Military Space Operations: Common Problems and Their Effects on  
Satellite and Related Acquisitions (02-JUN-03, GAO-03-825R).	 
                                                                 
In fiscal year 2003, the Department of Defense expects to spend  
more than $18 billion to develop, acquire, and operate satellites
and other space-related systems. Satellite systems collect	 
information on the capabilities and intentions of potential	 
adversaries. They enable military forces to be warned of a	 
missile attack and to communicate and navigate while avoiding	 
hostile action. And they provide information that allows forces  
to precisely attack targets in ways that minimize collateral	 
damage and loss of life. DOD's satellites also enable global	 
communications, television broadcasts, weather forecasting;	 
navigation of ships, planes, trucks, and cars; and		 
synchronization of computers, communications, and electric power 
grids. Congress requested that we review reports we issued on	 
satellite and other space-related programs over the past two	 
decades and identify common problems affecting these programs.	 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-03-825R					        
    ACCNO:   A07059						        
  TITLE:     Military Space Operations: Common Problems and Their     
Effects on Satellite and Related Acquisitions			 
     DATE:   06/02/2003 
  SUBJECT:   Military operations				 
	     Satellites 					 
	     Space exploration					 
	     Program evaluation 				 
	     Defense procurement				 

******************************************************************
** 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-03-825R

GAO- 03- 825R Satellite Acquisition Programs

United States General Accounting Office Washington, DC 20548

June 2, 2003 The Honorable Jerry Lewis Chairman, Subcommittee on Defense
Committee on Appropriations House of Representatives

Subject: Military Space Operations: Common Problems and Their Effects on
Satellite and Related Acquisitions Dear Mr. Chairman: In fiscal year 2003,
the Department of Defense expects to spend more than $18 billion to
develop, acquire, and operate satellites and other space- related systems.
Satellite systems collect information on the capabilities and intentions
of potential adversaries. They enable military forces to be warned of a
missile attack and to communicate and navigate while avoiding hostile
action. And they provide information that allows forces to precisely
attack targets in ways that minimize collateral damage and loss of life.
DOD*s satellites also enable global communications, television broadcasts,
weather forecasting; navigation of ships, planes, trucks, and cars; and
synchronization of computers, communications, and electric power grids.

You requested that we review reports we issued on satellite and other
space- related programs over the past two decades and identify common
problems affecting these programs. In addition to analyzing past reports,
we interviewed Air Force space acquisition officials and reviewed past DOD
studies as well as DOD*s selected acquisition reports to the Congress. As
agreed with your office, given the short timeframe of this assignment, we
did not thoroughly assess underlying causes of problems identified or the
effectiveness of actions being taken to address these problems. However,
we plan to do so as part of a follow- on study. To the extent possible, we
looked at the current status of programs we reviewed. However, because we
principally relied on past GAO and DOD reports, some recent changes in
status and cost may not be reflected. We conducted our review from April
2003 through May 2003 in accordance with generally accepted government
auditing standards.

GAO- 03- 825R Satellite Acquisition Programs Page 2

RESULTS IN BRIEF

The majority of satellite programs cost more than expected and took longer
to develop and launch than planned. In reviewing our past reports, we
found that these results were commonly tied to the following problems.

(1) Requirements for what the satellite needed to do and how well it must
perform were not adequately defined at the beginning of a program or were
changed significantly once the program had already begun. (2) Investment
practices were weak. For example, potentially more cost- effective

approaches were not examined and cost estimates were optimistic. (3)
Acquisition strategies were poorly executed. For example, competition was
reduced for the sake of schedule or DOD did not adequately oversee
contractors. (4) Technologies were not mature enough to be included in
product development.

Several factors contributed to these problems. First, DOD often took a
scheduledriven instead of a knowledge- driven approach to the acquisition
process. As a result, activities essential to containing costs, maximizing
competition among contractors and testing technologies were compressed or
not done. Second, there is a diverse array of organizations with competing
interests involved in overall satellite development* from the individual
military services, to testing organizations, contractors, civilian
agencies, and in some cases international partners. This created
challenges in making tough tradeoff decisions, particularly since, for
many years, there was no high- level official within the Office of the
Secretary of Defense dedicated to developing and enforcing an overall
investment strategy for space. Third, space acquisition programs have
historically attempted to satisfy all requirements in a single step,
regardless of the design challenge or the maturity of technologies to
achieve the full capability. This approach made it difficult to match
requirements to available resources (in terms of time, money, and
technology).

Other factors also created challenges for the satellite acquisition
programs we reviewed. These include a shrinking industrial base, a
declining space workforce, difficulties associated with testing satellites
in a realistic environment, as well as challenges associated with
launching satellites.

DOD has studied problems affecting its satellite acquisitions and is
undertaking efforts to address these problems. We plan to evaluate these
efforts in a subsequent review. Therefore, we are not making
recommendations in this report.

BACKGROUND

DOD*s current space network is comprised of constellations of satellites,
groundbased systems, and associated terminals and receivers. Among other
things, these assets are used to perform intelligence, surveillance, and
reconnaissance functions; perform missile warning; provide communication
services to DOD and other government users; provide weather and
environmental data; and provide positioning and precise timing data to U.
S. forces as well as national security, civil, and commercial users. Table
1 identifies specific satellite systems used for these purposes. Appendix
I describes these systems in more detail.

GAO- 03- 825R Satellite Acquisition Programs Page 3 Table 1: Current and
Planned Satellite Systems

Function Current Systems Planned Systems

Missile warning and tracking Defense Support Program

(DSP) Space- Based Infrared System (High)

Space Tracking and Surveillance System (STSS) Intelligence, Surveillance
and Reconnaissance National Reconnaissance

Office satellites (not covered in this review)

NRO satellites

DOD*s Space- based Radar (not covered in this review)

Communications

Wideband/ high capacity systems Protected systems

(antijam, survivable) Narrowband systems

Defense Satellite Communications System (DSCS) Global Broadcasting Service
(GBS) (not covered by GAO reports)

Milstar

Ultra High Frequency FollowOn satellite communications system (UFO) (not
covered by GAO reports)

Wideband Gapfiller Satellite (WGS)

Advanced Wideband System (AWS)

Advanced Extremely High Frequency (AEHF) Advanced Polar System

Mobile User Objective System (MUOS) (not covered by GAO reports)
Navigation, Positioning, Timing Global Positioning System

(GPS) Next Generation GPS Weather/ Environmental Defense Meteorological

Satellite Program (DMSP) National Polar- orbiting Operational
Environmental Satellite System (NPOESS) All of these systems are playing
an increasingly important role in military operations. According to DOD
officials, for example, in Operation Iraqi Freedom, approximately 70
percent of weapons were precision- guided, most of those utilizing GPS
capabilities. Weather satellites enabled warfighters to not only prepare
for, but also to take advantage of blinding sandstorms. Communications and
intelligence satellites were also heavily used to plan and carry out
attacks and to assess post- strike damage.

Some of DOD*s satellite systems* such as GPS* have also grown into
international use for civil and military applications and commercial and
personal uses. In addition, many satellites launched over the past two
decades have lasted longer than expected. For example, some of the later
DSP spacecraft have operated for more than 10 years* well past design
lifetime.

The Joint Staff and the Combatant Commands are responsible for
establishing overall requirements while the services are responsible for
satisfying these requirements to the maximum extent practical through
their individual planning, programming, and

GAO- 03- 825R Satellite Acquisition Programs Page 4 budgeting systems.
According to DOD, the Office of the Secretary of Defense and the

intelligence community*s Community Management Staff provide high- level
leadership for national security space activities. The Air Force is the
primary procurer and operator of space systems and spends the largest
share of defense space funds, annually averaging about 85 percent. The Air
Force Space Command is the major component providing space forces for the
U. S. Strategic Command.

The Army controls the Defense Satellite Communications System and operates
ground mobile terminals. The Navy operates the Ultra High Frequency
follow- on satellites, the Geosat follow- on satellites, a weather
satellite, and some space systems that contribute to surveillance and
warning. And the National Reconnaissance Office designs, procures, and
operates space systems for intelligence and defense activities.

In addition, the National Security Space Architect and National Security
Space Integration Directorate coordinate national security space
architectures and plans for future national security space activities. The
Office of the Secretary of Defense, the Marine Corps, and other DOD
agencies also participate in national security space activities.

COMMON PROBLEMS AFFECTING SATELLITE ACQUISITIONS The majority of satellite
programs we have reviewed over the past two decades

experienced problems during acquisition that drove up costs and schedules
and increased technical risks.

First, requirements for what the satellite needed to do and how well it
must perform were not adequately defined at the beginning of a program or
were changed significantly once the program had already begun. This made
it more difficult for programs to ensure that they could match their
requirements to their resources (in terms of money, time, and technology).
The more requirements were added or changed, the more that cost and
schedule increased.

Second, investment practices were weak. At times, programs did not explore
potentially more cost- effective investment approaches. Once they settled
on an approach, programs often did not develop realistic cost estimates.
From a broader

perspective, investments in programs were not made in accordance with an
overall space investment strategy for DOD. Funds were sometimes shifted
from healthier programs to pay for weaker ones. Further, according to DOD
officials, decisions external to the program office were sometimes imposed
that resulted in unexpected funding cuts.

Third, acquisition strategies were poorly executed. For example,
competition was reduced for the sake of schedule or DOD did not adequately
oversee contractors. At times, contract type was not suitable for the work
being done.

Fourth, programs did not always ensure that technologies were mature
before making heavy investments in the program. This often caused cost and
schedule increases due to the need to fix problems later in development. A
continuing problem is that software needs are poorly understood at the
beginning of a program.

GAO- 03- 825R Satellite Acquisition Programs Page 5 Table 2 identifies
examples of problems identified in our reports and affected systems.

Table 2: Specific Common Problems Identified in GAO Reports

Problems Systems Affected by One or More Problems Requirements* Defining
what the system needs to do and how well it needs to perform

Program did not adequately define requirements

Unresolved conflicts among users on requirements Frequent changes made to
requirements after product development began

DSP replacement programs

Milstar

AEHF

SBIRS- High

Investment Strategy* Choosing a path that offers the most cost- effective
solution and ensuring costs are contained

Program did not adequately analyze investment alternatives

Cost and/ or schedule estimates were optimistic

Funding was unstable

DSP replacement programs

SBIRS- Low/ STSS

Milstar

AEHF

SBIRS- High

GPS III

Acquisition Strategy* Maximizing competition and contractor reliability

Level of competition was reduced or eliminated

Contract type was not suitable for work being done Poor oversight over
contractors

AEHF

SBIRS- High

SBIRS- Low

STSS

EELV

Technology* Ensuring technology is mature before heavy investments are
made in the program

Technology not sufficiently mature at program start

Software needs poorly understood

Testing compressed, skipped, or done concurrently with production

DSP replacement programs

Milstar

SBIRS- Low

AEHF

SBIRS- High Several factors contributed to the problems identified in our
reports. First, DOD took a schedule- driven versus a knowledge- driven
approach to the acquisition process. As a result, activities essential to
containing costs, maximizing competition among contractors and testing
technologies were shortchanged. Second, there was a diverse array of
organizations with competing interests involved in overall satellite
development* from the individual military services, to testing
organizations, contractors, civilian agencies, and in some cases, even
international partners. This created challenges in making tough tradeoff
decisions, particularly since, for many years, there was no high- level
official within the Office of the Secretary of Defense dedicated to
developing and implementing an overall investment strategy for space. 1
Often, disagreements within DOD would go unresolved for a long period of
time.

Third, space acquisition programs have historically attempted to satisfy
all requirements in a single step, regardless of the design challenge or
the maturity of technologies to achieve the full capability. This approach
made it difficult to match requirements to available resources (in terms
of time, money, and technology).

1 In 1994, DOD established the Office of the Deputy Under Secretary of
Defense for Space. The Deputy was responsible for developing,
coordinating, and overseeing the implementation of space policy. The
Deputy also had oversight responsibility for space architectures as well
as space acquisition programs. In 1998, this office was dissolved and its
responsibilities divided and given to other offices within OSD and the
military services.

GAO- 03- 825R Satellite Acquisition Programs Page 6 Table 3 further
illustrates how these cross- cutting factors can contribute to problems

in requirements, investment strategy, acquisition strategy and technology.
Table 3: Cross- cutting Factors Contributing to Space Acquisition Problems
and Potential Outcomes

Cross- Cutting Factors Requirements Investment Acquisition

Strategy Technology Schedule driven vs. knowledge driven approach

Requirements not fully known at start of program

Changes drive up costs and schedule Planning is optimistic;

costs not fully known at start of program. Alternatives not analyzed or
eliminated to meet schedule pressures

Solution being pursued may not be the most costeffective;

decisionmakers lack insight into cost growth

Competition may be shortchanged in an attempt to accelerate development

Best technical solution may be ignored; costs go up

due to lack of competition Testing schedule is compressed to meet target
launch date or it

is done concurrently with production. Less time to fix problems that arise
during testing

Costs and schedule increase due to need to fix problems later in product
development

Multiple players (Air Force, Army, Naval, Space Commands, testing
organizations, contractors, other agencies, international

partners); no *honest broker* at OSD level.

Competing/ conflicting requirements set. Changes made

throughout product development

Requirements cannot be matched to resources

Original cost estimates become invalid. Investments not made in accordance
with overall space

investment strategy for DOD Overall investment in space may not be
optimized

Single- step development vs. evolutionary development

Requirements exceed resources (time, money, technology) at time of product
development

Technology too immature at product development

GAO- 03- 825R Satellite Acquisition Programs Page 7 Other Factors Created
Challenges for Acquisitions Other factors also created challenges for the
satellite acquisition programs we

reviewed. Specifically, as with other defense industry sectors, the
satellite industry has seen a high rate of consolidation resulting in
reduced levels of competition. In 1998, we reported that since 1990 the
number of defense satellite contractors shrunk from 8 to 5. Moreover, in
recent years, the U. S. commercial space industry has seen decreasing
demand and increasing international competition. Our work has found
varying levels of success in maintaining and promoting competition within
this environment.

DOD has also had difficulty in maintaining the capability to launch its
satellites* partly due to problems within the expendable launch sector and
partly due to a decision in the 1970s to fly all DOD spacecraft on NASA*s
space shuttle. According to a DOD report 2 , as a result of the latter,
DOD investments in space launch infrastructure and vehicle improvements
virtually halted until the Challenger accident of 1986. The accident
itself disrupted launch schedules for programs such as GPS. At the same
time, the lack of investment in launch capabilities for so many years
contributed to higher launch costs after the accident and serious
operational limitations due to aging and obsolete launch vehicle
components and a dependence on outdated launch vehicle production lines.
In 1998, we reported that the number of contractors in this sector fell
from 6 to 2.

Air Force officials also cited challenges related to DOD*s space
workforce. In 2001, a congressionally chartered commission looking at
space issues, known as the Space Commission, noted that from its inception
the defense space program has benefited from world- class scientists,
engineers, and operators, but now many experienced personnel are retiring
and recruitment and retention of qualified space personnel is a problem.
Further, the commission concluded that DOD does not have the strong
military space culture* including focused career development and education
and training* it needs to create and maintain a highly trained and
experienced cadre of space professionals who can master highly complex
technology as well as develop new concepts of operation for offensive and
defensive space operations.

Unique aspects of satellite development and testing also presented
challenges for programs we reviewed. For example, some testing on
satellites can be done on the ground in thermovac or other environmental
simulation chambers. Some systems can also be tested via aircraft.
However, the only way to test satellites in the true operational space
environment is to build one or more demonstrator satellites and launch
them into orbit. Launching demonstrators is costly and time- consuming but
it offers greater assurance that satellites will work as intended. Also, a
high degree of coordination between space and ground segments as well as
user equipment is necessary. Typically, satellite software is used to test
the satellite before it is shipped for launch. Ground control software is
typically installed/ fielded a year before launch to allow for training
and rehearsals. Therefore, scheduling slips within any one of

2 Aspin, Les, Secretary of Defense, Report on the Bottom- up Review,
October 1993.

GAO- 03- 825R Satellite Acquisition Programs Page 8 these activities can
cause problems for other activities. At the same time, the timing

of the launching of satellites must coincide with the deployment of ground
receivers, but this can be difficult to do when ground and space segments
are funded by different military services.

In addition, satellite programs require a significantly larger investment
in the acquisition phase than other weapons systems. This is because
satellites are RDT& E intensive, go through extensive development testing,
and need to have all of their sustainment capabilities on board when
launched. Once on orbit they require a reduced amount of funding to
operate when compared with the funding profile of a typical, large
production DOD program. Air Force space acquisition officials stated that
the funding profile for a satellite program is typically the reverse of
the funding profile for a typical DOD program. The notional DOD lifecycle
profile shows approximately 28 percent of a program*s budget funding its
development and 72 percent of its budget funding the production of
hundreds of units and paying for the operations and sustainment that goes
with it. For a satellite program, the funding profile is *front- loaded*
with 60- 70 percent of its budget funding development and launch with 30-
40 percent of the budget funding operations and maintenance of the
satellite system. According to Air Force officials, this sort of profile
makes it difficult to adapt to unknowns that arise since it is not
possible to trade out- year production funding to fund near- term problems
since the production numbers for satellite systems are so small.

HOW PROBLEMS AFFECTED SPECIFIC PROGRAMS

Nearly every program we reviewed over the past several decades experienced
one or more of the problems we identified and experienced cost and
scheduling increases as a result. Corrective actions were taken on some
programs to reduce cost, schedule or technical risks after they were
identified. For example, the NPOESS program took a range of actions to
reduce program risks, including deferring development of requirements,
deciding to rely on existing versus new technology for some sensors, and
using aircraft to test sensors. In other cases, problems were allowed to
persist to the point where DOD needed to step in a restructure the
program. SBIRS- Low, for example, was restructured after continuing to
experience cost growth and scheduling delays, and SBIRS- High was
restructured last year after experiencing continued cost growth and
schedule delays. In the 1990s, three separate programs designed to replace
DSP satellites were abandoned after it became clear that they would be
either too costly and/ or technically risky to pursue.

Recent cases are discussed in more detail below. A chronology of our
findings related to individual systems is also provided in appendix I.

GAO- 03- 825R Satellite Acquisition Programs Page 9 Advanced EHF Satellite
The AEHF is a satellite system intended to replace the existing Milstar
system and to

be DOD's next generation of higher speed, protected communication
satellites. We recently reported that cost estimates developed by the Air
Force for this program increased from $4.4 billion in January 1999 to $5.6
billion in June 2001 for five satellites. Moreover, DOD will not meet its
accelerated targeted date for launching

the first satellite in December 2004. In fact, the first satellite's new
launch date is December 2006. (DOD has since decided to purchase three
satellites with options to purchase the fourth and fifth. The December
2002 Selected Acquisition Report for the AEHF showed current program costs
at $4.7 billion for three satellites. )

Several factors contributed to cost and schedule overruns and performance
shortfalls. First, in the early phases of the AEHF program, DOD
substantially and frequently altered requirements. Although considered
necessary, many changes were substantial, leading to cost increases of
hundreds of millions of dollars because they required major design
modifications. Second, based on a satellite constellation gap caused by
the failure of a Milstar satellite, DOD decided to accelerate its plans to
build the AEHF satellites. The contractors proposed, and DOD accepted, a
high risk schedule that turned out to be overly optimistic and highly
compressed- leaving little room for error and depending on a chain of
events taking place at certain times. Substantial delays occurred when
some events did not occur on time. DOD decided to take this approach on
the grounds it offered a chance to meet unmet warfighter requirements
caused by the loss of the Milstar satellite. Third, at the time DOD
decided to accelerate the program, it did not have the funding needed to
support the activities and the manpower needed to design and build the
satellites quicker. The lack of funding also contributed to schedule
delays, which in turn, caused more cost increases.

Advanced Wideband Satellite System (AWS) AWS (also known as the
Transformational Communications Satellite or TSAT) is a fairly new program
focused on supplementing AEHF and replacing DOD*s Wideband Gapfiller
Satellite system (WGS). DOD plans to include laser crosslinks on the
satellite to significantly increase capacity. In 2003, GAO reported that
the AWS program is scheduled to enter product development with only one of
its five critical technologies mature according to best practice
standards. Four immature technologies were scheduled to reach maturity by
January 2006, more than 2 years after development start. Three of four
technologies have a backup technology in case of development difficulties.
But the Single Access Laser Communications technology has no backup, and
according to program officials, any delay in maturing this technology
would result in a slip in the expected launch date.

SBIRS- High SBIRS- High satellites are being developed to replace DOD*s
older missile warning satellites. In addition to missile warning and
missile defense missions, the satellites will perform technical
intelligence and battlespace characterization missions. After the program
was initiated in 1994, it faced cost, scheduling, and technology problems.

GAO- 03- 825R Satellite Acquisition Programs Page 10 GAO reports from 1995
through 2001, for example, noted that the program was facing

serious hardware and software design problems. In 2001, the program
reported that it had exceeded the 25 percent cost threshold established in
10 U. S. C. 2433. In 2002, an independent review team chartered by DOD to
examine the reasons behind cost and scheduling problems in the SBIRS- High
program reported that a key root cause was that system requirements were
not well- understood when the program began and as it evolved. In
addition, the requirements setting process was often adhoc with many
decisions being deferred to the contractor. The review team also found
that the program was too immature to enter system design and development.
Further, there was too much instability on the program after the contract
award* with DOD undertaking four major replanning efforts. DOD has since
restructured the program and taken corrective actions, but the team noted
that there were still risks within the program, including risks related to
the schedule.

SBIRS- Low SBIRS- Low satellites are to perform missile warning and
missile tracking functions. Because of their low- earth orbit, they may be
particularly useful in tracking missiles through the midcourse of their
flight* when missiles themselves have cooled down and become more
difficult to track.

SBIRS- Low has been restructured due to cost, scheduling, and technical
problems. Despite spending several billion dollars on these efforts, DOD
has not launched a single satellite or demonstrated any space- based
missile tracking capabilities from space using technologies similar to
those to be used by SBIRS- Low (now called the Space Tracking and
Surveillance System, or STSS). In 2001, GAO reported that DOD was not
adequately analyzing or identifying cost- effective alternatives to SBIRS-
Low that could satisfy critical missile defense requirements, such as a
Navy ship- based radar capability. At the time, other studies supported
the possibility that other types of sensors could be used to track
missiles in midcourse of their flight and to cue interceptors. In 2001,
GAO reported that the SBIRS- Low acquisition schedule was at high risk of
not delivering the system on time or at cost or within expected
performance. Satellite development and production, for example, were to be
done concurrently, leaving the Air Force at risk of having to correct
problems discovered during testing at late stages of the acquisition
process, when they are more expensive and time- consuming to fix. SBIRS-
Low also had high technical risks because some critical satellite
technologies were judged to be immature for the current stage of the
program, including the scanning infrared sensor, tracking infrared sensor,
and technologies used to cool down satellite sensors. As the program was
experiencing cost and schedule problems, DOD restructured the program,
moving it from the Air Force to the Missile Defense Agency to reflect the
increased focus on missile defense and renaming it the Space Tracking and
Surveillance System (STSS).

In May 2003, we reported that the STSS program was not considering two
potentially more cost effective alternatives*( 1) delaying the launch date
by one year and (2) stopping efforts to launch existing technology for
research purposes and concentrating instead on new technology. Moreover,
the program faced investment and scheduling risks since it recently
reduced competition within the program and it decided on a 2007 launch
date without knowing the extent of work that must be done on the satellite
equipment it plans to assemble and launch.

GAO- 03- 825R Satellite Acquisition Programs Page 11 National Polar-
orbiting Operational Environmental Satellite System (NPOESS) This program
essentially combined separate weather satellite efforts being pursued

by DOD and the National Oceanic and Atmospheric Administration (NOAA)
after it was determined that doing so could reduce duplication and save
money. Our earlier reviews identified potential requirements setting
problems attributable to the broad base of internal customers each agency
has and the diversity of requirements that needed to be met. DOD*s
selected acquisition report on NPOESS stated that coordination and
validation of the broad- based requirements took longer than anticipated
and delayed a request for proposal release by 6 months. In 1997, the
NPOESS program assessed specific technical, scheduling, and cost risks
facing the program, and determined there were risks within the interface
data processing segment, the space segment, and the overall system
integration segment. To reduce these risks, the program deferred
development of requirements either because the technology needed to
implement them did not exist or the requirement was too costly. It
undertook earlier development of some satellite sensors in order to allow
more time to mature technologies. It decided, in some cases, to use
existing sensor technologies instead of building new ones. It also
increased testing to demonstrate satellite sensors and to deliver early
data to users to that they could begin to work with the data.

Global Positioning System (GPS) GPS satellites, which provide positioning,
navigation, and timing information to military forces and civilian users,
have existed for over 25 years, but a full constellation of satellites has
been operational for only 7 years. In 1980, we reported that the cost to
acquire and maintain GPS satellites through 2000 increased from $1.7
billion to $8.6 billion due largely to estimates not previously included
for replenishment satellites, launches, and user equipment. In 1983, we
reported that costs might still be understated since system design changes
were being considered. Costs and schedule were significantly affected in
1987 as a result of the Challenger accident, since DOD was depending on
the space shuttle to launch GPS satellites. Reliability problems with GPS
receivers also affected schedule throughout the program. In 1991, for
example, we reported that DOD postponed full- rate production for receiver
sets by 2 years due to reliability problems. Last fall, according to GPS
program officials, the program was on track to launch the first GPS III
satellite in 2012. However, following a review by the Under Secretary of
the Air Force, funding for the program was zeroed for fiscal year 2004,
and $46 million was withheld from the fiscal year 2003 budget. Without a
full release of the withheld funding, the program office believes the
launch date may slip past 2012.

DOD HAS STUDIED ACQUISITION PROBLEMS

DOD has studied many of the problems related to satellite acquisitions
identified in our reviews and is making changes. A 1994 study performed by
the U. S. Space Command, for example, stated that DOD*s process of
defining requirements for space systems needed to be improved to ensure
greater Joint Staff and Service influence in decisionmaking. With
increasing budget pressures and dramatically different post

GAO- 03- 825R Satellite Acquisition Programs Page 12 Cold War strategies,
the U. S. Space Command also noted that it was essential for all

services to better understand the costs and benefits of requirements. A
1998 study performed by the United States Air Force Scientific Advisory
Board advocated adopting commercial practices such as business case
analysis, streamlined procurement, and spiral development of ground
segments as a way to improve acquisition practices. The study also called
for improved oversight by high- level officials, development of improved
cost/ performance models that increase visibility into program status and
emerging problems, and maintaining adequate budget reserves in acquisition
programs to minimize reprogramming actions and avoid program disruptions.

More recently, the U. S. Space Commission, chaired by Donald Rumsfeld,
found that DOD*s budgeting process and declining space workforce created
difficulties for acquisitions. Specifically, the Commission noted that
when satellite programs are funded in one budget and terminals in another,
the decentralized arrangement can result in program disconnects and
duplication. It can result in lack of synchronization in the acquisition
of satellites and their associated terminals. It can also be difficult for
user requirements to be incorporated into the satellite system if the
organization funding the system does not agree with and support those user
requirements.

Last year, the independent review team studying the SBIRS- High program
recognized that there were broad, systemic issues that need to be
addressed on space programs. These include: the need for pre- acquisition
rigor up front (requirements); increased funding stability; and the need
for block upgrades since preplanned product improvements are very
difficult for space systems, particularly for space craft. The team also
noted that space programs tend to have *inclusive* requirements supporting
multiple DOD and warfighting needs with many mission partners.

A range of actions are being undertaken by DOD and individual military
services to streamline space acquisition. For example, the Air Force has
developed a new space system acquisition process designed to shorten
timeframes for technical assessments and facilitate faster decisionmaking.
This approach will establish key decision points earlier in the
acquisition process, as compared to the acquisition process for nonspace
systems, and will provide more oversight earlier in the development of
complex satellite technology. According to DOD, the new process will
conduct an independent cost estimate as part of the key decision point
(KDP) authorizing the start of the system design effort and will then also
conduct another cost estimate after the design is complete as part of the
KDP prior to the start of system build, test and launch activities. A key
feature of the new process is that it will use an independent program
assessment team composed of members with appropriate expertise to
thoroughly review a space program before each KDP. The assessment will be
done on a full- time basis over a two to four week period in an effort to
perform relevant technical and programmatic reviews in less time than the

traditional, part- time, multi- layered integrated product team approach.
We plan to study DOD*s new space policy as part of our follow- on review
and to assess whether DOD will have adequate knowledge about technology,
design, and costs for making its decisions.

GAO- 03- 825R Satellite Acquisition Programs Page 13 To strengthen space
planning, DOD undertook efforts to develop a plan that would

set overall objectives for space and provide a high- level 10- to 15- year
roadmap for the direction of space program. The plan is expected to be
completed sometime in fiscal year 2003. In response to the Space
Commission*s recommendation, 3 the Secretary of Defense also designated
the Air Force to be the executive agent for space within DOD, with
departmentwide responsibility for planning, programming, and acquiring
space systems. In October 2001, DOD established a *virtual* major force
program for space to increase visibility of resources allocated for space
activities. The virtual major force program identifies spending on space
activities within the other major force programs in DOD*s Future Years
Defense Budget and provides information by functional area. Further, in
recent testimony, the Under Secretary of the Air Force noted that the Air
Force was working with the Director of OSD Cost Analysis Improvement Group
to form a national security space cost assessment team to provide a
useful, accurate, and timely independent cost estimate with common
methodology in support of space acquisition.

We plan to review these and other actions being taken to address satellite
acquisition problems in a subsequent review.

AGENCY COMMENTS

DOD provided technical comments on a draft of this letter. These comments
were largely focused on ensuring technical accuracy in our reporting of
individual systems and providing updated information. We incorporated
these comments where possible. DOD did not comment on our overall
findings.

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

If you or your staff have any questions concerning this report, please
contact me at (202) 512- 4841. Key contributors to this report were
Cristina Chaplain, Jean Harker, Natalie Britton, Bradley Terry and Art
Gallegos.

Katherine V. Schinasi Director, Acquisition and Sourcing Management

3 We recently reported on the status of DOD*s efforts to implement the
Commission*s recommendations. See Defense Space Activities: Organizational
Changes Initiated, but Further Management Actions Needed (GAO- 03- 379,
April 2003).

GAO- 03- 825R Satellite Acquisition Programs Page 14 Appendix I Profiles
of Satellite Acquisitions This appendix profiles satellite programs
covered by GAO reviews during the past

two decades. It also profiles two launch systems, given their importance
to the success of satellite programs. Among other things, the profiles
describe the programs*

Mission

Primary users

Manager

Architecture and key technologies

Contractors/ contract type

Original cost/ quantity and current cost/ quantity 4 Total spent/ percent
total spent 5 The profiles also identify key GAO findings related to
requirements, investment planning, acquisition strategy, and technology. A
summary of these findings and our report coverage are highlighted below.
In addition to analyzing past GAO reports, we also relied on DOD Selected
Acquisition Reports to the Congress and several DOD studies.

Table I. 1 Summary of GAO Coverage and Key Findings Mission Program
Require-

ments Invest-

ment Acquisition Technology

DSP proposed replacements (FEWS, ALARM) a a a SBIRS High a a a a Missile

Warning/ Tracking

STSS (including Brillant Eyes, SMTS, SBIRS- Low) a a a a Milstar a a a
DSCS a AEHF a a a a WGS (new effort, covered in a recent,

broader GAO assessment of major weapon system programs)

a Communi- cations

AWS (new effort, covered in a recent, broader GAO assessment of major
weapon system programs)

a Navigation GPS a a Weather NPOESS a a Launch Titan IV a a EELV a a a a 4
Original and current cost estimates were inflated from the base year
reported in the SAR to 2003 current dollars using DOD escalation factors.
For older SARs with very early base years such as DSP, inflating the
dollar amounts may be subject to error based on accuracy of escalation
factors. In some cases, DOD provided us with updated cost information. 5
Total dollars spent were inflated from the year the SAR was issued to 2003
dollars. The percent total spent was taken from the latest SAR available
and was not calculated by GAO. In some cases, DOD provided us with updated
cost information.

GAO- 03- 825R Satellite Acquisition Programs Page 15

Mission: Missile Warning Program: Defense Support Program (DSP) and early
proposed replacements

Background information DSP is a strategic surveillance and warning
satellite system with an infrared capability to detect ballistic missile
launches (intercontinental and submarine- launched). It provides near
real- time detection information in support of DOD*S integrated tactical
warning and attack assessment (ITW/ AA) mission. DSP began in 1967, and
the first operational satellite was deployed in 1971. The most recent DSP
satellite launch (number 21) was in August 2001. In the late 1970s, DOD
decided that DSP should be replaced since the system did not satisfy all
the validated military requirements for a space- based ITW/ AA sensor. It
followed this decision with several attempts to develop replacement
systems, but these efforts failed due to high costs and technology
immaturity. DOD eventually made enhancements to DSP. The SBIRS- High
program is focused on replacing DSP. Architecture/ Key Technologies

The number of DSP satellites in orbit is classified SECRET. DSP satellites
use infrared sensors to detect heat from missile and booster plumes
against the earth's background. Over the last 29 years, there have been
five major design changes. Historically, DSP satellites have been launched
atop the Titan III & IV family of launch vehicles; one was launched aboard
the Space Shuttle. Currently, DSP

satellites are launched into geo- synchronous orbit using a Titan IV- B
launch vehicle with an Inertial Upper Stage. DSP Flight 23 will be
launched on an Evolved Expendable Launch Vehicle (EELV).

Users Strategic and tactical forces across military services Original
cost/ quantity:

$10.8 billion 19 satellites

Manager Air Force Current cost/ quantity:

$14.7 billion 23 satellites

Contractors/ Contract type

TRW for satellites Fixed price with Incentive Gencorp, Aerojet for sensors

Fixed price with Incentive

(note: contract has since changed)

Total spent/ % total spent

$7.8 Billion 75. 4% Source: 12/ 31/ 1996 Selected Acquisition Report (all
dollar amounts in 2003 dollars) and DOD

provided updates

Key Issues Affecting Program

Technology immaturity

Unanticipated costs

Lack of adequate analysis of alternatives Note: Issues mostly affecting
DSP replacement programs

Chronology of Key Findings

1992 GAO reported that DOD was not adequately analyzing alternatives to
DSP. DOD first proposed replacing DSP with a system called the Advanced
Warning System (AWS), but this proposal never fully materialized because
of immature technology and high costs. A subsequent proposal, the Boost
Surveillance and Tracking System was discontinued after DOD decided to
pursue other technologies for tracking ballistic missiles. AWS was
proposed for remaining tactical warning and attack assessment missions in
1990 but was later scaled down to a less costly and less capable system
called the Follow- on Early Warning system (FEWS). GAO reported that while
the current proposal for FEWS may provide more capability than the
existing DSP system, DOD still needed to consider other alternatives,
including an enhanced DSP which

could be nearly as effective and cost billions dollars less than a fully
capable FEWS. Several DOD studies supported this point. 1993 GAO reported
that adding global processing capability* which would enable processing of
data generated by the satellite constellation network to be done in a
single station-- in upgrades to ground processing stations for DSP might
not be cost- effective. One reason was that there were

GAO- 03- 825R Satellite Acquisition Programs Page 16 no corresponding
plans to reduce the number of ground stations. Another reason was that

operational requirements were not yet complete.

1994 GAO reported that Congress had appropriated $515 million for FEWS for
fiscal years 1992 through 1994, but terminated the program in late 1993
based on affordability reasons. In late 1994, the Air Force selected ALARM
(Alert, Locate, and Report Missiles system) to be DSP*s replacement. ALARM
was to be smaller than DSP and less capable than FEWS with an emphasis on
greater support to tactical forces. At the time of GAO*s review, concerns
were that DOD was

about to make a substantial investment in ALARM without fully defining
operational requirements. Moreover, while DOD cost estimates showed ALARM
to be more affordable than FEWS in the short term, the total life cycle
costs lead GAO to question whether ALARM, with projected upgrades, would
actually be a more expensive system.

1994 GAO reported that the Air Force plans to accelerate ALARM schedule by
2 years from 2004 to 2002 could add costs to the program which in turn
could put DOD in a similar unaffordable position when it rejected the FEWS
program. At the time, the program office had identified an additional $434
million that would be needed to support the new schedule. Accelerating
schedule could also save as much as $700 million because it could obviate
the need to procure an additional DSP satellite, its launcher, and an
inertial upper stage. However, acceleration could also create program
risks by shortening the demonstration and validation phase of the
acquisition process by 10 months and performing the critical design review
a full year ahead of the original schedule. Air Force officials contended
that previous engineering efforts on DSP earlier

replacement programs provided enough experience to offset this risk.

1994 GAO reported that funds for developing two critical technologies for
ALARM* infrared focal plane array and radiation- hardened electronics*
were frozen. Contractors stated that no private sector funds would be
available for these technologies.

2003 CRS report recapped history of DSP, noting that none of the proposed
replacement programs reached fruition, and instead, enhancements were made
to the DSP series. For example, DSP was designed to detect launches of
strategic long range missiles (such as intercontinental ballistic
missiles) but following the Persian Gulf War DOD recognized that the
threat was changing from intercontinental ballistic missiles to tactical
missiles like the SCUD- C. In 1995, DOD added the ALERT (Attack and Launch
Early Reporting to Theater) system, a

ground- processing center that uses DSP data, to augment its missile
warning capabilities.

GAO Reports GAO/ NSIAD- 92- 39, GAO/ NSIAD- 93- 148, GAO/ T- NSIAD- 94-
108, GAO/ T- NSIAD- 94- 164, GAO/ NSIAD- 94- 253

GAO- 03- 825R Satellite Acquisition Programs Page 17

Mission: Missile Warning

Program: Space Based Infrared System- High (SBIRS- High) Background
information The SBIRS system was initiated in 1994 as an effort to replace
DSP, the current system used to detect missile launches. Until recently,
SBIRS had two components: SBIRS- High, which would consist of launch
detection satellites in geo- synchronous and highly elliptical orbits and
SBIRS- Low which would consist of launch detection and tracking satellites
in low earth orbits. In 2000, SBIRS- Low was shifted back to the Ballistic
Missile Defense Organization, which is now the Missile Defense Agency.
SBIRS- Low is primarily focused on supporting the missile defense mission.
SBIRS- High is being managed by the Air Force. It is focused on missile
warning, missile defense, technical intelligence, and battlespace
characterization.

Architecture/ Key Technologies

SBIRS- High features a mix of four geo- synchronous earth orbit (GEO)
satellites and a spare, two highly elliptical earth orbit (HEO) payloads,
and associated ground hardware and software. SBIRS- High will have both
improved sensor flexibility and sensitivity over DSP. Sensors will cover
short- wave infrared like its predecessor, expanded mid- wave infrared and
see- to- the- ground bands allowing it to perform a broader set of
capabilities as compared to DSP. Currently in the engineering,
manufacturing, and development phase, the first SBIRS- High HEO payload is
scheduled for delivery in 2003 and the first GEO satellite is expected to

launch in 2006. Users Strategic and tactical forces across military
services Original cost estimate/ quantity:

$4. 1 billion 5 satellites Manager Air Force Current

cost/ quantity:

$8. 5 billion 5 satellites Contractors/

Contract type

RDT& E -- SBIRS- High EMD Mod: Lockheed Martin Space Systems Cost Plus
Award Fee October 1995

(note: contract has since changed) Total spent/ % total spent

$3. 0 billion 34. 9% Source: 12/ 31/ 2002 Selected Acquisition Report (all
dollar amounts in 2003 dollars) and DOD provided updates

Key Issues Affecting Program

Requirements definition

Technology immaturity

Unanticipated software growth Significant cost growth

Schedule delay Program instability Chronology of Key Findings

1995- 2001 GAO reports found the program was facing serious hardware and
software design problems including sensor jitter, inadequate infrared
sensitivity, and stray sunlight.

2001 DOD selected acquisition report stated the program experienced
significant cost growth and schedule delays. Driven by poor cost and
schedule performance and the contractor's projection of a fiscal year 2002
funding shortfall, the System Program Office and Lockheed Martin Space
Systems Company (LMSSC) completed a preliminary Estimate at Completion
(EAC) exercise in October 2001. The preliminary EAC results indicated
potential cost growth in excess of $2 billion across the Engineering and
Manufacturing Development contract and schedule delays of 12 to 36 months.

GAO- 03- 825R Satellite Acquisition Programs Page 18

2001 Secretary of Air Force reported a Nunn McCurdy Unit Cost Breach (10
U. S. C. 2433) exceeding 25 percent to Congress. House Appropriations
Committee report (House Report 107- 298) cited scheduling, cost, and
technology problems, including unanticipated software code growth, high
number of discrepancy reports in ground mission software, unbudgeted
payload redesign activities, notable schedule slippages.

2002 An Independent Review Team (IRT) was chartered by DOD to look at the
reasons behind significant cost increases, and program management and
execution problems affecting the program. Key root causes identified
included: (1) the program was too immature to enter system design and
development, (2) system requirements decomposition and flowdown were not
well understood as the program evolved, and (3) there was a significant
breakdown in execution management. 2002 IRT reported that in general, the
complexity, schedule, and resources required to develop SBIRS were, in
hindsight, misunderstood. This led to an immature understanding of how
requirements translate into detailed engineering solutions. In addition,
the requirements setting process was often ad hoc with many decisions
being deferred to the contractor. While SBIRS- High adopted a more
commercial approach to doing business within the defense related industry*
the winning contractor assumed Total System Performance Responsibility
(TSPR) for the integrated architecture* TSPR was not properly understood
or implemented on the SBIRS- High program. The way TSPR was initially

applied circumvented traditional program management and integrated product
team roles and responsibilities. 2002 IRT also observed that there had
been far too much instability on the program since the contract award. In
a 5- year timeframe, the program underwent four major replanning efforts
and four program directors. The team acknowledged that corrective actions
were being taken on the program, but noted that there were still
significant risks within the program, including risks related to the
schedule for first high- elliptical orbit launch and ground software.

2002 Under Secretary of Defense for Acquisition, Technology, and Logistics
certified SBIRS- High to Congress as essential to national security, no
alternatives offering equal or greater military capability at same or
lower costs existed, new cost estimates were reasonable, and management
structure was adequate to manage and control unit costs.

2003 CRS reported that SBIRS- High has become controversial because of
cost growth and schedule slippage caused by technical challenges that have
been encountered in developing the sensors and satellites.

2003 GAO reported that three critical technologies* the infrared sensor,
thermal management, and onboard processor* are now mature. When the
program began in 1996, none of its critical technologies were mature. GAO
could not assess design stability relative to best practices, because
program was not

tracking the number of releasable drawings and did not know how many total
drawings were expected for SBIRS- High. However, GAO reported that design
stability has been an issue for this program. GAO could not assess
production maturity relative to best practices because the contractor does
not use statistical process control to assure that production processes
are stable. GAO Reports Three reports from 1995- 2001 and GAO- 03- 476

GAO- 03- 825R Satellite Acquisition Programs Page 19

Mission: Missile Warning/ Tracking Program: Space Based Infrared System-
Low (SBIRS- Low); now known as the Space Tracking and Surveillance System
(STSS) Background information STSS started in 1990 as Brilliant Eyes, was
transferred in 1993 from the Ballistic Missile Defense Organization (BMDO)
to the Air Force and renamed the Space and Missile Tracking System (SMTS).
In 1994, DOD terminated the SMTS program, consolidated its infrared space
requirements, and selected SBIRS as a *system of systems* approach with
two components: SBIRS- High, which would consist of launch detection
satellites in geo- synchronous and highly elliptical orbits and SBIRS-
Low, which would consist of launch detection and tracking satellites in
low earth orbits. In 2000, SBIRSLow was shifted back from the Air Force to
the BMDO, which is now the Missile Defense Agency (MDA). In 2002, SBIRS-
Low was renamed STSS. While STSS is primarily focused on supporting the

missile defense mission, SBIRS- High is focused on missile warning,
missile defense, technical intelligence, and battlespace characterization
and is managed by the Air Force. Architecture/ Key Technologies

STSS is a capabilities- based development. STSS will build a few
satellites at a time with later satellites being more capable than earlier
ones. Using the advantage of a lower operational altitude, STSS will track
tactical and strategic ballistic missiles against the cold background of
space. The satellite*s sensors will operate across long and short- wave
infrared, as well as the visible light spectrum. These wavebands allow the
sensors to acquire and track missiles during the boost phase as well as in
midcourse. STSS is expected to launch its first satellites in 2007. Users
Strategic and tactical forces across military services Original cost
estimate/ quantity:

Not Available Manager Missile Defense Agency Current cost/ quantity:

Quantity undetermined but more than 20 satellites would be needed for
worldwide coverage Contractors/

Contract type

Prime Contractor: Northrop Grumman Cost Plus Award Fee Total spent/

% total spent

Not available Source: GAO analysis.

Key Issues Affecting Program

Requirements definition

Technology immaturity

Lack of competition

Cost growth

Inadequate analysis of alternatives

Note: Problems mostly affecting past SBIRS- Low efforts

Chronology of Key Findings

1997 GAO assessed various options for accelerating SBIRS- Low deployment
date, which had been set for 2006, given congressional concerns about
direction of the program. GAO reported that moving up the date by 3 or 4
years would result in high program risk because of the high degree of
concurrent activities between planned flight demonstrations and
development and fabrication of satellites. Additional funding might also
be required. Moving up the date 2 years would reduce the need for
concurrency, and therefore lower risks, but still require additional funds
to account for schedule compression. Moving up the date 1 year would
reduce scheduling risks and could require less funding. DOD subsequently
changed deployment date

to 2004.

2001 GAO reported that SBIRS- Low acquisition schedule was at high risk of
not delivering the system on time or at cost or within expected
performance because satellite development and production, for example, was
expected to be done concurrently. SBIRS- Low program also had high
technical risks because some critical satellite technologies were judged
to be immature for the current stage of the program, including scanning
infrared sensor, tracking infrared

GAO- 03- 825R Satellite Acquisition Programs Page 20 sensor, and
technologies used to cool down satellite systems.

2001 GAO also reported that DOD was not adequately analyzing or
identifying cost- effective alternatives to SBIRS- Low that could satisfy
critical missile defense requirements, such as a Navy ship- based radar
capability. At the time, other studies supported the possibility that
other types of sensors could be used to track missiles in midcourse of
their flight and to cue interceptors. Subsequent to 2001 GAO report, DOD
restructured the SBIRS- Low program because of cost

and scheduling problems, and put the equipment it had partially built into
storage. In 2000, the Congress directed the Air Force to transfer the
program to the Ballistic Missile Defense Organization (now MDA). DOD was
also directed to study alternatives (such as ground- based radar systems)
to SBIRS- Low. May 2003 GAO reported that DOD believed that a
discrimination capability (that is, the ability to

detect and track multiple objects and differentiate the threatening
warhead from decoys) would significantly enhance a space- based missile
tracking system like STSS. However, DOD deferred plans to achieve this
capability for STSS given technical challenges. GAO also reported that
DOD's unwillingness to relax requirements for capabilities such as
discrimination during earlier

SIBRS- low efforts contributed to cost and scheduling problems. May 2003
GAO reported that in taking on the restructured SBIRS- Low program, now
called Space Tracking and Surveillance System (STSS), MDA purposely set
out to adopt a strategy that would evolve STSS over time, deferring some
requirements, and calling for competition in development of sensors aboard
the satellite. However, recent decisions were limiting MDA*s ability to
achieve its original goals as well as knowledge that could be gained from
its satellite demonstrations. For example, plans were eliminated to have
contractors compete for production of the sensor to detect missile
launches. If it chose to keep STSS as part of the missile defense system,
STSS could end up being more expensive in the future because MDA could be
locked into a single contractor for the design and product of the larger
constellation of satellites. May 2003 GAO reported that MDA was focused on
launching its satellites by 2007 in order to assess its performance in the
missile defense tests. However, it made this decision without

completing its assessment of the working condition of the equipment it
planned to assemble and use to demonstrate STSS capabilities. Also, MDA
was not considering other approaches to demonstrating capabilities because
they would not allow STSS to participate in 2006- 2007 missile defense
tests. These include (1) launching satellites in 2008 instead of 2007 and
(2) dropping effort to demonstrate capabilities with legacy satellites
that were based on older technology and focusing instead on developing new
technology. Both approaches would enable MDA to inject more competition
into STSS program, reduce scheduling risks, and demonstrate more
capabilities. However, they also have drawbacks; primarily, they would

delay MDA*s ability to make informed tradeoffs on missile defense sensors.

GAO Reports GAO/ NSIAD- 97- 16, GAO- 01- 6, GAO- 03- 597

GAO- 03- 825R Satellite Acquisition Programs Page 21

Mission: Current Communication Systems

Programs: Defense Satellite Communication System (DSCS) and Milstar

Background information DSCS and Milstar are current DOD communication
satellite systems that provide protected communications to support
globally distributed military users. The Air Force began launching the
current DSCS satellites in 1982. The Air Force initiated the Milstar
program in 1981, but the first Milstar satellite was launched in 1994 and
the last one in April 2003. Architecture/ Key Technologies

Currently, ten DSCS satellites and five Milstar satellites operate in geo-
synchronous orbit. The DSCS satellites utilize super high frequency
transponder channels that provide the highest data capability, but require
large antennas (4 to 60 feet) for receiving large amounts of data. The
Milstar satellites utilize extremely high frequency transponder channels
that provide low to medium data rate communications but require small
antennas (5 inches to 10 feet) and provide communications that are more
survivable and resistant to jamming than the DSCS. The Milstar satellites
are launched

onthe Titan IV and weigh about 10,000 pounds. The last two DSCS satellites
will be launched by the EELV and weigh about 2,500 pounds.

Users Strategic and tactical forces across military services Original
cost/ quantity:

Not Available * Milstar $1.7 billion -DSCS 14 satellites

Manager Air Force Current cost/ quantity:

Not Available * Milstar $2.7 billion- DSCS 14 satellites

Contractors/ Contract type

RDT& E -- Milstar II Satellites: Lockheed MSL & Space Co October 1992

Cost plus award fee DSCS III Production: General Electric Co November 1984
Firm fixed price Total spent/%

total spent

Not Available * Milstar and DSCS Source: Milstar * 12/ 31/ 1999 Selected
Acquisition Report and DSCS * 9/ 30/ 91 (all dollar amounts in

2003 dollars)

Key Issues Affecting Programs

Cost growth Requirements changes Chronology of Key Findings

1986 GAO reported that in late 1982 the Air Force realized that the
Milstar configuration could not be achieved given existing schedule and
budgetary constraints. As a result, the program office began rescoping the
program to conform to the budgetary constraints in a design- to- budget
exercise. In 1983 the program office rescoped the program for a second
time* this time adding requirements due to user input and concerns.

1986 GAO reported that DOD revised the acquisition strategy from a total
system integration package to an associate contractor approach because the
teaming of TRW and Hughes (they had previously performed the majority of
extremely high frequency work) presented an insurmountable challenge to
other contractors. Under the associate contractor approach, rather than
contracting for the whole system with a prime contractor, the government
contracts with different firms for components of the system.

GAO- 03- 825R Satellite Acquisition Programs Page 22

1992 GAO reported that the National Defense Authorization Act for FY1991
directed the Secretary of Defense to develop or carry out a plan for
either a restructured Milstar or an alternative advanced communications
satellite program that would substantially reduce program costs. DOD chose
to restructure the program and lower costs by reducing the constellation
size from 8 to 6 satellites, the number of control stations from 25 to 9,
and the number of terminals from 1, 721 to 1,467. To provide greater
system utility to tactical forces, DOD decided to add a medium data rate
capability to the satellite (this would increase the volume of information
that

could be processed through the satellites).

1992 GAO reported that some satellite issues related to the Army*s
tactical use of Milstar had not been resolved. For example, formal
agreement had not been reached on sufficient capacity that the Army
claimed it needed. While DOD expected the medium data rate capacity to
allow about

40 million bits of information to be passed through the satellite each
second, Army representatives stated that to satisfy critical Army
communication requirements, at least 34. 4 million bits per second would
be needed* about 86 percent of the total planned throughput capacity for
each satellite. After considering the multiservice aspects of the Milstar
program, the Army concluded that to justify its participation in the
Milstar program, the minimum throughput

capacity acceptable would be 30.7 million bits per second* about 77
percent of the total planned capacity for each satellite. The remaining
capacity would be allocated among the Air Force, the Navy, and the Marine
Corps.

1993 GAO reported that in 1991 as directed by Congress, DOD published its
military satellite communications architecture study that identified 12
alternatives for various communications approaches that ranged from using
all commercial to all military satellite programs. From among the 12
alternatives, DOD selected an all military approach consisting of existing
systems. GAO reported that DOD did not select one alternative, the dual
common bus that provided a better way to demonstrate advanced
technologies.

1994 GAO reported that in response to our 1993 report, DOD agreed with the
need to move away from customized, unique busses toward common busses and
stated that the most cost effective approach for inserting modern
technology was to begin developing an advanced, lower cost, lower weight
payload capability. 1994 GAO reported that congressional directives and
national policy emphasized greater use of commercial satellite services to
reduce costs of military satellite services. However, a new criterion used
by DOD for establishing communication requirements reduced general purpose
requirements by over 40 percent. This change has reduced the potential for
using commercial satellite communication services. (It should be noted,
according to DOD officials, that there were some pointed objections in the
past year to the DOD's use of commercial satellite systems such as
INTELSAT and INMARSAT because they were "part owned" by countries such as
Iraq and Iran.)

1997 GAO reported that during the next decade, DOD anticipated a
significant increase in its high- capacity satellite communications (DSCS)
because of the shift in the national military strategy and availability of
advanced technologies. DOD planned to replenish the existing DSCS

constellation during fiscal year 1997- 2003 with the five satellites
remaining in inventory. DOD was modifying four of these satellites to
double each satellite*s capacity from 100 megabits per second (MBPS) to
about 200 MBPS and to replace potentially defective parts with improved
electronic components. Even so DSCS*s replenishment satellites were not
expected to keep pace with the projected requirements, thus an alternative
would have been to lease satellite communications from commercial
providers. However, according to DOD analysis, commercial

leasing was more costly than acquiring equivalent commercial like
capabilities.

1999 GAO reported that in 1998 a draft operational test report identified
four limitations associated with Milstar I capabilities to support
strategic missions. While DOD had identified corrective actions, final
resolutions were dependent on approval of requirements, verification
through testing, a certification process, or obtaining necessary funds.
Regarding tactical missions, the Air Force had encountered schedule delays
related to software development for a critical Milstar component* called
the automated communications management system* that

GAO- 03- 825R Satellite Acquisition Programs Page 23 could adversely
affect Milstar II*s timely support to tactical forces.

2003 GAO reported that in 2000, DOD recognized the need to address the
capabilities and coverage gap caused by the April 1999 Milstar launch
failure and adopted a high- risk accelerated schedule for the Advanced
Extremely High Frequency (AEHF) satellite system.

GAO Reports GAO/ NSIAD- 86- 45S- 15, GAO/ NSIAD- 92- 121, GAO/ T- NSIAD-
92- 39, GAO/ NSIAD- 94- 48, GAO/ NSIAD- 97159, GAO/ NSIAD- 99- 2, GAO/
NSIAD- 93- 216, GAO/ T- NSIAD- 94- 108, GAO/ T- NSIAD- 94- 164, GAO/
NSIAD94- 253

GAO- 03- 825R Satellite Acquisition Programs Page 24

Mission: Planned Communication Systems

Programs: Advanced Extremely High Frequency (AEHF) Communications
Satellite, Wideband Gapfiller Satellite (WGS), and Advanced Wideband
Satellite (AWS) Background information The current military satellite
communications network represents decades- old technology. To meet

the heightened demands of national security in the coming years, newer and
more powerful systems are being developed. The AEHF is a satellite system
intended to replace the existing Milstar system and to be DOD*s next
generation of higher speed, protected communication satellites. WGS will
augment communications services currently provided by the Defense
Satellite Communications System (DSCS), which provides super high
frequency wideband communications. WGS will provide an interim solution to
assure DOD*s existing worldwide communication support is maintained until
the development and deployment of the Advanced Wideband Satellite System
(AWS) also known as TSAT. AWS is intended to become the cornerstone of
DOD*s future communications architecture that includes supplementing the
AEHF system and replacing the WGS system.

Architecture/ Key Technologies

AEHF started in 1998 and the constellation will consist of three
satellites in low inclined geosynchronous orbits (requirements still call
for five satellites- four operational and one spare) that can transmit
data to each other via cross- links. AEHF entered the Engineering
Manufacturing Development/ Production acquisition phase in November 2001.
Each satellite will be launched with the Evolved Expendable Launch Vehicle
(EELV); the initial launch is planned for December 2006. WGS started in
2001 and the constellation was planned to have 3 satellites, but the
program recently added two more satellites because the initial capability
of AWS, which is intended to replace AEHF

and some aspects of WGS, may not be able to support all the super high
frequency services that the users require. Thus additional WGS spacecraft
are being acquired to bridge this gap. WGS combines commercial
capabilities* phased array antennas and digital signal processing
technology* into a

flexible architecture that will allow WGS to evolve and satisfy the
growing wideband communication requirements of the warfighter. WGS is
currently in full rate production with the first satellite scheduled for a
June 2004 launch aboard an EELV vehicle. AWS* final configuration has not
yet solidified under ongoing milsatcom transformational efforts, but

the concept is one of applied technology and engineering that will remove
capacity as a constraint on warfare communications. AWS plans to take
advantage of the commercial and government technology advances of the
first half of this decade to meet expected needs. Some of the

technologies that AWS plans to use are laser crosslinks, space- based data
processing and routing systems, and highly agile multibeam/ phased- array
antennas. DOD plans for the program to enter product development in
October 2003 with the first satellite to be launched at the end of 2009. A
key program review is planned for November 2004 to determine if sufficient
technology development has

occurred to warrant continuing the program at its planned schedule or
whether the 4 th and 5 th AEHF satellites should be acquired.

Users Military Strategic and Tactical Original cost/ quantity:

$5.4 billion - AEHF 5 satellites

$1.0 billion * WGS 3 satellites

GAO- 03- 825R Satellite Acquisition Programs Page 25

Manager Air Force Current cost/ quantity:

$4.8 billion -AEHF 3 satellites

$1.5 billion -WGS 5 satellites

Contractors/ Contract types

AEHF*s system development and demonstration: Lockheed Martin November 2001
Cost plus award fee WGS* RDT& E and procurement: Boeing Satellite Systems
January 2001 Firm fixed price

Total spent/ % total spent

$1.1 billion - AEHF 21. 2% $. 28 billion * WGS

17. 5% Source: AEHF and WGS -12/ 31/ 2002 Selected Acquisition Report (all
dollars amounts in 2003 dollars)

Key Issues Affecting Programs

Cost growth

Scheduling risks

Requirements Changes

Immature technology

Note: Problems reported primarily affect AEHF.

Chronology of Key Findings

AEHF

2002 DOD selected acquisition report commented on funding cuts. In fiscal
year 2002, AEHF sustained a $70 million fiscal year 2002 congressional
reduction to RDT& E funding. The AEHF space segment was a firm fixed price
contract. According to DOD, this sizable reduction would

likely result in a six- month launch delay to satellites l- 3, breach of
initial operational capability and a significant overall program cost
increase. In 2002, the Deputy Secretary of Defense decided to change the
acquisition strategy of AEHF from a 5- satellite program to a 3- satellite
program. Under the revised strategy, full capability may no longer be
satisfied by an AEHF- only constellation. (According to DOD officials, the
current DOD plan is to meet the full AEHF operational capability
requirement with three AEHF

spacecraft and a combination of one or two AWS spacecraft and zero, one or
two Advanced Polar System spacecraft * this plan is driving the AWS first
launch date of late 2009.)

2003 GAO reported in the early phases of the program, DOD substantially
and frequently altered its requirements; the system design changed. While
considered necessary, some changes increased costs by hundred of millions
of dollars and caused scheduling delays.

2003 GAO reported that in December 1999, the two contactor teams that had
been awarded engineering manufacturing and development contracts a few
months earlier offered to form a *national team* to accelerate the AEHF
program. DOD agreed to the national team proposal even though DOD
recognized it meant lack of benefits from competition.

2003 GAO reported that once DOD decided to accelerate its plans to build
the satellites, the contractors proposed and DOD agreed to support a high-
risk schedule that turned out to be overly optimistic and highly
compressed* leaving little room for error and depending on a chain of
events taking place at certain times. Substantial delays occurred when
some events, such as the award of the contract or the availability of
equipment, did not occur on time. In commenting

on the AEHF report, DOD noted the decision to accelerate the program was
based on a satellite constellation gap caused by the loss of a Milstar
satellite. DOD also stated many in DOD expressed concern about the risks,
but believed the risk was acceptable based on information known at the
time.

2003 GAO reported that at the time DOD decided to accelerate the program,
it did not have the funding needed to support the activities and manpower
needed to design and build the satellites

GAO- 03- 825R Satellite Acquisition Programs Page 26 quicker. The lack of
funding also contributed to schedule delays, which in turn, caused more
cost increases.

2003 GAO reported that the program demonstrated most technology knowledge
at development with 11 of 12 critical technologies having reached maturity
according to best practice standards. However, the program office did not
project achieving maturity on the remaining technology*

the phased array antenna* by the design review in June 2004 and did not
have a backup capability. Program officials assessed the software
development for the mission control system as moderate risk and have
developed a risk mitigation strategy. However, until these mitigation
actions are completed, software may be at risk for unplanned cost and
schedule growth.

2003 GAO reported that significant design changes affected cost and
delayed the AEHF schedule. For example, software growth occurred as more
requirements were added and as the design of the system stabilized. These
increases in software requirements for both the satellite and the mission
control segments increased the software cost estimate by over 77 percent
or about $223 million.

2003 GAO reported in the area of production maturity that any future
problems with the fabrication of the communications and transmission
security microprocessor, a component designed to limit access to satellite
transmissions to authorized users, could delay the production schedule and
the launch of the first satellite planned for December 2006.

WGS

2003 GAO reported that WGS* critical technologies, design, and production
processes are mature. DOD plans to rely on commercial technologies that
will not require extensive product development. Program officials were
concerned about WGS production risk that was to be reduced during
production of commercial satellite orders. However, due to drastic loss of
commercial satellite orders, only one commercial satellite with similar
technologies as WGS is now leading WGS in the manufacturing schedule.
Recently identified problems found on the *leader* program will impact WGS
manufacturing schedule and might result in a first launch schedule delay
of four to six months.

2003 GAO reported that the 4 th and 5 th satellites have been directed by
DOD to be launched in fiscal year 2009 and fiscal year 2010 respectively.
These dates are outside the allowable dates of the WGS contract options
clauses and will require renegotiation to finalize their cost. These later

launch dates could result in cost increases to compensate for loss of
learning curve from over a three- year break in production, parts
obsolescence, and inflation. AWS

2003 GAO reported that AWS is scheduled to enter product development with
only one of its five critical technologies mature. The four immature
technologies are scheduled to reach maturity by January 2006, more than
two years after development start. Three of the four technologies have

a backup technology in case of development difficulties. However, the
Single Access Laser Communications technology has no backup and according
to program officials any delay in maturing this technology would result in
a slip in the expected launch date.

2003 GAO reported that the program plans an aggressive development cycle
even though the AWS is expected to provide a transformational leap in
satellite communications capability.

GAO Reports GAO- 03- 476, a report that covers multiple systems, and an
AEHF report in 2003.

GAO- 03- 825R Satellite Acquisition Programs Page 27

Mission: Navigation Program: NAVSTAR Global Positioning System (GPS)

Background information GPS is a space- based radio- positioning system
nominally consisting of a 24- satellite constellation that provides
navigation and timing information to military and civilian users
worldwide. The full constellation of GPS satellites has been operational
for 7 years. Total program investment over a 43year period (through 2016)
is estimated at $18.4 billion.

Architecture/ Key Technologies

GPS satellites, in one of six medium earth orbits, circle the earth every
12 hours emitting continuous navigation signals on two different
frequencies. In addition to the satellites, the system consists of a
worldwide satellite control network and GPS receiver units that acquire
the satellite*s signals and translate them into precise position and
timing information. Four generations of GPS satellites have flown in the
constellation: the Block I, the Block II, the Block IIA, and the Block
IIR. Block I satellites were used to test the principles of space- based
navigation, and lessons learned from these 11 satellites were incorporated
into later blocks. Block II, IIA and IIR satellites make up the current
constellation. Block IIRs began replacing older Block II/ IIAs in 1997.
There are currently eight Block IIR satellites on orbit and they have
reprogramable satellite processors enabling problem fixes and upgrades in
flight. Up to eight IIR satellites are being modified to radiate both a
new civil signal (L2C) and a new military signal (M- Code) for a more
robust and capable signal structure. The first

modified Block IIR (designated as the IIR- M) is planned for launch in
2004. Block IIF satellites are the next generation of GPS satellites.
Block IIF provides all the capabilities of the previous blocks with some
additional benefits as well. Improvements include an extended design life
of 12 years, faster processors with more memory, and a new civil signal on
a third frequency. The first Block IIF

satellite is scheduled to launch in 2006. The Delta II has launched the
Block II, IIA, and IIR satellites, and the EELV (Delta IV and Atlas V)
will launch the Block IIF satellites.

GPS Blocks IIF and IIR

Users Military and Civilian Original cost/ quantity:

$5.3 billion 33 satellites

Manager Air Force Current cost/ quantity:

$5.8 billion 37 satellites

Contractors/ contract type GPS IIF OCS/ MOSC development:

BOEING NORTH AMERICAN, April 22, 1996 Cost Plus Award Fee Block IIR SAT
development:

Lockheed Martin August 2000 Firm fixed price/ cost plus incentive

Total spent/ % total spent

$2.3 billion 39. 7% Source: 12/ 31/ 2002 Selected Acquisition Report (all
dollar amounts in 2003 dollars) and DOD

provided updates

Key Issues Affecting Program

Cost Growth

Schedule risk

Component reliability problems

GAO- 03- 825R Satellite Acquisition Programs Page 28

Chronology of Key Findings

1980 GAO reported program cost (to acquire and maintain the program
through the year 2000) increased from $1. 7 billion to $8. 6 billion due
largely to estimates not previously included for replenishment satellites,
launches, and user equipment. Beginning in 1983, DOD planned to use the
Space Shuttle to launch the NAVSTAR satellites. In the event of Space
Shuttle problems, Atlas or Titan launches would need to be used as an
alternative at an additional cost of $12

million to $38 million per satellite launch. The original full operational
capability date of August 1985 slipped 25 months. 1980 GAO reported that
survivability of GPS satellites was a concern due to Soviet testing of an

anti- satellite system and reliability of GPS satellite atomic clocks
emerged during the demonstration and validation phase when 80 percent
either failed or acted abnormally. 1983 GAO reported that the multiyear
procurement estimate of $1. 4 billion was likely understated because
indications are that the prime contractor would propose a higher cost and
that multiyear procurement savings were not correctly calculated using the
present value analysis method. System design changes were being considered
that would add considerable cost to the program.

The program office expressed concern about the lack of backup launch
vehicles in the event of problems with the Space Shuttle. 1983 GAO
reported that integration testing of the spacecraft with the qualification
test vehicle was scheduled to begin 7 to 18 months after the planned March
1983 award date of the production contract. The consequences of
concurrency could lead to design changes and additional costs. The program
office was considering two design changes to the production spacecraft, a
W- sensor and enhancements related to GPS survivability. 1987 GAO reported
that following the Challenger accident in January 1986, the Air Force
reduced the number of GPS satellites planned for launch on the Space
Shuttle from 28 to 8, because it had awarded a contract to McDonnell
Douglas to build and launch 7 medium expendable launch vehicles with an
option to purchase up to 13 more. 1987 GAO reported GPS acquisition
changes after the Space Shuttle Challenger*s accident: (1) NASA slipped
the date for the first launch schedule for the Block II satellites from
January 1987

to June 1989, (2) since the GPS program was in the production and
deployment phase, the Air Force began stretching out the procurement
process, and (3) the Air Force postponed a planned buy of 20 Block II- R
replenishment satellites because the program office*s estimated need date
for these replenishment satellites had slipped 3 years. 1987 GAO reported
that since development of GPS user equipment (consists of 1-, 2-, and 5-
channel radio receiver sets) was almost 3 years behind schedule due to
technical problems, the Challenger loss caused no further adjustment to
user equipment production. 1987 GAO reported that even though user
equipment technology was changing rapidly with

miniaturized and less costly sets currently available from several
manufacturers, program office officials expressed concern about incurring
substantial costs by changing to the new equipment and that the new
equipment would not meet military specifications. 1991 GAO reported that
DOD postponed full- rate production for receiver sets from March 1989 to
September 1991 due to lingering receiver set reliability problems and
reevaluation of program

requirements. During development testing the Army discovered reliability
problems with the one- and two- channel GPS receiver sets. One 5- channel
set experienced a number of failures during multiservice testing and this
led to a marginal rating of all 5- channel receivers. GAO Reports GAO/
PSAD- 80- 21, GAO/ MASAD- 83- 9, GAO/ NSIAD- 87- 209BR, GAO/ NSIAD- 91- 74

GAO- 03- 825R Satellite Acquisition Programs Page 29

Mission: Weather

Programs: Defense Meteorological Satellite Program (DMSP) and National
Polar- orbiting Operational Environmental Satellite System (NPOESS)

Background information Since the 1960s, the U. S. has operated two
separate polar- orbiting meteorological satellite systems. These systems
are known as the Polar- orbiting Operational Environmental Satellites
(POES), managed by the National Oceanic and Atmospheric Administration
(NOAA), and the Defense Meteorological Satellite Program (DMSP), managed
by DOD. These satellites obtain environmental data that are the
predominate input to numerical weather prediction models* all used by
weather forecasters, the military and the public. Polar satellites also
provide data used to monitor

environmental phenomena as well as data that are used by researchers for a
variety of other studies, such as climate monitoring. Given the
expectation that converging the POES and DMSP program would reduce
duplication and result in sizable cost savings, a May 1994 Presidential
Decision Directive required NOAA and DOD to converge the two satellite
programs into a single program capable of satisfying both military and
civilian requirements. The converged program is called the National Polar-
orbiting Operational Environmental Satellite System (NPOESS).

Architecture/ Key Technologies

DMSP satellites circle the Earth at an altitude of about 500 miles in a
near- polar, sun- synchronous orbit. Each scans an area 1,800 miles wide
and covers the entire Earth in about 12 hours. Pointing accuracy of the
satellites is maintained by four reaction wheel assemblies that provide
three- axis stabilization. The primary sensor on board is the Operational
Linescan System that observes clouds via visible and infrared imagery for
use in worldwide forecasts. A second important sensor is the Special
Sensor Microwave Imager, which provides all- weather capability for
worldwide tactical operations and is particularly useful in typing and
forecasting severe storm activity. DMSP satellites

also carry a suite of additional sensors, which collect a broad range of
meteorological and space environmental data for forecasting and analysis.
Historically DMSP satellites have been launched on Titan II boosters from
Vandenberg Air Force Base with the most recent launch occurring on
December 12, 1999. One more DMSP satellite will be launched on a Titan II
booster. The remaining

four DMSP satellites will be launched on Evolved Expendable Launch Vehicle
(EELV) boosters from Vandenberg Air Force Base. There are two operational
DMSP satellites. NPOESS program acquisition plans call for the procurement
and launch of six NPOESS satellites over the life of the program and the
integration of 14 instruments, including 12 environmental sensors.
Together, the sensors and spacecraft receive and transmit data on
atmospheric, cloud cover, environmental, climate, oceanographic, and
solar- geophysical observations. Additional instruments

are carried to support search and rescue efforts and data collection from
a variety of globally deployed transmitters. NPOESS will be a launch- on-
demand system, and satellites must be available to back up the planned
launches of the final POES and DMSP satellites. The first NPOESS
satellite* designated C1* is scheduled for delivery in late 2009,
according to Air Force officials. Users DMSP focuses on military users.

NPOESS will be available to military, civil, and international users.
Original cost estimate/ quantity:

$5. 6 billion 6 satellites (NPOESS only)

Manager DMSP is managed by the Air Force. NPOESS is managed tri- agency
integrated program office (DOD, DOC, NASA), located within NOAA.

Current cost/ quantity:

$6. 1 billion 6 satellites (NPOESS only)

GAO- 03- 825R Satellite Acquisition Programs Page 30

Contractors/ Contract type

Engineering and Manufacturing Development/ Production and Operations
Northrop Grumman, August 2002 Cost plus award fee/ performance incentive,
Fixed price incentive production options, Fixed price

operation and support options

Total spent/% total spent

$857. 9 million 14.0 percent (NPOESS only)

Source: 12/ 31/ 2002 Selected Acquisition Report (All dollar amounts in
2003 dollars) and DOD provided updates

Key Issues Affecting Program

Requirements definition/ meeting user needs

Technical/ scheduling risks Note: Problems reported affect NPOESS rather
than DMSP

Chronology of Key Findings

1987 GAO reported that the program could save millions of dollars by
converging NOAA and DOD weather satellite programs, which would reduce the
number of satellites from four to three. 1987 GAO reported that NOAA and
Air Force requirements were diverging in several respects, making the
effort to converge the two programs more difficult. For example, NOAA
wanted to change its approach from using expendable convention satellites
to installing sensors on

serviceable platforms. The Air Force plans to continue using its current,
conventional design of DMSPs (expendable and rocket launched) into the
late 1990s before redesigning a new system. NOAA and Air Force also
differed on quality standards for electronic components.

1995 GAO reported that while the planned delivery date for the first
satellite was 2004, transferring two DMSP satellites to NOAA might require
that delivery be accelerated to as early as 2001. Such an action would
increase both technical and schedule risks and require substantial
increases in the convergence program*s near- term budget.

1995 GAO reported that interchangeable components between DMSP and NOAA
satellites were less than earlier estimated. Of 63 platform components,
only 15 (24 percent), such as the inertial measurement unit and earth and
sun sensing equipment, could be used on NOAA satellites without
modifications. Another 13 components (21 percent), such as the power
supply electronics, battery charge assembly, and solar array electronics,
could be used if they were modified, at additional cost. The remaining 35
components (55 percent) were either substantially different or unique and
had no value to NOAA. Additionally, DMSP mission sensors could not be used
because they are unique and would not satisfy NOAA* s requirements.

1997 NPOESS integrated program office determined that there were
scheduling, technical and cost risks associated with the interface data
processing segment and overall system integration and with the space
segment.

2001 DOD selected acquisition report commented on schedule delays being
reported to Congress. Specifically, DOD stated that the Joint Agency
Requirements Group final review of the updated NPOESS requirements took
longer than planned. As a result the engineering and manufacturing
development request for proposal release, initiation of the life cycle
cost estimate update, and the final release of the technical requirements
document were delayed. The milestone decision was moved from February 2002
to August 2002.

2002 GAO reported that technical, schedule, and cost risks were being
reduced by deferring development of requirements, initiating earlier
development of sensors and/ or relying on existing versus new technology,
conducting ground- based demonstrations of data processing system, and
using aircraft to test sensors, among other activities. 2002 GAO reported
that processing centers face challenges in handling the massive increase
in

GAO- 03- 825R Satellite Acquisition Programs Page 31 the volume of data
that would be sent by the new satellites. Whereas current polar satellites

produce approximately 10 gigabytes of data per day, NPOESS is expected to
provide 10 times that amount. Agencies involved in the program were
working to address this problem by improving data management
infrastructure, but more could be done to coordinate and further define
these efforts. 2003 GAO reported that NPOESS entered product development
in August 2002 with most of its

technologies mature. The program also completed a significant portion of
the engineering drawings well in advance of the design review; however,
the total number has yet to be determined. Over 5 years ago, program
officials considered the program to have several high- risk

areas. Since then, officials have implemented several efforts, which are
expected to reduce all program areas to low risk by the first NPOESS
launch, currently scheduled for the 2008- 2009 time frame. GAO Reports
GAO/ NSIAD- 87- 107, GAO/ NSIAD- 95- 87R, GAO- 02- 684, GAO/ NSIAD- 94-
253

GAO- 03- 825R Satellite Acquisition Programs Page 32

Mission: Launch Programs: Titan IV and Evolved Expendable Launch Vehicle
(EELV)

Background information Over the years DOD has used a fleet of expendable
launch vehicles* Delta, Atlas, and Titan* to transport a variety of
satellites into space. The Titan IV is a heavy- lift space launch vehicle
used to carry DOD payloads such as Defense Support Program (DSP) and
Milstar satellites into space. The Titan IV was designed to complement the
National Space Transportation System (Space Shuttle) and serve as an
independent vehicle system to assist in assuring DOD access to space. Air
Force contracted for a total of 41 Titan IV vehicles with the last launch
scheduled for 2004. DOD considers

these launch vehicles to currently operate at or near their maximum
performance capacity and to be very costly to produce and launch. Since
1987, the government has made several attempts to develop a new launch
vehicle, but these attempts were canceled either because of funding
issues, changing requirements, or controversy regarding the best solution.
In 1994, by congressional direction, DOD developed a space launch
modernization plan that led to the initiation of the Evolved Expendable
Launch Vehicle (EELV) program. With EELV, the Air Force hoped to cut its
heavy- lift mission costs by about 50 percent and its overall launch
mission costs by at least 25 percent. The intent of the EELV program was
to develop a family of launch vehicles, using common components, standard
services and supporting systems that would significantly reduce the life-
cycle cost compared to today's systems. Due to a sudden projected increase
in commercial demand that was forecast in 1997, Air Force approved a plan
to develop the Atlas V and Delta IV

EELVs, rather than just one of them. The additional cost of maintaining
two EELV launch infrastructures was intended to be offset by more
competitive pricing. The successful launches of the medium- lift models of
the Atlas V and Delta IV rockets in 2002 fulfilled part of the
engineering, manufacturing, and development segment of the Air Force EELV
contract to Boeing and Lockheed Martin. In the initial launch service
award (1998) Boeing was awarded 19 launch services and

Lockheed Martin was awarded 9 launch services. Current launch services
awards have been modified after the 2000 EELV restructure to 19 missions
for Boeing and 7 missions for Lockheed Martin. Both contractors plan to
deploy their commercial launch service to launch both commercial and
government missions.

Architecture/ Key Technologies

Each Titan launch vehicle is made up of a core, a fairing, and a set of
solid rocket motors. Solid rocket motors along with liquid rockets in the
core provide the propulsion for the Titan IV. The Titan IV may also have
an optional upper stage to provide the additional booster capacity that
some satellite payloads require to reach their intended orbit. The EELV
will use the Delta IV launch vehicle built by Boeing and the Atlas V built
by Lockheed Martin. Boeing developed the RS- 68 liquid- oxygen/ liquid-
hydrogen main engine, for the Delta IV, which is the first cryogenic
engine built in the United

States since the Space Shuttle Main Engine. Lockheed Martin*s main engine,
the RD- 180, is a liquidoxygen/ kerosene engine developed in a joint
venture between NPOEnergomash, a Russian company, and UTC/ Pratt and
Whitney. Users Military satellites are launched by Titan IV Military and
commercial satellites are launched by EELV

Original cost/ quantity:

$14.7 billion - EELV 181 launch vehicles

$3.2 billion - Titan IV 10 launch vehicles

Manager Air Force Current cost/ quantity:

$18.0 billion -EELV 182 launch vehicles

$20.1 billion * Titan IV 39 launch vehicles

GAO- 03- 825R Satellite Acquisition Programs Page 33

Contractors/ Contract type

EELV- Boeing and Lockheed Martin for EMD and initial launch services RDT&
E: Other Transaction Launch Services: Firm Fixed Price (note: contract has
since changed)

Titan IV- Production: Lockheed Martin April 1996 Fixed- price incentive
fee

Total spent/ % total spent

$2.0 billion - EELV 9.7% $16.4 billion * Titan IV

90.9 % Source: Titan 12/ 31/ 2001 and EELV 12/ 31/ 2002 Selected
Acquisition Report (all dollar amounts in 2003 dollars)

Key Issues Affecting Programs

Schedule risk with transition to new launch vehicle Acquisition strategy
changed DOD oversight role Cost reductions uncertain

Note: Problems report affect EELV rather than Titan IV

Chronology of Key Findings

Titan IV

1991 GAO reported that slowing down Titan IV production may eventually
result in an overall increase in program costs, but that budgetary
requirements may be reduced by $47 million in FY1992 and $11 million in
FY1993. 1991 GAO reported that the Air Force planned to slowdown
production of the Titan IV launch vehicle to better synchronize production
and launch schedules. This restructuring of the program would result in
slowing down production from 8- 10 vehicles per year to not more than 6
vehicles per year beginning in 1992. The Titan IV has an optional upper
stage, the Inertial Upper Stage (IUS) and the newer Centaur, to provide
addition booster capacity for some satellite

payloads like the DSP. However, the DSP satellites to be boosted by the
IUS were not under contract and their launch was expected to be delayed.
In addition, planned production of the IUS vehicles for 1992 would likely
slip to 1995.

1991 GAO reported that numerous problems had delayed the transition of the
solid rocket motor upgrade program from development and testing to
production. For example, during the first static firing test of the rocket
motor upgrade the test motor exploded which would likely result in at
least a one- year delay in production from October 1991.

1993 DOD Bottom- Up Review noted that there are two types of requirements
for space launch: (1) performance* the ability to deliver a satellite
reliably to a specific orbit, and (2) operational flexibility. This review
reported that current launch systems generally met the first objective but

not the second. Performance and flexibility was inadequate because of (1)
the need to sustain three separate launch teams and associated equipment;
(2) the aging and obsolescence of major ELV components; and (3) continued
dependence on outdated launch vehicle production lines and manpower-
intensive launch processes. This report also found that there was
overcapacity in the

American space launch industry. As a result, the three manufacturers
operated at less than 50% capacity, which raised the unit cost of each
launch vehicle. The ability to sustain three launch suppliers over the
long term was in doubt. Foreign competition was also a factor. DOD
examined three options to address these issues: (1) extend the life of the
current launch vehicle fleet to the year 2030; (2) develop a new family of
expendable launch vehicles to replace the current fleet starting in 2004;
and (3) pursue a technology- focused effort to develop a reusable launch
vehicle. Option 1 was selected as the most cost- effective option in the
near- term while meeting DOD*s

requirements.

1994 DOD Space Launch Modernization Plan sought to develop roadmap options
establishing priorities, goals, and milestones for the modernization of U.
S. space launch capabilities. This report cited the growing sense within
Congress and others that while space launch is a critical issue for
America*s future in space, there is no coherent national plan to guide our
actions into the next century. The study developed 15 recommendations
concerning, among others, the

industrial base, investment, requirements, and coordination. The most
consistent theme of the study is that space launch is the key enabling
capability for the Nation to exploit and explore

GAO- 03- 825R Satellite Acquisition Programs Page 34 space.

1994 GAO reported that according to the April 1994 Moorman report, fewer
satellites, with longer lives, perform more work, which has resulted in
decreased launch rates and excess launch vehicle production and processing
capacity. The accompanying negative effect is low, inefficient

production rates that raise unit costs.

1994 GAO reported that DOD lacked an adequate and validated set of
requirements for a future launch system. While DOD desired to improve and
evolve the existing expendable launch vehicle fleet, it hadn*t established
an approach for acquiring and evaluating Russian launch vehicle components
and technologies to incorporate into future designs.

EELV

1997 GAO reported that cost risk was inherent in the vehicle acquisition
plan because production could be initiated from 1 to 2 years before the
first system development test flight. Such a strategy could result in
costly modifications to the production vehicles. Since there was
uncertainty in program cost the potential exists for program cost
increases. Cost dictated that there would not be any launches for
operational test and evaluation purposes.

1997 GAO reported that the program had schedule risk because DOD would
purchase the last of its existing expendable launch vehicles before the
first system development test flight was scheduled to occur. If the test
flight was unsuccessful, coupled with the expiration of existing
contracts, this could create a void in DOD*s launch capability. GAO had
reported on numerous occasions about the risks associated with program
concurrency and initiating production without

adequate testing.

1997 GAO reported that the Air Force had identified vehicle propulsion,
systems integration, and software as technical risk areas. Propulsion
systems were expected to require significant development. Integrating all
design, engineering, testing, manufacturing, and launch functions and the
software information system were expected to be challenging tasks. The
commercial application of the EELV posed a unique situation for the
government with the winning contractor

potentially enjoying an enhanced competitive edge (the demand for
commercial launches has not materialized and two contractors were awarded
EELV contracts) from DOD*s investment in the program. 1998 GAO reported
that the primary benefits associated with the EELV program should be

reduced cost to the government, but that DOD*s cost reduction estimate was
uncertain due to fluctuations in number, type and timing of launches.

1998 GAO reported that meeting launch site facility preparation schedules
as the primary program risk because construction had to begin shortly
after the milestone II decision in June 1998 to support the first EELV
launch in fiscal year 2002.

1998 GAO reported that DOD*s use of other transaction instruments, a
relatively new acquisition method, would challenge DOD in determining how
best to protect the government*s interests. Other transactions are
generally not subject to the federal laws and regulations governing
standard procurement contracts. Consequently, when using other transaction
(10 U. S. C. 2731) authority, contracting officials are not required to
include standard contract provisions that typically address such issues as
financial management or intellectual property rights, but rather

may structure the agreements as they consider appropriate. In addition,
the two contractors were not willing to guarantee system performance
because DOD*s financial risk was to be capped at $500 million per
contractor, while the contractor*s financial risk would be an open- ended
commitment. As a result, the contractors would not guarantee a launch
vehicle capability to

meet the government*s requirements (would only agree to provide a *best
effort*).

2001 DOD selected acquisition report commented on satellite weight growth
for the Wideband Gapfiller Satellite (WGS) and Advanced Extremely High
Frequency (AEHF) satellites. For example, the WGS spacecraft weight growth
had driven a need to upgrade from Medium to Intermediate for both Delta IV
and Atlas V launch vehicle configurations for the first three WGS

GAO- 03- 825R Satellite Acquisition Programs Page 35 missions. Spacecraft
weight growth on the AEHF satellite had also resulted in additional
funding

being added to the budget in order to upgrade to an Intermediate class
vehicle. GAO Reports GAO/ NSIAD- 91- 271, GAO/ NSIAD- 94- 253, GAO/ NSIAD-
97- 130, GAO/ NSIAD- 98- 151

GAO- 03- 825R Satellite Acquisition Programs Page 36 Appendix II Related
GAO Reports

Missile Warning and Tracking Missile Defense: Alternative Approaches to
Space Tracking and Surveillance System Need to be Considered. GAO- 03-
597. Washington, D. C.: May 23, 2003.

Defense Acquisitions: Space- Based Infrared System- low at Risk of Missing
Initial Deployment Date. GAO- 01- 6. Washington, D. C.: February 28, 2001.

National Missile Defense: Risk and Funding Implications for the Space-
Based Infrared Low Component. GAO/ NSIAD- 97- 16. Washington, D. C.:
February 25, 1997.

Defense Support Program: Ground Station Upgrades Not Based on Validated
Requirements. GAO/ NSIAD- 93- 148. Washington, D. C.: May 21, 1993.

Early Warning Satellites: Funding for Follow- on System Is Premature.

GAO/ NSIAD- 92- 39. Washington, D. C.: November 7, 1991. Communications
Military Satellite Communications: Concerns With Milstar's Support to
Strategic and Tactical Forces. GAO/ NSIAD- 99- 2. Washington, D. C.:
November 10, 1998. Defense Satellite Communications: Alternative to DOD's
Satellite Replacement Plan Would Be Less Costly. GAO/ NSIAD- 97- 159.
Washington, D. C.: July 16, 1997. Military Satellite Communications: DOD
Needs to Review Requirements and Strengthen Leasing Practices. GAO/ NSIAD-
94- 48. Washington, D. C.: February 24, 1994. Military Satellite
Communications: Opportunity to Save Billions of Dollars. GAO/ NSIAD- 93-
216. Washington, D. C.: July 9, 1993.

Military Satellite Communications: Milstar Program Issues and Cost- Saving
Opportunities. GAO/ NSIAD- 92- 121. Washington, D. C.: June 26, 1992.
Military Satellite Communications: Potential for Greater Use of Commercial

Satellite Capabilities. GAO/ T- NSIAD- 92- 39. Washington, D. C.: May 22,
1992. DOD Acquisition: Case Study of the MILSTAR Satellite Communications
System.

GAO/ NSIAD- 86- 45S- 15. Washington, D. C.: July 31, 1986. Navigation
Global Positioning System: Production Should Be Limited Until Receiver
Reliability Problems Are Resolved. GAO/ NSIAD- 91- 74. Washington, D. C.:
March 20, 1991.

GAO- 03- 825R Satellite Acquisition Programs Page 37

Satellite Acquisition: Global Positioning System Acquisition Changes After
Challenger's Accident. GAO/ NSIAD- 87- 209BR. Washington, D. C.: September
30, 1987. Issues Concerning the Department of Defense's Global Positioning
System as It

Enters Production. GAO/ MASAD- 83- 9. Washington, D. C.: January 26, 1983.
NAVSTAR Should Improve the Effectiveness of Military Missions-- Cost Has
Increased. GAO/ PSAD- 80- 21. Washington, D. C.: February 15, 1980.
Weather Polar- Orbiting Environmental Satellites: Status, Plans, and
Future Data

Management Challenges. GAO- 02- 684T. Washington, D. C.: July 24, 2002.

Meteorological Satellites. GAO/ NSIAD- 95- 87R. Washington, D. C.:
February 6, 1995.

Weather Satellites: Economies Available by Converging Government
Meteorological Satellites. GAO/ NSIAD- 87- 107. Washington, D. C.: April
23, 1987.

Launch Evolved Expendable Launch Vehicle: DOD Guidance Needed to Protect
Government's Interest. GAO/ NSIAD- 98- 151. Washington, D. C.: June 11,
1998.

Access to Space: Issues Associated With DOD's Evolved Expendable Launch
Vehicle Program. GAO/ NSIAD- 97- 130. Washington, D. C.: June 24, 1997.

Titan IV Launch Vehicle: Restructured Program Could Reduce Fiscal Year
1992 Funding Needs. GAO/ NSIAD- 91- 271. Washington, D. C.: September 6,
1991.

Reports Covering Multiple Space Programs and Management Issues Defense
Acquisitions: Assessments of Major Weapon Programs. GAO- 03- 476.
Washington, D. C.: May 15, 2003.

Defense Space Activities: Organizational Changes Initiated, but Further
Management Actions Needed. GAO- 03- 379. Washington, D. C.: April 18,
2003.

Military Space Operations: Planning, Funding, and Acquisition Challenges
Facing Efforts to Strengthen Space Control. GAO- 02- 738. Washington, D.
C.: September 23, 2002.

Defense Industry: Consolidation and Options for Preserving Competition.
GAO/ NSIAD- 98- 141. Washington, D. C.: April 1, 1998.

National Space Issues: Observations on Defense Space Programs and
Activities. GAO/ NSIAD- 94- 253. Washington, D. C.: August 16, 1994.

GAO- 03- 825R Satellite Acquisition Programs Page 38

Military Space Programs: Comprehensive Analysis Needed and Cost Savings
Available. GAO/ T- NSIAD- 94- 164. Washington, D. C.: April 14, 1994.

Military Space Programs: Opportunities to Reduce Missile Warning and
Communication Satellites' Costs. GAO/ T- NSIAD- 94- 108. Washington, D.
C.: February 2, 1994.

Military Space Programs: An Unclassified Overview of Defense Satellite
Programs and Launch Activities. GAO/ NSIAD- 90- 154FS. Washington, D. C.:
June 29, 1990.

(120236)

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.

The General Accounting 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 the Internet. GAO*s Web site (www. gao. gov) contains abstracts
and fulltext files of current reports and testimony and an expanding
archive of older products. The Web site features a search engine to help
you locate documents using key words and phrases. You can print these
documents in their entirety, including charts and other graphics.

Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as *Today*s Reports,* on its
Web site daily. The list contains links to the full- text document files.
To have GAO e- mail

this list to you every afternoon, go to www. gao. gov and select
*Subscribe to daily E- mail alert for newly released products* under the
GAO Reports heading.

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. General Accounting 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: Web site: www. gao. gov/ fraudnet/ fraudnet. htm E- mail:
fraudnet@ gao. gov Automated answering system: (800) 424- 5454 or (202)
512- 7470 Jeff Nelligan, Managing Director, NelliganJ@ gao. gov (202) 512-
4800

U. S. General Accounting Office, 441 G Street NW, Room 7149 Washington, D.
C. 20548 GAO*s Mission Obtaining Copies of

GAO Reports and Testimony

Order by Mail or Phone To Report Fraud, Waste, and Abuse in Federal
Programs Public Affairs
*** End of document. ***