 1.

                          Figure 1: Global Hawk RQ-4A

                           Source: Northrup Grumman.

                 Differences between the Two Global Hawk Models

In March 2002, DOD restructured the acquisition strategy to include a
second Global Hawk model, the RQ-4B. The new strategy includes 51 air
vehicles, 10 ground stations, multiple intelligence sensors, support
equipment, and facilities at a cost of $6.3 billion. Of the 51 air
vehicles to be purchased, 7 are RQ-4As and 44 are RQ-4Bs. Separately, the
Navy is procuring 2 RQ-4As and a ground station for about $300 million
(including development costs) to evaluate the vehicles' potential for the
Broad Area Maritime Surveillance Program. In December 2002, DOD
restructured the program again. Instead of buying all RQ-4Bs with multiple
intelligence capability, the RQ-4Bs will now have a mix of multimission
and single-mission capabilities. The two restructurings also increased
low-rate initial production quantities to 19 (recently increased to 20)
air vehicles: 7 RQ-4As and 12 (now 13) RQ-4Bs.

The RQ-4A and the RQ-4B differ significantly. The new RQ-4B model is
intended to have 50 percent greater payload capacity, a longer fuselage
and longer wing span and will be heavier than the A model. DOD considered
these changes necessary to carry new advanced sensor payloads and to
provide multi-intelligence capabilities on a single RQ-4B. Even though the
RQ-4B is bigger and heavier, it will use the same engine as the RQ-4A.
Table 1 shows the key differences in the two models.

       Table 1: Key Characteristics of Global Hawk RQ-4A and RQ-4B Models

     Key characteristics          RQ-4A                 RQ-4B                 
     Payload capacity             2,000 pounds          3,000 pounds          
     Take-off weight              26,750 pounds         32,250 pounds         
     Wingspan                     116.2 feet            130.9 feet            
     Fuselage length              44.4 feet             47.6 feet             
     Endurance                    31 hours              33 hours              
     Time at 60,000 feet          14 hours              4 hours               
     Average speed at 60,000 feet 340 knots             310 knots             
     Approximate range            10,000 nautical miles 10,000 nautical miles 

               Sources: Northrop Grumman (data); GAO (analysis).

In addition to the differences shown in table 1, the RQ-4B includes new
requirements for advanced sensors payloads,4 enhancements to
communications and ground stations, a new multiplatform common data link,
and an open systems architecture. The new design will use more advanced
technologies (such as lithium batteries and electric brakes), will require
a larger power-generating capability, and will incorporate new landing
gears that fold into the wing. Also, the design changes require new
manufacturing processes and investments in new production tooling-the
factory equipment and manufacturing items used to build large quantities
of major weapon systems, such as Global Hawk.

Global Hawk's restructuring has impacted the acquisition program in a
number of significant ways: the time span for funding has been compressed
into roughly half the time and the overall funding amount has increased;
concurrent development and production is causing the Air Force to invest
in almost half the total fleet of the new and improved Global Hawk vehicle
before a production model has proven that it will work as intended; and
development costs have tripled because of the need to develop a new and
improved vehicle. In addition, the program has deferred some capabilities
and incurred delays that could affect the Air Force's ability to deliver
Global Hawk to the warfighter.

  Restructured Global Hawk Program Attempts to Do More in Less Time

    Annual Funding Needs Are Higher under Restructured Program

The restructured program requires greater up-front investment, a faster
ramp-up in funding, and a larger total budget. The development period was
extended from 7 years to 12 years, and development funding increased
significantly to develop the RQ-4B and to integrate advanced sensor and
communication technologies. Procurement is now concentrated into 11 years
instead of the 20 years of relatively level procurement set out in the
original plan. The restructuring triples Global Hawk's budgetary
requirements in some years. Figure 2 illustrates the restructuring's
compression of the program and impact on annual funding requirements.

4 In addition to enhancements to existing imagery sensors, the RQ-4B will
eventually incorporate the Airborne Signals Intelligence Payload and the
MultiPlatform Radar Technology Insertion Program.

      Page 7 GAO-05-6 Global Hawk Acquisition

Figure 2: Global Hawk's Annual Funding Requirements Funding, as of
September 2001, before restructuring (in millions of then-year dollars)
800 700 600 500 400

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
                              2017 2018 2019 2020

Years Funding, as of December 2003, after restructuring (in millions of
then-year dollars)

2001

        2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

                         2015 2016 2017 2018 2019 2020

        Years

Development Production Military construction

                   Sources: Air Force (data); GAO (analysis).

    Restructured Program Increased Concurrency between Development and
    Production

Compared with the original plan, the restructured plan has much higher
annual funding requirements, placing more budgeted funds at risk of not
being fully funded when competing for the defense dollar. In their
respective peak years of budget requirements, the original plan would have
required $353 million (fiscal year 2010), while the restructured plan
expects to request $781 million (fiscal year 2007). The upcoming fiscal
year 2006 requirement is currently about $750 million, three times higher
than the original plan for that same fiscal year. Cumulatively, the
restructured plan requires $6.3 billion to be completed in fiscal year
2012, whereas the original plan would only have needed $3.4 billion by
that year.

Significant concurrency now exists between development and production that
covers the period from fiscal years 2004 to 2010. The Air Force now plans
to invest in almost half of the total RQ-4B fleet before a production
model is flight-tested and operational evaluations are completed to show
that the air vehicle design works as required. Full-rate production will
begin before the airborne signals intelligence and multiplatform radar
complete development and are flight-tested to prove the integrated system
will work as intended. The primary reason for building the RQ-4B model was
to integrate and carry the advanced sensors to provide added capability to
the warfighter. Additionally, schedule delays have already occurred in the
restructured plan that will continue to add pressure in the program.

Collectively, the actions to restructure the program have materially
changed the underpinnings in the original business case decision developed
to justify the start of system development and low-rate production. The
business case should provide sufficient evidence that resources are
available to meet warfighter needs. This case would include technology and
design demonstrations that added confidence that the integrated product
can be developed within time and money constraints. The original plan was
to first acquire basic RQ-4A systems very similar to the demonstrators and
then slowly and incrementally develop and acquire systems with more
advanced sensor capabilities while using the same air vehicle. This
strategy incorporated an evolutionary approach in that a basic capability
was to be produced in a first block of aircraft and a second, more
advanced block was to be acquired once the new technologies were mature.
Each block had separate decision points and testing plans and significant
risk was removed from the program because the demonstrators had been
built, tested, and extensively flown (and later used successfully in
actual combat operations in Afghanistan and Iraq). While testing showed it
needed some improvements, the RQ-4A was significantly more mature and
proven than the RQ-4B model to begin production.

Figure 3 illustrates the significantly greater concurrency of development
and production activities resulting from the program's restructuring
compared with the original plan. Historically, programs with high degrees
of concurrency are at greater risk of cost, schedule, and performance
problems than programs with less overlap of development and production.
The original acquisition strategy planned to complete most development
testing prior to beginning production, thereby taking advantage of product
knowledge. The restructured program added the new RQ-4B model,
substantially increased low-rate production quantities, and established
highly concurrent development and production cycles to acquire and test
several different RQ-4B configurations over the life of the program. The
Air Force plans to invest in 20 RQ-4Bs before completion of initial
operational test and evaluation. The reason for designing a larger and
heavier Global Hawk was to satisfy warfighter needs for the new advanced
sensors. However, integration and operational testing of the advanced
sensors on the fully configured air vehicle are not scheduled to be
completed and reported on until fiscal year 2009 for the advanced signal
intelligence sensor and fiscal year 2011 for the multiplatform radar. By
this time, the entire RQ-4B fleet will already be produced or on order.

  Figure 3: Restructured 2004 Plan Is Highly Concurrent Compared with Original
                                Global Hawk Plan

                      Original 2001 plan (in fiscal years)

Basic RQ-4A

Advanced RQ-4A

2001 '03 '04 '05 '06 '07'02  2021'08 '09 '10 '11 '12 '13 '14 Key           
                                '15 '16 '17 '18 '19 '20         attributes    
D evel                                                       12 air        
          op.                                                   vehicles with 
          P                                                     basic imagery 
              rodu ction                                        sensors       
                De velo  pme nt                                 25 air        
                                                                vehicles with 
                                                                advanced      
                                                                imagery       
                                                                sensors and   
                                                                26 air        
                                                                vehicles with 
                                                                signals       
                                           Pr oduc  tion        intelligence  
                                                                Total air     
                                                                vehicles: 63  

                    Restructured 2004 plan (in fiscal years)

Basic RQ-4A

Advanced RQ-4B

(phase 1)

Advanced RQ-4B

(phase 2)

Advanced RQ-4B

(phase 3)

Advanced RQ-4B

(phase 4)

                                                                 2021'12                 
                                                                 '13 '14   Key           
2001 '03 '04 '05 '06 '07 '08 '09 '10 '11'02                      '15 '16   attributes
                                                                 '17 '18   
                                                                 '19 '20   
D                                                                          7 air         
  Pevel roduop.                                                            vehicles with 
                                                                           basic imagery 
                ction                                                      sensors       
            Dev                                                            9 air         
                                                                           vehicles with 
                                                                           improved      
                elopPro menducti ton                                       imagery       
                                                                           sensors and   
                                                                           limited       
                                                                           signals       
                                                                           intelligence  
                    Dev elop                                               8 air         
                                                                           vehicles with 
                                                                           improved      
                                                                           sensors and   
                                                                           improved      
                                                                           signals       
                                 mentProd uctio n                          intelligence  
                                                                           to be         
                                                                           retrofitted   
                                                                           later with    
                                                                           advanced      
                                                                           signals       
                                                                           intelligence  
                                                                           15 air        
                                                                           vehicles with 
                                                                           improved      
                                                                           sensors and   
                              De     velo pme   nt                         advanced      
                                                  P rodu ction             signals       
                                                                           intelligence  
                                 Deve     lopm  ent                        12 air        
                                                                           vehicles with 
                                                                           advanced      
                                                                           multiplatform 
                                                                           radar         
                                                    Prod uctio n           technology    
                                                                           Total air     
                                                                           vehicles: 51  

Sources: Air Force (data); GAO (analysis).

Note: In this figure, development includes both technology and system
development.

    Restructured Program Added Requirements and Increased Costs

Global Hawk's development cost estimates have increased almost threefold,
from $906.2 million in March 2001 to about $2.6 billion in March 2004,
mostly due to the requirement for the new RQ-4B's inclusion in the
program. Total program costs have continued to increase, including an
increase of $466 million since March 2003. The program acquisition unit
cost increased 44 percent since program start, from $85.6 million to
$123.2 million. Increasing costs for Global Hawk raises affordability
issues and questions about employing the vehicle in medium- and
high-threat environments because of its high replacement costs and limited
numbers. Total procurement cost estimates decreased from program start due
to the cut in quantities from 63 to 51 and inflation savings resulting
from compressing the program and cutting 9 years of future procurement
activities. Table 2 shows how costs have changed since March 2001 in
millions of then-year dollars.

         Table 2: Global Hawk Program's Cost, Quantity, and Unit Costs

Total costa                     March 2002 (1st March 2003 (2nd March 2004 
                                   restructuring)  restructuring)  (status    
                                                                   this year) 
                      $906.2 March                                            
Development      2001 (original        $2,311.0        $2,395.6   $2,587.9
                             plan)                                 
Procurement             4,459.8         4,388.9         3,278.5    3,552.2 
Military                   28.0           146.7           140.8      140.8 
construction                                                    
Total program          $5,394.0        $6,846.6        $5,814.9   $6,280.9 
Air vehicles                 63              51              51         51 
Quantity                                                        
Ground stations              14              10              10         10 
Total program             $85.6          $134.2          $114.0     $123.2 
Unit costsa,b                                                   
Procurement               $70.8           $86.0           $64.2      $69.6 
only                                                            

Sources: Air Force (data); GAO (analysis).

aAll costs are expressed in millions of then-year dollars, which include
inflation and represent the Air Force's budget plans.

bTotal program unit cost is calculated by dividing the total cost of
development, procurement, and system-specific military construction for
the acquisition program by the quantity of air vehicles to be produced.
Procurement unit cost is the total amount for procurement divided by the
number of air vehicles to be procured. It does not include costs for
development and military construction.

    Restructured Program Has Created Other New Challenges

The following factors caused the Global Hawk program's cost estimates to
change between 2001 and 2004:

     o March 2001 cost estimate: Based on the original acquisition strategy
       to slowly and incrementally develop and acquire improved versions of
       the demonstrator model. The RQ-4B model was not yet part of the
       acquisition strategy.
     o March 2002 cost estimate: Reflects changes for the first restructuring
       of the program, which introduced the RQ-4B. Development costs
       increased significantly because of plans to quickly build advanced
       capabilities into the RQ-4B. While the quantity of air vehicles-the
       RQ-4A and RQ-4B models-and ground stations decreased because of
       revised user requirements, total procurement costs increased because
       of the higher cost for the RQ-4Bs and the plan at that time to equip
       all the larger platforms with multi-intelligence mission capabilities.
     o March 2003 cost estimate: Reflects a second restructuring for
       affordability reasons. In December 2002, DOD officials decided to
       switch from all multimission capabilities to a mix of multimission and
       single-mission RQ-4Bs. This switch lowered procurement costs by
       decreasing the required number of sensors.
     o March 2004 cost estimate: Between March 2003 and March 2004, total
       program cost increased by $466 million, and officials added another 18
       months to the development program to accomplish requirements deferred
       from prior years and to accommodate new requirements. Development
       costs increased to cover the extended schedule and additional
       requirements. Procurement costs increased primarily because of higher
       costs for structural components and for labor to build the RQ-4B.

Space, weight, and power constraints of the RQ-4B limit what capabilities
can be included now or added in the future. Some capabilities have already
been eliminated or deferred to later years. For example, the warfighter
wanted a defensive subsystem for Global Hawk, but development has been
delayed and may be dropped because of weight limitations in the air
vehicle, already at or near capacity with some of the new advanced sensor
payloads. Also, the RQ-4B configured with the airborne signals
intelligence payload is projected to have no capacity for future growth
because this payload weighs more than allocated in the design of the air
vehicle.

 Delays in Key Events Since Restructuring Can Impact Delivery of New Capability

Other development tasks have similarly been delayed or pushed out beyond
the budget years, including efforts related to demonstrating that Global
Hawk can operate in areas with extreme temperatures.

The Air Force's overall acquisition approach to add new technologies
whenever they are deemed ready was designed to allow flexibility in
responding to changes in priorities and new requirements. However, Global
Hawk's vehicle limitations and changing requirements have increased
development challenges. For example, despite the space, weight, and power
limitations of the RQ-4B, Air Force officials stated that Global Hawk
users and other DOD officials continue to identify potential future
technologies and capabilities for possible incorporation into Global Hawk.
Absent major downsizing of the advanced sensors or other payloads, the Air
Force will need to consider dedicating the RQ-4B to an increasing number
of single and specific-rather than multi-intelligence-missions, if the
goal is to utilize new and unproven emerging technologies not currently
part of the Global Hawk plan.

The new schedule for some key events and activities has slipped because of
programmatic, budget, or external issues. Air Force and contractor
officials say that a significant contributor to schedule delays was the
episodic deployment of Global Hawk's earlier model in Afghanistan and
Iraq. The Global Hawk system-including considerable numbers of Air Force
and contractor personnel, ground stations, and supporting equipment-has
been used to support combat operations and is subject to future deployment
orders.

Some examples of program events that have been delayed and others whose
future schedules have slipped include:

     o government acceptance of the second RQ-4A production aircraft due to
       quality and performance problems identified during tests;
     o delivery of the equipment and support needed to begin initial
       operations at the Global Hawk's home base, Beale Air Force Base in
       California;
     o the operational assessment of the RQ-4A;
     o completion of the first phase of combined developmental and
       operational testing of Global Hawk;
     o acquisition of production tooling, establishing manufacturing
       processes, and delivering parts needed for production;
     o delivery schedules projected for RQ-4B air vehicles; and
     o the expected start of initial operational test and evaluation to
       support the full-rate production decision.

Delays and deficiencies in scheduled development testing could compromise
upcoming decisions in the program. According to test officials in the
Office of the Secretary of Defense, the first of five phases of Global
Hawk's combined development and operational testing is not as robust as
originally planned and is taking significantly longer than expected. As of
July 2004, only about 10 percent of the required flight test points had
been completed and nearly 70 percent of the remaining test points were
either on hold or not fully defined. The approved test plan required this
testing to be completed by September 2004, but testing officials do not
expect it to be completed until March 2005. Test delays are occurring due
to late delivery of key subsystems, lack of resources, deployments in
support of the global war on terrorism, other program priorities, and
unexpected testing problems. Test officials told us that the lack of
quality test data is hampering their ability to provide meaningful
oversight.

The results from this first phase of development testing were to be used
in the operational assessment of the first two production RQ-4A aircraft
starting in September 2004 to assess the Global Hawk's mission readiness
and suitability. Because of phase one delays, the start of the assessment
has slipped until at least March 2005. Test officials believe further
delays are likely because of other higher priorities, including the
start-up activities at Beale Air Force Base. At this time, a firm date for
the testing has not been scheduled and the unapproved test plan still
lacks the necessary details to ensure effective testing. Test officials
believe the operational assessment is in jeopardy of being cancelled or
cut back in order to start the dedicated initial operational test and
evaluation on time. The officials say that eliminating the operational
assessment, or reducing its scope, would add risk to the program. Entering
the next phase involves testing the new, larger RQ-4B aircraft and
advanced sensor payloads, and, without having the assurances the
production aircraft are mission-ready, additional tests will likely be
required.

  Global Hawk Program's Current Management Approach Sets Stage for Additional
  Risks

In attempting to get advanced capabilities to the warfighter sooner, the
Air Force's restructured acquisition strategy for the Global Hawk program
does not fully follow best practices and DOD acquisition guidance for an
evolutionary, knowledge-based acquisition process. DOD recently rewrote
its acquisition policy specifically to encourage acquisitions to develop
and deliver increased capability to the warfighter incrementally (or on an
evolutionary basis), only when appropriate knowledge concerning
technology, design, and manufacturing has been attained. Compared with the
original strategy, the new Global Hawk acquisition strategy has yielded
less product knowledge in each of these areas, thereby raising the
likelihood of future negative impacts on cost, schedule, and performance.
Air Force and contractor officials acknowledge that-with its highly
compressed and concurrent schedule-the program is risky and presents major
management challenges. The Air Force has established management controls
and processes intended to mitigate risks; however, without a disciplined
process to capture and base investment decisions on key technology,
design, and manufacturing knowledge, the controls are less robust and the
risks remain high.

    Restructured Acquisition Strategy Does Not Fully Capture Product Knowledge
    at Key Decision Points

By approving the start of system development and low-rate production at
the same time, Global Hawk's restructured acquisition strategy skipped the
critical decision points that require the capture of key product knowledge
used to inform decisions to move forward in an acquisition program.
Skipping the necessary steps to capture technology, design, and
manufacturing knowledge has added risk to the program. The Air Force would
have captured more knowledge under the original March 2001 strategy, which
more closely followed the knowledge-based approach. At that time, the plan
was to acquire basic air vehicles and ground systems very similar to the
demonstrators that had already been built, extensively flown, and (later)
used in combat. The Air Force then planned to upgrade sensor and
performance capabilities for the next production lot as the technologies
matured while retaining the same airframe. Since the decision to start the
program, additional information and experience have closed some of the
gaps, but a substantial lack of knowledge continues to add risk to the
RQ-4B acquisition.

GAO has a body of work focused on best practices in product development
and weapon systems acquisition.5 We have found that when program managers
capture key product knowledge at three critical knowledge points during a
major acquisition, the probability of meeting expected performance within
cost and schedule objectives increases. Each of the points builds on
previously attained knowledge. The acquired knowledge is used to identify
and reduce any risks before moving a weapon system to the next stage of
development. This approach to developing new products-commercial and
defense-has been shown over time to continually produce successful
outcomes in terms of cost, schedule, and performance.

In recent years, DOD revised its acquisition policy to embrace an
evolutionary and knowledge-based approach, which we believe provides a
sound framework for the acquisition of major weapon systems. This policy
covers most of DOD's major acquisition programs. As noted in our November
2003 report,6 this revised policy is a step in the right direction. The
acquisition policy states that program managers shall provide knowledge
about the key aspects of the system at key decision points in the
acquisition process7 and an evolutionary or incremental development
approach should be used to establish a more manageable environment for
attaining and applying knowledge. The customer may not get the ultimate
capability right away, but the initial product is available sooner and at
a lower cost. The policy adopts the essence of the following points from
the knowledge-based approach:

o  Knowledge point 1: Should occur when the acquisition program is
scheduled to start, when the customer's requirements are clearly defined,
and resources-proven technology, engineering capability,

5 GAO, Defense Acquisitions: Assessments of Major Weapon Programs,
GAO-04-248 (Washington, D.C.: Mar. 31, 2004). This report includes an
assessment of the Global Hawk program against the knowledge-based
approach. Other recent reports discussing best practices include GAO, Best
Practices: Capturing Design and Manufacturing Knowledge Early Improves
Acquisition Outcomes, GAO-02-701 (Washington, D.C.: July 15, 2002) and
Defense Acquisitions: DOD Faces Challenges in Implementing Best Practices,
GAO-02-469T (Washington, D.C.: Feb. 27, 2002).

6 GAO, Defense Acquisitions: DOD's Revised Policy Emphasizes Best
Practices, but More Controls Are Needed, GAO-04-53 (Washington, D.C.: Nov.
10, 2003).

7 For example, the policy states that unless some other factor is
overriding in its impact, the maturity of the technology shall determine
the path to be followed by the program in entering the system development
phase of the acquisition cycle.

time, and money-exist to satisfy them. This match should support the
business case for starting system development and demonstration.
Technology should be mature before starting a program, and, therefore, the
technology development phase of an acquisition should be separate from the
system development phase.

     o Knowledge point 2: Should occur at the design readiness review, about
       halfway through the system development phase, when the product's
       design is determined to be capable of meeting product requirements-the
       design is stable and ready to begin initial manufacturing of
       prototypes.
     o Knowledge point 3: Should occur when managers commit to starting
       production, when information is available to determine that a reliable
       product can be produced repeatedly within established cost, schedule,
       and manufacturing quality targets.

Figure 4 shows a generalized depiction of DOD's acquisition policy, where
DOD's key milestones are anchored along a typical program's acquisition
path and where the three knowledge points from the knowledge-based
approach fit along this path. Also shown in figure 4 is how the Global
Hawk program overlaps technology and system development and begins
production before the necessary knowledge is achieved.

Figure 4: Global Hawk Program Is Not Fully Aligned with DOD's Acquisition Policy
                          and Knowledge-Based Approach

Sources: DOD and Air Force (data); GAO (analysis).

Global Hawk's new strategy approved initial production of the improved
RQ-4B well in advance of completing technology maturation and approved
developing and integrating the vehicle's design with the various sensor
payloads desired by the warfighter. Furthermore, low-rate production was
approved without ensuring the quality and reliability of manufacturing
processes. This approach added significant risk in that sensor
technologies and final design may not meet the space, weight, and power
limitations of the RQ-4B, which is in low-rate production, and may not
satisfy the warfighter's requirements. By not closing knowledge gaps in
the integrated product design (air vehicle, sensor payloads, and data
links) needed to meet requirements, there is increased risk that sensor
development schedules may need to be extended to achieve form, fit, and
function for an integrated Global Hawk system. Otherwise, the program
office may have to go back to the warfighter and further negotiate
requirements. Table 3 compares the product knowledge available to support
key decision points under the original plan in March 2001 with the
knowledge obtained at the start of RQ-4B production in July 2004. A black
dot indicates product knowledge meets best practice standards from
knowledge-based approach.

 Table 3: How Global Hawk Product Knowledge Compares with Three Critical Points
                        in the Knowledge-Based Approach

                          Global Hawk program status:

                           Date March 2001 July 2004

Status At combined system development and At start of RQ-4B production for
basic limited production decision for RQ-4A RQ-4B air vehicle design with
limited signals intelligence

 Three critical points of knowledge-based approach and best practice standards:

Knowledge point 1: Technologies needed to meet essential product
requirements have been demonstrated to work in their intended environment,
and the producer has completed a preliminary design of the product.

                Technologies matured to high readiness levels  o

                     Preliminary design established  o   o

Knowledge point 2: Design is stable and has been demonstrated through
prototype testing. Ninety percent of engineering drawings releasable to
manufacturing organizations.

a

90 percent of engineering drawings released  o

Prototype demonstration that design should meet  o  b requirements

                Reliability targets and growth curve established

Knowledge point 3: Product is ready to be manufactured within cost,
schedule, and quality targets. All key manufacturing processes have come
under statistical process control and product reliability has been
demonstrated.

Fully integrated system representative prototype  o  demonstrated to work
in operational environment

             Critical processes capable and in statistical control

                            Reliability demonstrated

Sources: Air Force (data); GAO (analysis).

Note: A black dot ( o ) indicates product knowledge meets best practice
standards from knowledgebased approach.

a75 percent of drawings released at design readiness review in April 2004.

bSystem representative prototype will not be built. Since the basic design
evolved from RQ-4A, the Air Force and contractor are conducting modeling
efforts and component tests, such as wind tunnel testing of the new wing,
to validate that the RQ-4B air vehicle design should meet requirements.

The table shows that the level of product knowledge approached best
practice standards when the decision was made in March 2001 to start
system development and low-rate production of the RQ-4A. The program's
restructurings in 2002, however, created substantial gaps in technology,
design, and manufacturing knowledge that have not yet been closed by the
start of RQ-4B production. Lack of product knowledge increases risks of
poor cost, schedule, and performance outcomes. Appendix III includes a
more detailed discussion of knowledge gaps at each knowledge point.
Following are brief examples of knowledge gaps as they relate to each of
the three critical knowledge points.

Technology maturity: Using best practices, at the start of system
development, a program's critical technologies should be in the form, fit,
and function needed for the intended product and should be demonstrated in
a realistic environment. The RQ-4B development program is struggling to
meet these criteria for several of its most critical technologies. Nearly
2 years after development began, the technologies required for the RQ-4B
to perform its operational mission including enhanced imaging sensors,
signals intelligence, multiplatform radar, and open system architecture
are immature, basically at a functional rather than form or fit
configuration. For example, the airborne signals intelligence payload and
multiplatform radar technology insertion program are still in development
under separate Air Force programs. These subsystems are key to providing
the advanced intelligence, surveillance, and reconnaissance capabilities
for which the RQ-4B is being developed. At the time of our review, neither
of these technologies had been demonstrated in an operational environment
using a system prototype. Air Force officials expect them to be mature by
the time they begin buying sensors to incorporate them into the Global
Hawk production line in fiscal years 2008 and 2009. However, by this time
most of the air vehicles will have already been bought. Also, operational
testing to evaluate performance in a realistic operating environment is
not scheduled until late fiscal year 2008 for the signals intelligence
sensor and late 2010 for the radar. Nevertheless, the Air Force continues
to build the RQ-4B platform lacking solid assurance that these critical
subsystems will work as planned.

Design maturity: The program had completed 75 percent of RQ-4B model
drawings by the design readiness review in comparison with the 90 percent
completion standard for best practices. While the Air Force anticipated
the design and experience on the RQ-4A would add assurances and speed
efforts to mature the new RQ-4B design, the two vehicles ultimately had
only about 10 percent commonality. While drawings completed were

    Joint Efforts Will Help Manage Risk in the Global Hawk Program

approaching best practice standards, the Air Force did not build a
prototype of the RQ-4B design to demonstrate a stable design.
Demonstration of the design is a key factor in ensuring a stable design.
The Air Force had not established a reliability growth goal or plan and
had not identified critical manufacturing processes, both essential to the
next phase of production and needed to ensure quality and cost targets can
be met.

Production maturity: Officials have started to identify the critical
manufacturing processes for the RQ-4B but do not intend to collect and use
statistical process control data to ensure the manufacturing could deliver
quality products. The new RQ-4B requires new manufacturing processes
because of major differences from the RQ-4A. In addition, Officials from
the program office, the prime contractor, and the Defense Contract
Management Agency continue to identify problems and concerns about the
performance and quality of work by several key subcontractors, including
those producing the wing, the advanced sensor suite, and the vertical tail
and aft fuselage parts. This latter subcontractor is new to large-scale
manufacturing using advanced composite materials and has experienced
significant start-up and quality problems. According to best practices,
the subcontractor's critical processes must be demonstrated to ensure good
quality and limit rework. The prime contractor and DOD sent special teams
of advisors to help develop the firm's manufacturing processes and to
train employees. Creating another gap in production maturity, a fully
integrated system representative prototype was not tested before starting
production and will not have been demonstrated before full-rate
production, scheduled in 2007. By then, 45 percent of the RQ-4B planned
quantities will be under contract.

Air Force and contractor officials agree that the restructured program
significantly increased program and technical risks. They acknowledge that
the use of the approach to insert technology periodically affects all
aspects of the program, making it more challenging to manage functional
areas, including logistics support, contracting, program integration, and
testing. To better manage the risks and challenges created by this
acquisition approach and environment, the Global Hawk management team
provided the following as examples of actions they are taking

     o better teaming practices between the government and contractor to
       manage the program at all levels;
     o better controls for the release of funds on both development and
       production contracts;
     o allocation of higher amounts of management reserve funding during
       contract performance;
     o use of a "buy to budget" concept that limits activity in the program
       to a ceiling amount of funds planned for the total program; and
     o use of a risk management database to focus the attention of management
       on the most critical risks facing the program.

These are all management practices that can be used to manage any product
development program and will likely identify and help manage risks in the
Global Hawk program. Nevertheless, using a knowledge-based approach that
captures critical knowledge at key junctures in a program has been shown
time and again in both commercial and defense acquisition programs to
consistently produce successful outcomes-cost, schedule, quality, and
performance.

In March 2001, DOD approved the start of development and production for

  Conclusions

Global Hawk on the basis of a business case that matched requirements with
resources-technologies, engineering capabilities, time, and funding. The
first increment of Global Hawk was based on mature technologies and a
design proven to meet the warfighter's need through actual combat use of
the technology demonstrator. The plan included a reasonable funding
profile and embraced a knowledge-based acquisition strategy that completed
development before entering production. The plan included future
improvements to the baseline capability as technologies and funding became
available. By December 2002, the Air Force had dramatically changed the
Global Hawk's acquisition plan and the knowledge-based foundation for the
earlier decision to proceed into development and production. This change
created large gaps between Global Hawk's requirements and the resources
available to meet them. The new plan required a new, larger, and heavier
air vehicle with only 10 percent commonality with the previous proven
design; increased development time; and accelerated production time,
creating significant concurrency between development and production. To
accommodate the changes, the plan calls for twice the annual funding
amounts in peak years over the old plan. Overall, the new plan has
increased risks significantly. Subsequent reviews by DOD have acknowledged
the changes in the program have increased uncertainty. The new design has
not been demonstrated to work using a prototype model; technologies to
support the advanced sensor payloads that drove the need for a new Global
Hawk design are still immature; and the Air Force will be requesting about
$750 million in funding next year for the program. Yet, the Air Force has
awarded a contract to start the production of the new, larger Global Hawk
B model with the hope that simulations and analysis will be sufficient to
allow decision makers to manage risk. The history of DOD managed programs
suggests otherwise.

To decrease risks of poor outcomes and to increase the chances of

  Recommendations for

delivering required warfighter capabilities with the funds available, we
are making recommendations to the Secretary of Defense to take the
following two actions

     o direct the Air Force to revisit the decision to begin concurrent
       development and production of the Global Hawk B design and direct the
       Air Force to create and present a new business case that defines the
       warfighter's needs that can be accommodated given current available
       resources of technology, engineering capability, time, and money, and
     o delay further procurement of the Global Hawk B, other than units
       needed for testing, until a new business case is completed that
       reduces risk and justifies further investments based on a
       knowledge-based acquisition strategy.

  Agency Comments and Our Evaluation

DOD provided us with written comments on a draft of this report. The
comments appear in appendix II. DOD stated that it did not concur with our
two recommendations. Separately, DOD provided one technical comment that
we incorporated in this report to more accurately characterize the issue
of affordability and use of Global Hawks in threat conditions.

Regarding our first recommendation on completing a new business case to
justify and guide concurrent development and production of the RQ-4B
model, DOD stated its belief that the Global Hawk's acquisition strategy
balances acquisition risks with the department's demands to rapidly field
new capabilities to the warfighter, thereby obviating the need for a new
business case. Furthermore, by following what officials call an
evolutionary development process, DOD said it is providing
transformational warfighting capabilities to ongoing military operations
without disrupting Global Hawk's current development and production
activities. DOD said it is effectively managing risk with the help of
regular oversight meetings and by requiring monthly and quarterly activity
reports.

We continue to believe that a new business case is needed to support
further investments and to improve oversight by Congress and DOD decision
makers. The program today is much different than the one supported by the
original business case. The Air Force started with an advanced concept
technology demonstration program that proved the capability of a smaller
and lighter Global Hawk air vehicle. Use of this vehicle on numerous
occasions in actual combat situations has saved lives, according to Air
Force and contractor officials. However, this is not the vehicle that the
Air Force now plans to produce. Instead, the Air Force dramatically
changed the acquisition strategy for the Global Hawk program and is not
gaining some key knowledge before production. The Air Force plans to
concurrently design and produce a new Global Hawk air vehicle that is
significantly larger and heavier than the earlier version used in combat.
The larger air vehicle is intended to accommodate new, heavier, and larger
sensors that will not be available until the 2008 to 2009 time frame. In
implementing the restructured strategy, the Air Force is not fully
following a knowledge-based approach for developing the RQ-4B Global Hawk
as called for by best practices and DOD's new defense acquisition
guidance. The new guidance clearly states that knowledge reduces risks,
and we agree.

While the Air Force believes it can manage the risk of a concurrent
development and production program by holding regular meetings with
acquisition executives and by issuing management reports, DOD's own
experience has shown this to be risky and a factor that led DOD to change
its acquisition policy to a knowledge-based approach. History has shown
concurrency usually delays the delivery of a needed capability and results
in higher costs. From March 2003 to March 2004, estimated program costs
have increased by $466 million, and the sensors and the new air vehicle
are still being developed. Stepping back from this rush to produce the new
air vehicle and establishing a new business case designed to capture key
product knowledge before costly investments in production would better
inform DOD decision makers and Congress about what is feasible with
available technology and dollar resources to meet warfighter needs and to
better assess the extent and/or severity of program acquisition risks.

Regarding our second recommendation to delay further procurement of the
RQ-4B (other than units needed for testing) until a new knowledge-based
and risk-reducing business case is prepared, DOD stated that its current
acquisition strategy effectively manages risk and fosters the rapid
delivery of needed capabilities to the warfighter. DOD said we overstated
risks from RQ-4B development, design changes, and insertion of advanced
sensor capabilities. DOD further stated that our recommendation would
result in a production break with serious cost and schedule complications
and that GAO's sequential knowledge-based approach does not consider
real-world events, such as the September 11, 2001, terrorist attack in the
United States or issues related to North Korea and Iraq.

We believe the risks in the Global Hawk program are real and continue to
support delaying the near-term procurement of air vehicles not needed for
testing. We think this is a prudent way for the program to gain knowledge
before significantly increased resource investments and to reduce risks
until a new air vehicle integrated with the advanced signals intelligence
payload and the multiplatform radar can be demonstrated through testing to
meet warfighter requirements. Our report notes that operational testing of
the air vehicle's performance and suitability will not take place until
almost half the fleet is already purchased and that integration and
testing of the advanced sensors will not occur until late in the program
after the fullrate production decision is made and most systems are
bought. DOD's comments appear to decouple the air vehicle from the
advanced sensors by stating that, if a sensor diverges from its current
plan, alternate future payloads could fill the RQ-4B's greater payload
capacity. However, the need for designing a new larger air vehicle was
predicated on its ability to carry these specific sensors to meet the
warfighter's requirements. Therefore, we believe that knowledge based on a
demonstration of the integrated capability is key to supporting production
and delivery of the product within estimated cost and schedule.
Additionally, the new Global Hawk program strategy requires significantly
greater amounts of funding earlier, putting that investment at risk should
changes occur as development and testing is completed.

Regarding a potential break in production, our analysis indicates that a
break is neither impending nor certain if our recommendation were adopted.
We are not recommending that DOD stop production or reduce the total
quantity but rather a near-term delay in procuring the portion of annual
buys for air vehicles not needed for testing. Funds currently on contract
and approved appropriations for fiscal year 2005 would continue production
on the Air Force's planned schedule through mid-fiscal year 2007 at least.
Only then would a production break or slowdown happen, and only if the Air
Force has not yet prepared a business case to justify its investments
beyond that point based on demonstrated product knowledge of the new air
vehicle. If the current acquisition strategy and financial plan are
feasible and appropriate, the Air Force would be able to prepare and
justify a comprehensive business plan for the RQ-4B well in advance of a
potential break.

DOD indicated that our knowledge-based acquisition approach was untimely
and not adaptive to fast-changing world events. When we developed the
knowledge-based approach, our high priority was to focus on better ways to
deliver capability to the warfighter more quickly through incremental, or
evolutionary, development. Our approach is based on a careful study of
historical DOD acquisition programs and the best efforts in the private
sector. Our prior work shows that proceeding without requisite knowledge
ultimately costs programs more money and takes longer to complete than
those adopting a more rigorous and comprehensive strategy basing
investment decisions on key product knowledge-technology, design, and
production maturity levels. DOD agreed with our findings and changed its
acquisition guidance to reflect a knowledge-based approach. We note in
this report that the original Global Hawks were produced through a
successful demonstration program that effectively and quickly provided the
warfighter with transformational intelligence, surveillance, and
reconnaissance capabilities. Defense Contract Management Agency reports,
contract cost reports, corporate briefings, design drawing changes, new
tooling and new production processes, and the evident need for Air Force
and prime contractor task teams to be extensively deployed to
subcontractor facilities, all indicate that the RQ-4B program entails
higher degrees of risk, greater management challenges, and significant
changes from production of the RQ-4A and earlier demonstrators.

We believe that our recommendation to delay further procurement of the
RQ-4B until a new knowledge-based and risk-reducing business case is
prepared prudently balances real-world internal investment risks with
military demands from real-world external events. The Air Force could have
continued to deliver the capability of the Global Hawk that was the direct
outgrowth of the demonstration program while allowing the sensor and radar
technology time to mature before investing in a new larger and more risky
Global Hawk program. This would have allowed continued delivery of the
enhanced RQ-4A capability to the warfighter while minimizing the impacts
of design changes that come out of normal development and testing and that
grow more costly as a product enters the production environment. The heavy
cost of design changes after production is underway could impact DOD's
ability to respond to other warfighter needs in the post-9/11 world.

As arranged with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days from
its issue date. At that time, we will send copies to interested
congressional committees, the Secretary of Defense, the Secretary of the
Air Force, the Secretary of the Navy, and the Director, Office of
Management and Budget. In addition, the report will be available on the
GAO Web site at h ttp://www.gao.gov.

If you or your staff has any questions concerning this report, please
contact me at (202) 512-4163 or Michael J. Hazard at (937) 258-7917. Other
staff making key contributions to this report were Lily J. Chin, Bruce D.
Fairbairn, Steven M. Hunter, Matthew B. Lea, Charlie Shivers, and Adam
Vodraska.

Michael J. Sullivan
Director (Acting)
Acquisition and Sourcing Management

Appendix I

Scope and Methodology

To determine the effects of Global Hawk's restructuring on cost, schedule,
and performance goals, we compared the original acquisition strategy, two
major revisions, and the current acquisition strategy as implemented. We
identified changes in cost, quantity, fleet composition, and sensor
capability mixes as well as overall consequences of restructuring on total
funding requirements, annual budget requests, and program cycles for
developing, testing, and producing the Global Hawk. We reviewed management
plans, cost reports, contract files, progress briefings, and risk data to
identify program execution efforts and results to date. We identified cost
changes, schedule delays, and performance issues.

To evaluate whether the current acquisition approach can help forestall
risks, we applied GAO's methodology for assessing risks in major weapon
systems. This methodology is derived from the best practices and
experiences of leading commercial firms and successful defense acquisition
programs. We reviewed program office and prime contractor organizations,
processes, and management actions. We extracted and evaluated program and
technical risks maintained in a risk database used by the program office
and contractor to identify major risks and the steps taken to mitigate
risks. We compared the program office's plans and results to date against
best practice standards in achieving product knowledge in terms of
technology, design, and production maturity information and in applying
knowledge to support major program decisions. We identified gaps in
product knowledge, reasons contributing to those gaps, and the elevated
risks expected as a consequence of inadequate product knowledge. We
further analyzed original and current acquisition approaches to
demonstrate the high concurrency of development, production, and testing
and the elevated risks imposed as a result.

In performing our work, we obtained information and interviewed officials
from the Global Hawk System Program Office, Wright-Patterson Air Force
Base, Ohio; 452nd Flight Test Squadron, Air Force Flight Test Center, and
Detachment 5, Air Force Operational Test and Evaluation Center, Edwards
Air Force Base, CA; Defense Contract Management Agency, San Diego and
Palmdale, CA; Northrop Grumman Integrated Systems, Rancho Bernardo and
Palmdale, CA; and offices of the Director, Operational Test and
Evaluation, and Unmanned Aerial Vehicle Planning Office, which are part of
the Office of the Secretary of Defense in Washington, D.C.

We conducted our work from February to September 2004 in accordance with
generally accepted government auditing standards.

                                  Appendix II

                    Comments from the Department of Defense

Appendix II Comments from the Department of Defense

Note: Page numbers in the draft report may differ from those in this
report.

Appendix II
Comments from the Department of Defense

Appendix II
Comments from the Department of Defense

Appendix II
Comments from the Department of Defense

Appendix III

                  Knowledge Gaps at Critical Knowledge Points

  Technology Maturity Gap- Knowledge Point 1

Achieving a high level of technology maturity at the start of system
development is a particularly important best practice. This means that the
critical technologies needed to meet essential product requirements are in
the form, fit, and function needed for the intended product and have been
demonstrated to work in their intended environment. The RQ-4B development
program is struggling to meet these criteria for several of its most
critical technologies. More than 2 years after development began, the
technologies required for the RQ-4B to perform its operational mission
including enhanced imaging sensors, signals intelligence, multiplatform
radar, and open system architecture are immature, basically at a
functional rather than form or fit configuration. Nevertheless, the Air
Force continues to build the RQ-4B platform, lacking solid assurance that
these critical subsystems will work as planned.

In particular, the airborne signals intelligence payload and multiplatform
radar technology insertion program are still in development under separate
Air Force programs and will be purchased by the Global Hawk program as
government furnished equipment. These subsystems are key to providing the
advanced intelligence, surveillance, and reconnaissance capabilities for
which the RQ-4B is being developed. At the time of our review, neither of
these technologies had been demonstrated in an operational environment
using a system prototype. Air Force officials characterized their current
stages of development as laboratory settings demonstrating basic
performance, technical feasibility, and functionality but not form or fit
(size, weight, materials, etc.). Technology maturity of the sensors is
critical because the basic design of the RQ-4B has been completed and
allocates limited space, weight, and power for the new capability. If the
new sensors cannot be developed within these constraints, some performance
trade-offs-such as reduced frequency coverage-are likely. The airborne
signals intelligence payload currently exceeds the weight allocated for
its integration into the RQ-4B, while the multiplatform radar uses most of
the vehicle's available power-generation capability.

Officials expect them to be mature by the time they begin buying sensors
to incorporate them into the Global Hawk production line in fiscal years
2008 and 2009. However, by this time most of the air vehicles will have
already been bought; additional time and money might be needed to fix or
retrofit any remaining differences. Also, operational testing to evaluate
performance in a realistic operating environment is not scheduled until
late fiscal year 2008 for the signals intelligence sensor and late 2010
for the radar. Any changes or delays in these programs would likely impact
Global Hawk cost, schedule, and/or performance.

            Appendix III Knowledge Gaps at Critical Knowledge Points

  Design Maturity Gap- Knowledge Point 2

Seventy-five percent of engineering drawings were released at the Global
Hawk design readiness review that triggered the start of RQ-4B
manufacturing and assembly. This figure is 15 percent less than the best
practices' standard of 90 percent. The Air Force and contractor had
anticipated being able to use much of the design work and production
experience on the RQ-4A to prove the design and decrease the time and
extent of engineering work on the RQ-4B. However, officials found out that
the two models had much less in common than anticipated. About 90 percent
of the airframe had to be redesigned-only 10 percent was common to both
models. Therefore, relying on the experience of the RQ-4A increased the
risk of poor program outcomes because the RQ-4B is substantially heavier;
incorporates a new wing, fuselage, and vertical tail; has a 50 percent
greater payload capacity to carry advanced sensors still in development;
and requires new production tooling, new materials, and changed
manufacturing processes.

The Air Force also did not build an RQ-4B prototype-a best practice to
demonstrate design stability-before awarding a contract to start
production. An analysis of the development contract performance, as of May
2004, shows that development and integration efforts needed to finalize
the design and prepare the RQ-4B for production is behind schedule and
over cost. The planned work efforts were just over one-half completed, but
two-thirds of the budget allocated for these efforts was expended. Defense
Contract Management Agency analysts cited cost growth in labor and
materials and problems in finalizing and releasing design drawings as
causes for the problems.

Neither the original nor the current plan established comprehensive
reliability targets and growth curves. Reliability growth is the result of
an iterative design, build, test, analyze, and fix process. Improvements
in reliability of a product's design can be measured by tracking
reliability metrics and comparing the product's actual reliability with
the growth plan and, ultimately, to the overall goal. Although both models
are in production, the Air Force did not establish reliability growth
programs to measure how reliability is improving and to uncover design
problems so fixes could be incorporated before the design was frozen and
before committing to production.

            Appendix III Knowledge Gaps at Critical Knowledge Points

  Production Maturity Gap- Knowledge Point 3

Officials have started to identify the critical manufacturing processes
for the RQ-4B but do not intend to collect and use statistical process
control data to ensure that the manufacturing could deliver quality
products within best practices quality standards and that the end product
meets the design and specifications. The officials' assessments of the
program continue to identify significant concerns about the quality,
performance, and timeliness of the work of several subcontractors. For
example, the subcontractor building the vertical tail and main parts of
the fuselage is new to large-scale manufacture using advanced composite
materials. The firm experienced significant start-up problems and the
prime contractor and DOD sent special teams of advisors to help develop
the firm's manufacturing processes and to train employees. The
subcontractor's critical processes must be demonstrated to ensure good
quality and limit rework. Officials have identified similar concerns with
the subcontractors building the wing and imaging sensor.

The Air Force started producing the A and B models without first
demonstrating that the systems would meet reliability goals. Reliability
is a function of the specific elements of a product's design and making
changes after production begins is costly and inefficient. Best practices
for system development require reliability to be demonstrated by the start
of production. The RQ-4A is a production version of the demonstrators with
few changes. Testing of the demonstrators identified a need to evaluate
reliability under a stressful operating tempo. Air Force officials told us
that reliability improvements on the RQ-4A were constrained, as were
demonstrations of reliability. The RQ-4B design has incorporated
improvements in such areas as flight control actuators, mission computers,
avionics, and structures that officials expect will fix some of the
identified problems and improve reliability, but these have not been
demonstrated.

Finally, the Air Force did not acquire and test a fully integrated system
representative prototype before committing to production. The contract for
the first three units was awarded and work began in late fiscal year 2004.
Budget plans call for procuring 13 RQ-4Bs in low-rate production through
the fiscal year 2006. The Air Force has also programmed advance
procurement funds in fiscal year 2006 for 7 more, meaning that the
government will have made investments in 20 RQ-4Bs-45 percent of the
entire RQ-4B fleet-before the basic air vehicle is flight tested and
before evaluations are made leading to the full-rate production decision,
scheduled in fiscal year 2007. The Air Force also plans to enter full-rate
production without complete testing to demonstrate that a fully integrated
system-with advanced sensors and data links-will work as intended, is

Appendix III Knowledge Gaps at Critical Knowledge Points

reliable, and can be produced within cost, schedule, and quality targets.
Initial operational test and evaluation will only test the RQ-4B air
vehicle with its basic imagery intelligence payloads. Complete operational
testing and incorporation of the advanced signals intelligence payload and
the multiplatform radar capabilities-the reasons for acquiring the larger
model in the first place-will not occur until later in the program, after
the full-rate decision is made.

In the absence of specific product knowledge required by best practices
and DOD acquisition guidance, the Air Force and its contractor are
depending on the operational experience of the demonstrators, lab modeling
and simulation efforts, and production of the RQ-4A to help "close the
gaps" and provide some assurance on the RQ-4B design maturity, its
reliability, and its producibility within cost, schedule, and quality
targets. Although the demonstrator program had notable successes, testing
identified significant improvements were needed before producing
operationally effective and suitable air vehicles. Areas needing
improvement included reliability under a stressful operating tempo,
performance of sensors, mission planning, and communications bandwidth
burden. We also note that the RQ-4A is a production version of the
demonstrators with few changes and that government acceptance of the
second production RQ-4A was delayed due to deficiencies, including flight
problems. Moreover, as previously discussed, the RQ-4B is significantly
different than the RQ-4A and requires investing in new tooling and changed
manufacturing processes. These factors contribute to increased risks of
poor cost, schedule, and performance outcomes due to incomplete product
knowledge.

Related GAO Products

Defense Acquisitions: Assessments of Major Weapon Programs.

GAO-04-248. Washington, D.C. Washington, D.C.: March 31, 2004.

Force Structure: Improved Strategic Planning Can Enhance DOD's Unmanned
Aerial Vehicles Efforts. GAO-04-342 . Washington, D.C.: March 17, 2004.

Defense Acquisitions: DOD's Revised Policy Emphasizes Best Practices, but
More Controls Are Needed. GAO-04-53. Washington, D.C.: November 10, 2003.

Defense Acquisitions: Matching Resources with Requirements Is Key to the
Unmanned Combat Air Vehicle Program's Success. GAO-03-598 . Washington,
D.C.: June 30, 2003.

Best Practices: Setting Requirements Differently Could Reduce Weapon
Systems' Total Ownership Costs. GAO-03-57. Washington, D.C.: February 11,
2003.

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

Defense Acquisitions: DOD Faces Challenges in Implementing Best Practices.
GAO-02-469T. Washington, D.C.: February 27, 2002.

Best Practices: Better Matching of Needs and Resources Will Lead to Better
Weapons System Outcomes. GAO-01-288 . Washington, D.C.: March 8, 2001

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

Defense Acquisition: Employing Best Practices Can Shape Better Weapon
System Decisions. GAO/T-NSIAD-00-137. Washington, D.C.: April 26, 2000.

Unmanned Aerial Vehicles: Progress of the Global Hawk Advanced Concept
Technology Demonstration. GAO/NSIAD-00-78. Washington, D.C.: April 25,
2000.

Related GAO Products

Unmanned Aerial Vehicles: DOD's Demonstration Approach Has Improved
Project Outcomes. GAO/NSIAD-99-33 . Washington, D.C.: August 16, 1999.

Best Practices: DOD Training Can Do More to Help Weapon System Program
Implement Best Practices. GAO/NSIAD-99-206 . Washington, D.C.: August
16,1999.

Best Practices: Better Management of Technology Development Can Improve
Weapon System Outcomes. GAO/NSIAD-99-162 . Washington, D.C.: July 30,
1999.

Defense Acquisitions: Best Commercial Practices Can Improve Program
Outcomes. GAO/T-NSIAD-99-116 . Washington, D.C.: March 17, 1999.

Defense Acquisition: Improved Program Outcomes Are Possible.

GAO/T-NSIAD-98-123. Washington, D.C.: March 18, 1998.

Best Practices: DOD Can Help Suppliers Contribute More to Weapon System
Programs. GAO/NSIAD-98-87 . Washington, D.C. March 17, 1998.

Best Practices: Successful Application to Weapon Acquisition Requires
Changes in DOD's Environment. GAO/NSIAD-98-56. Washington, D.C.: February
24, 1998.

Major Acquisitions: Significant Changes Underway in DOD's Earned Value
Management Process. GAO/NSIAD-97-108 . Washington, D.C.: May 5, 1997.

Best Practices: Commercial Quality Assurance Practices Offer Improvements
for DOD. GAO/NSIAD-96-162 . Washington, D.C.: August 26,1996.

  GAO's Mission

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

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

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

                             Order by Mail or Phone

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

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

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

Contact:

To Report Fraud, Web site: www.gao.gov/fraudnet/fraudnet.htm

  E-mail: [email protected]

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

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

Susan Becker, Acting Manager, [email protected] (202) 512-4800

  Public Affairs

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

           Presorted Standard Postage & Fees Paid GAO Permit No. GI00

United States
Government Accountability Office
Washington, D.C. 20548-0001

Official Business Penalty for Private Use $300

Address Service Requested
*** End of document. ***