Nuclear Waste: Absence of Key Management Reforms on Hanford's
Cleanup Project Adds to Challenges of Achieving Cost and Schedule
Goals (09-JUN-04, GAO-04-611).
The Department of Energy's (DOE) Hanford Site in Washington State
houses DOE's largest and most complex nuclear cleanup
project--treating and preparing for disposal 55 million gallons
of high-level radioactive waste. In 2000, DOE awarded an 11-year,
$4.3 billion contract to design, construct, and test treatment
facilities at Hanford. GAO was asked to review (1) efforts to
accelerate the project's completion, (2) implementation on this
project of agencywide management reforms, and (3) the challenges
resulting from any unimplemented reforms.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-04-611
ACCNO: A10430
TITLE: Nuclear Waste: Absence of Key Management Reforms on
Hanford's Cleanup Project Adds to Challenges of Achieving Cost
and Schedule Goals
DATE: 06/09/2004
SUBJECT: Construction contracts
Construction costs
Contract administration
Contract oversight
Contract performance
Cost analysis
Cost control
Facility construction
Nuclear waste disposal
Nuclear waste management
Nuclear waste storage
Performance measures
Radioactive waste disposal
Radioactive wastes
Federal procurement
Environmental cleanups
******************************************************************
** 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-04-611
United States General Accounting Office
GAO Report to the Committee on Government Reform, House of Representatives
June 2004
NUCLEAR WASTE
Absence of Key Management Reforms on Hanford's Cleanup Project Adds to
Challenges of Achieving Cost and Schedule Goals
a
GAO-04-611
Highlights of GAO-04-611, a report to the Committee on Government Reform,
House of Representatives
The Department of Energy's (DOE) Hanford Site in Washington State houses
DOE's largest and most complex nuclear cleanup project- treating and
preparing for disposal 55 million gallons of high-level radioactive waste.
In 2000, DOE awarded an 11-year, $4.3 billion contract to design,
construct, and test treatment facilities at Hanford. GAO was asked to
review (1) efforts to accelerate the project's completion, (2)
implementation on this project of agencywide management reforms, and (3)
the challenges resulting from any unimplemented reforms.
GAO recommends that DOE (1) follow more closely its project management
guidance when acquiring complex nuclear waste treatment plants, especially
by avoiding a fast-track, concurrent design-build approach, and (2)
develop and provide to Congress a plan that includes an estimate of the
costs and time frames needed to treat and dispose of DOE's high-level tank
wastes if most of these wastes must be disposed of in an underground
high-level waste repository. In commenting on the report, DOE generally
agreed with the recommendations, including improving its cost estimates,
but was unwilling to develop an alternative treatment plan for highlevel
waste until the legal issues are decided. GAO believes that any cost
estimate DOE develops should be based on a specific plan.
www.gao.gov/cgi-bin/getrpt?GAO-04-611.
To view the full product, including the scope and methodology, click on
the link above. For more information, contact Robin M. Nazzaro at (202)
512-3841 or [email protected].
June 2004
NUCLEAR WASTE
Absence of Key Management Reforms on Hanford's Cleanup Project Adds to
Challenges of Achieving Cost and Schedule Goals
DOE's initial approach called for treating 10 percent of the site's
high-level waste by 2018 and for operating the plant until treatment was
completed in 2046-well past a regulatory deadline to complete treatment by
2028. In 2002, DOE decided to accelerate cleanup by about 20 years and
reduce the project's $56 billion cost by $20 billion. In the short term,
however, several factors, including the accelerated approach and
contractor performance problems, have lengthened construction time and
raised contract costs by $1.4 billion to $5.7 billion.
Because of long-standing problems that preceded Hanford's contract, DOE
has instituted reforms in both contract and project management. DOE's 2000
Hanford contract implemented the contract performance reforms, including
linking contractor fees to cost and schedule performance. The contract did
not, however, implement project management reforms, such as an overall
plan to accomplish waste treatment by the regulatory deadline.
Not implementing project management reforms at the outset has added to the
risks in cleaning up Hanford's tank waste. First, to start quickly, DOE
committed to a "fast-track" process in which design, technology
development, and construction are performed concurrently on different
aspects of the project. For projects of Hanford's complexity, this
approach is not compatible with controlling costs and schedules. Second,
DOE has delayed completing analyses needed to determine the most
cost-effective approach to waste separation and may have missed savings
opportunities of at least $50 million a year. Third, DOE has not
adequately defined or communicated the potential effects of a legal
challenge to its overall plan for minimizing how much high-level waste is
disposed of in an underground repository. Unless effectively managed, an
adverse legal outcome could increase project costs by tens of billions of
dollars.
High-level Vitrification Plant at Hanford's Waste Treatment Construction
Site, March 2004
Contents
Letter
Results in Brief
Background
DOE Has Revised Its Cleanup Approach to Reduce Costs and Save
Time in the Long Term but Has Increased the Cost of Its Major Construction
Project in the Short Term Contract Performance Reforms Were in Place at
Start of Project, Project Management Reforms Were Added Later Not
Implementing Key Project Management Reforms at Start of the
Construction Project Has Added to DOE's Challenges Conclusions
Recommendations for Executive Action Agency Comments and Our Evaluation
1 3 7
12
19
26 37 38 38
Appendixes
Appendix I: Appendix II: Appendix III: Appendix IV:
Scope and Methodology
Analysis of DOE's Cost-Saving Estimates
Comments from the Department of Energy
GAO Contacts and Staff Acknowledgments
GAO Contacts
Staff Acknowledgments
41
44
47
49 49 49
Tables Table 1: Key Events in the Hanford Waste Treatment Project 10
Table 2: Comparison between DOE's Previous Approach
(December 2000) and Accelerated Approach (August 2002)
for Processing Hanford's Tank Waste 16
Table 3: GAO Analysis of DOE's Cost-Savings Estimate for the
Hanford Tank Waste Treatment Project 46
Figures Figure 1: Aerial View of the Waste Treatment Plant, March 2004 14
Figure 2: Key Construction Project Dates: DOE's Previous
Approach Compared with Its Accelerated Approach 18
Contents
Abbreviations
AEA Atomic Energy Act of 1954
AEC Atomic Energy Commission
CHG CH2M Hill Hanford Group
DOE Department of Energy
EPA Environmental Protection Agency
ERDA Energy Research and Development Administration
NRC Nuclear Regulatory Commission
OMB Office of Management and Budget
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.
A
United States General Accounting Office Washington, D.C. 20548
June 9, 2004
The Honorable Thomas M. Davis Chairman The Honorable Henry A. Waxman
Ranking Minority Member Committee on Government Reform House of
Representatives
The Department of Energy (DOE) spends almost $20 billion a year-about 90
percent of its fiscal year 2003 budget-on contracts, including those for
operating its installations. These operations include extensive projects
for treating and storing radioactive and hazardous wastes accumulated over
more than 50 years of nuclear weapons production-a task DOE has estimated
will cost more than $142 billion (fiscal year 2003 dollars) and will not
be completed until 2035. DOE's success has been mixed, with some cleanup
projects going well but others producing few or no results after the
expenditure of hundreds of millions of dollars. We and others have
criticized DOE for failing to hold its contractors accountable for results
and for inadequate management and oversight of its projects. In response,
starting in the mid-1990s, DOE undertook two main reform initiatives- both
were designed to improve contractor performance in carrying out major
projects. The contract performance initiative, which DOE formalized in
1996, included ensuring that DOE begins projects by using the appropriate
contract type, encouraging greater competition in contract bids, and
linking contract award fees more closely to contractor performance. The
project management initiative, started in 1999 and resulting in new
management guidance in 2003, required DOE and its contractors to implement
certain management practices, such as following a more disciplined
decision-making process for defining, planning, and carrying out major
projects.
You asked us to review the impact of these management reform efforts at
DOE's largest and most complex cleanup project-the waste treatment
project1 at DOE's Hanford Site in Washington State. Hanford's waste
treatment project involves more than 55 million gallons of radioactive and
hazardous waste-enough to fill an area the size of a football field to a
depth of more than 150 feet-stored in 177 underground tanks. DOE has
managed this waste over the years as high-level waste.2 The overall
project involves two major steps: (1) designing, constructing, and testing
a waste treatment plant3 (the construction project) and (2) operating this
plant and others in subsequent years to process and prepare the tank waste
for permanent disposal (the operations project). In 2000, after DOE had
announced its agencywide contract and project management reforms, it
awarded an 11-year, $4.3 billion contract for the construction project. In
April 2003-as part of a nationwide effort that DOE started in 2002 to
reduce the costs and schedule for cleaning up many of its sites-DOE
revised Hanford's waste treatment project, accelerating it to help meet a
2028 deadline that the department had agreed to with Washington State in
1989. Our analysis examines the interaction of these four separate
efforts- the contract performance initiative, the project management
initiative, DOE efforts to accelerate cleanup at its sites, and the waste
treatment project at Hanford. Specifically, this report examines (1) the
DOE accelerated approach for addressing Hanford's tank waste, including
any changes in the construction project's cost or schedule; (2) the extent
to which DOE's contract and project management reforms have been
implemented on the construction project; and (3) the challenges resulting
from any unimplemented reforms on the potential for completing the waste
treatment project successfully.
This report is based on detailed work we conducted at DOE's Hanford Site
near Richland, Washington, and on our analysis of cost and schedule
information and legal documents pertaining to the waste treatment project.
We also reviewed DOE's contract reform and project management initiatives,
including our prior reports on those initiatives. To evaluate the
1This project is officially known as the River Protection Project. The
Columbia River flows through the site, and the cleanup is designed in part
to keep contamination from reaching the river. In this report, we refer to
the project as the "waste treatment project."
2For this report, we use the term "high-level waste" to refer to the waste
that DOE is managing as high-level waste. DOE's Hanford Site is one of
three DOE sites with high-level wastes needing treatment; the other two
are the Savannah River Site in South Carolina and the Idaho National
Laboratory.
3The waste treatment plant consists of one facility to separate the waste,
two facilities to treat separated portions of the waste, and one
laboratory and other supporting facilities.
technical aspects of the Hanford waste treatment project, we obtained the
assistance of a physicist with extensive experience in the nuclear field.
A detailed discussion of our scope and methodology appears in appendix I.
We conducted our review from July 2003 through May 2004 in accordance with
generally accepted government auditing standards.
Results in Brief DOE's accelerated waste treatment approach calls for
significant changes in treating and disposing of Hanford's tank wastes as
a way of saving both time and money when the waste treatment facilities
begin operation. Under the approach of its December 2000, $4.3 billion
contract with Bechtel National, Inc. (Bechtel National), DOE estimated
that cleaning up the waste would cost about $56 billion and take until
about 2046. The accelerated approach calls for completing waste treatment
by 2028; by reducing cleanup by almost 20 years, DOE estimates it will
save about $20 billion. The accelerated approach came about 2 years after
the construction contractor, operating under the assumptions of DOE's
previous approach, had begun to design, build, and test the waste
treatment plant. DOE increased the contract amount for the construction
project by $1.4 billion, to $5.7 billion, to reflect the accelerated
approach's added production capacity, contractor performance problems,
additional design work, and better estimates. DOE also lengthened by 16
months the time for designing and building the plant. To keep the
construction project on schedule despite the increase, however, DOE
decided to shorten the commissioning phase-the period for testing the
plant to ensure that it will work properly once operations begin-by nearly
the same amount. As a result, DOE has maintained its schedule to begin
operation of the waste treatment plant by 2011.
DOE's 2000 contract implemented the department's contract performance
initiative but not its project management initiative. Elements of the
contract initiative in place at the contract's start included ensuring
that a competitive process for bidding the 2000 contract was followed and
paying contractor performance fees on the basis of the cost and time
required to successfully complete the plant. Because it was under pressure
to meet regulatory milestones and keep the project moving forward,
however, DOE awarded the contract without fully implementing its project
management initiative, which the department had developed but not yet
issued in final form. For example, the initiative calls for developing, at
the beginning of a project, an approach that can meet mission
requirements. The initial contract DOE awarded, however, called for
designing and building a plant that would not be able to complete the
cleanup by 2028, the time frame
DOE agreed to with Washington State. Similarly, the project management
initiative calls for a project's cost and schedule to be validated before
setting a firm project price. At the time of contract award, however,
DOE's cost and schedule estimates had not been fully validated to
determine if they were realistic. Since 2002, DOE has implemented various
steps called for in the project management initiative, such as developing
a more realistic baseline and increasing project contingency.
DOE's decision not to implement the project management initiative at the
start of the construction project has added to the risk and uncertainty
the department faces in completing both the construction and operations
projects. The initiative's main purpose is to minimize problems that have
plagued DOE in the past by taking such steps as avoiding or mitigating
high-risk strategies for constructing facilities, as well as by conducting
rigorous analyses to support key decisions. In awarding the 2000 contract
without these steps being fully implemented, DOE has adopted a high-risk
strategy at Hanford. According to Hanford project officials, implementing
project management steps subsequently, as DOE has begun to do under its
accelerated approach, is intended to provide a more disciplined method for
keeping the project's cost and schedule on track. In our view, however,
these steps cannot eliminate the risk and uncertainty resulting from the
department's decision not to implement the project management initiative
from the start. This risk and uncertainty comes from three main sources:
o Using a fast-track management approach. DOE has committed to a
"fast-track" process in which many design, technology development, and
construction activities are performed simultaneously, thereby
significantly increasing the risk of cost increases and schedule delays as
the construction project progresses. Performance so far on the
construction project has been mixed-problems have already contributed to
the $1.4 billion construction cost increase. Furthermore, our review
indicates that cost, schedule, and performance problems continue, although
they have not yet affected the revised project baseline. For example,
construction activities, such as building interior walls, have outpaced
design, leading to delays in finishing the walls and the need to
temporarily reassign construction workers to other tasks. In addition,
efforts to resolve key technical challenges, including incorporating
alternative treatment technologies, continue to fall behind and threaten
to affect the overall project's baseline. Hanford's construction project
manager acknowledged that the contract schedule for building and testing
the plant and concurrently resolving technical issues was and still is
high risk. Although he and other DOE officials
stated they have taken adequate steps to mitigate this risk, DOE, since
contract award, has continued to encounter significant technical and other
problems. Given the risk inherent in this project, we and outside experts
believe that further cost increases, as well as schedule delays, are
likely.
o Not fully evaluating cost-saving possibilities beyond the construction
phase. DOE started and is moving forward on the waste treatment project
without fully completing or conducting some of the rigorous analyses
needed to determine the most desirable approach for separating waste
components before further treatment. For example, as specified in its
construction contract, DOE plans to use a material called "resin" to
separate high-level components from the waste. The resin DOE selected in
2000 is available only from a single supplier, and DOE officials have been
slow to study alternative resins that could reduce operating costs of the
separation facility. Instead, DOE has continued to depend on the
production capability and pricing constraints of the single supplier, even
though other-potentially less expensive- alternative resins exist.
Finally, in late February 2004, DOE decided to more fully assess the cost
and risks of relying on a single supplier for the resin. But because of
the time required to test and certify an alternative resin, the new
material may not be available in time for commissioning and beginning
operation of the separation facility. As a result, DOE's project manager
estimated that this lost opportunity could increase project operation
costs by at least $50 million a year. DOE's slowness in pursuing an
alternative resin stemmed from its focus on achieving the near-term goal
of having an operating plant by 2011 and its belief that pursuing the
alternative, cost-saving option could jeopardize achieving that goal.
o Inadequate planning to assess and mitigate the effects of a legal
challenge to DOE's overall approach to treating and disposing of highlevel
radioactive waste. DOE's strategy for treating the waste-not only at
Hanford, but also at its other high-level waste sites at Savannah River
and the Idaho National Laboratory-is predicated on a key legal assumption
that has been successfully challenged in court. The treatment strategy
rests heavily on DOE's ability to determine that a majority of its tank
waste can be classified as other than high-level waste and treated with
less expensive technologies. DOE recently lost a court decision on this
matter and is appealing the ruling. DOE has also pursued, unsuccessfully
as of May 2004, a legislative remedy that has instead raised concerns
about whether the department is attempting to
avoid treating some of its waste in an appropriate manner. If DOE has to
change its current approach to treating and disposing of its waste, the
change would have a major impact on cleanup of Hanford's tank wastes as
well as those at DOE's other high-level waste sites. DOE has developed
only rough estimates of potential cost and schedule impacts that this
legal challenge poses, including near-term delays and overall effects on
cleanup. These estimates for the Hanford Site range from about $350
million for delays if the lawsuit is resolved in DOE's favor, to possibly
more than $19 billion in delays and changes to the program if the decision
against DOE is upheld. The Hanford estimate is part of a complexwide
estimate of more than $100 billion if the lawsuit is not resolved in DOE's
favor. Our review indicates that these estimates may be significantly
understated. During our review, DOE officials said they believe that their
cost estimates and risk-mitigation plans are adequate, and that no further
analysis is needed because DOE will ultimately succeed in the appeal or
its legislative efforts. In our view, however, a more thorough analysis
and full disclosure are needed concerning the potential risk this legal
issue poses to the waste treatment project at Hanford and DOE's other
sites, including potential impacts on the project's cost and schedule and
the environmental risks associated with further delays. We believe full
disclosure is important so that policy makers and others can undertake a
more informed debate about DOE's high-level waste program.
We are recommending that DOE (1) more closely follow its project
management order and implementing guidance when developing and carrying
out complex nuclear waste cleanup projects at Hanford and other DOE sites
and (2) develop and disclose to Congress a full and complete estimate of
the costs and time frames required to dispose of Hanford's and the rest of
DOE's high-level tank wastes if, to comply with the law, DOE must dispose
of a majority of its tank wastes in a high-level waste repository.
DOE agreed to follow more closely its project management order and
guidance and to develop more complete information on the costs if DOE is
required to dispose of a majority of its tank wastes in a high-level waste
repository. However, DOE said it was unwilling to develop an alternative
treatment and disposal plan until the outcome of the legal appeal has been
determined. We believe that to be meaningful, any cost and schedule
estimates DOE develops should be based on a specific alternative treatment
plan.
Background DOE's Hanford Site in southeastern Washington State was
established in 1943 to produce nuclear materials for the nation's defense.
This production mission entailed dissolving used ("spent") nuclear fuel to
remove plutonium, uranium, and other materials and generated large volumes
of high-level radioactive wastes. The wastes' radioactive components decay
in a process in which atoms of a radioactive element (also known as a
"radionuclide") spontaneously release dangerous radiation. Even short
periods of exposure to intense radiation can cause health problems,
including radiation sickness, burns, and even death. Because of the
intense radiation emitted from high-level waste, the waste must be
isolated and handled remotely behind heavy shielding. Some of the
radioactive components can be very mobile in the environment and, if not
checked, may migrate quickly to contaminate soils and groundwater. In
addition to radioactive components, DOE's high-level waste generally also
contains hazardous components-including solvents, caustic soda, and heavy
metals such as chromium. Radioactive waste components combined with
hazardous components are referred to as "mixed wastes."
Although DOE stopped producing nuclear material at Hanford in 1989,
high-level waste tanks on the site now contain more than 55 million
gallons of sludge, liquid, and a sandlike material called "saltcake." This
Hanford tank waste represents about 35 percent of the radioactivity and
almost 60 percent of the high-level waste inventory (by volume) within
DOE. Hanford's 177 underground waste tanks were constructed between 1943
and 1986. The oldest 149 tanks have single-layer walls, or shells, and DOE
has reported that 67 of these tanks are assumed to have leaked waste into
the soil. All of the single-shell tanks are beyond their design life. The
28 newest tanks have double-shells and are still in use, and DOE reports
that these tanks have not leaked. Since 1989, DOE has spent about $7
billion to manage the waste and explore possible ways to treat and dispose
of the wastes. As of May 2004, none of the high-level waste at Hanford has
been treated for final disposal.
Treatment and disposal of high-level waste produced at DOE facilities,
including Hanford, are governed by a number of federal laws that define
the role of DOE and the Nuclear Regulatory Commission (NRC) in waste
management. The Atomic Energy Act of 1954 (AEA) and the Energy
Reorganization Act of 1974 established responsibility for the regulatory
control of radioactive materials, including DOE's high-level wastes.4 The
Energy Reorganization Act of 1974 assigned NRC the function of licensing
facilities that are authorized for long-term storage of high-level
radioactive waste generated by DOE and others.5 The Nuclear Waste Policy
Act of 1982, as amended, defines high-level radioactive waste as "highly
radioactive material resulting from the reprocessing of spent nuclear
fuel, including liquid waste produced directly in reprocessing, and any
solid material derived from such liquid waste that contains fission
products in sufficient concentrations, and...other highly radioactive
material that the [NRC]...determines...requires permanent isolation."6 The
act also established a process for developing and siting a geologic
repository (a permanent site deep underground) for the disposal of
high-level waste and spent fuel.
DOE's past efforts to treat and dispose of Hanford's tank wastes have been
costly but have resulted in little progress. DOE's initial cleanup
strategy, developed in 1989, was to immobilize the wastes stored in the 28
most modern tanks because the department knew the most about the waste
constituents in those tanks. DOE planned to defer until at least 2003 in
deciding what to do with the waste in the remaining 149 tanks. The
department spent about $23 million on renovating a World War II-era
facility, in which it planned to conduct initial waste processing, before
determining that the facility could not be upgraded to meet environmental
standards. DOE abandoned this project in 1991. DOE continued designing a
waste treatment facility, known as the Hanford Waste Vitrification Plant,7
that, beginning in late 1991, was expected to process the waste from all
177 tanks. The resulting design, however, was too small to treat all of
the waste in a time frame acceptable to regulators. In addition, under
this scenario,
4The AEA authorized the Atomic Energy Commission (AEC) to provide for the
safe storage of radioactive waste from defense-related activities. 42
U.S.C. S: 2121(a)(3). Later, the Energy Reorganization Act of 1974
abolished the AEC, transferring responsibilities to the Energy Research
and Development Administration (ERDA)-DOE's predecessor-and NRC. 42 U.S.C.
S:S: 5814, 5841. In 1977, ERDA was abolished, and its functions were
transferred to the newly established DOE, explicitly leaving the
management of the government's radioactive waste in the hands of DOE. 42
U.S.C. S:S: 7151(a), 7133(a)(8).
542 U.S.C. S: 5842.
642 U.S.C. S: 10101(12).
7Vitrification is a thermal process of mixing the waste with glass-forming
materials and melting it into glass. The glass is then poured into
canisters for long-term storage or disposal.
DOE would have completed the treatment facility before other aspects of
the waste treatment program were fully developed-such as retrieving the
wastes from the tanks. DOE abandoned this plan in 1993, after spending
$418 million. DOE's next cleanup attempt, begun in 1995, involved shifting
more of the project and performance risk from DOE itself to its contractor
through an approach called "privatization." Under a fixed-price contract
using this approach, the contractor would design, finance, build,
commission, and operate waste treatment facilities on the Hanford Site for
DOE. The department would pay the contractor for successfully processed
waste placed in canisters. The first phase of this project was to involve
treating about 10 percent of the waste, at a contract price of $3.2
billion. Between 1996 and 2000, however, the proposed contract price
soared to more than $14 billion. In June 2000, DOE canceled the contract,
after spending about $300 million, mostly on plant design. In December
2000, DOE awarded a new $4.3 billion "cost-reimbursable" contract with
performance fee to complete the waste treatment plant that the previous
contractor had begun to design.8 In August 2002, DOE revised the project
to more effectively meet regulatory milestones. DOE renegotiated the
contract in April 2003 to reflect this revision and to address
construction problems. In this report, we refer to DOE's December 2000
plan as the "previous approach" and the August 2002 strategy as the
"accelerated approach." (Table 1 summarizes key project events discussed
in this report.)
8Cost-reimbursement contracts provide for payment of allowable incurred
costs as prescribed in the contract. These contracts establish an estimate
of total cost and set a ceiling that the contractor may not exceed, except
at its own risk, without the approval of the contracting officer.
Table 1: Key Events in the Hanford Waste Treatment Project
Date Activity
December 2000 Award of plant construction contract to Bechtel National
(previous approach)
July and September 2002 Independent reviews of the construction contract
cost and schedule
April 2003 Bechtel National contract modification approved (accelerated
approach)
May 2005 Alternative treatment demonstration to begin
July 2011 Bechtel National contract planned completion date
February 2018 Regulatory milestone to complete treatment of 10 percent of
the waste by volume (25 percent of the radioactivity)
December 2028 Regulatory milestone to complete treatment of Hanford's tank
waste
2046 Previous DOE baseline date for completing treatment of Hanford's tank
waste
Source: Compiled by GAO from DOE data.
Problems at Hanford and other DOE sites have led DOE to institute reforms
in contract and project management. DOE relies almost entirely on
contractors to carry out its production, research, and cleanup missions.
The department's history of inadequate management and oversight and of
failure to hold its contractors accountable for results led us in the
early 1990s to designate DOE contract and project management as a
high-risk area vulnerable to fraud, waste, abuse, and mismanagement. In
response to these and other criticisms, DOE took steps to reform its
contracting and project management practices. On the contracting side, in
February 1994, DOE issued a report-Making Contracting Work Better and Cost
Less- containing 48 recommendations in three key areas: selecting
alternatives to traditional contracting arrangements used for management
and operation of its sites, increasing competition to improve performance,
and developing and using performance-based contracting tools. On the
project management side, after a series of critical reviews by the
National Academy of Sciences and others, DOE issued order 413.3 in October
2000, which defined requirements for DOE project management. These
requirements include following a formal planning and decision process;
developing key management tools, including an acquisition strategy;9 and
implementing effective management practices, such as minimizing project
risk by developing mitigation strategies. DOE has implemented order 413.3
through detailed guidance it adopted in March 2003.10
DOE carries out its tank waste cleanup program under the leadership of
DOE's Assistant Secretary for Environmental Management and in consultation
with a variety of regional and local stakeholders. In addition to the
Environmental Protection Agency (EPA) and state environmental agencies,
which have regulatory authority in the states where the sites are located,
stakeholders include county and local governmental agencies, citizen
groups, advisory groups, and Native American tribes. These stakeholders
make known their views through various public involvement processes,
including site-specific advisory boards. Over the years, much of the
cleanup activity has been implemented under compliance agreements between
DOE and regulatory agencies. These compliance agreements provide for
establishing legally enforceable schedule milestones governing the work to
be done. The operation of Hanford's tank waste program is regulated under
the Hanford Federal Facility Agreement and Consent Order between DOE and
Washington State's Department of Ecology and EPA. This agreement, commonly
called the "Tri-Party Agreement," was signed in May 1989.11 At Hanford,
DOE manages the project through its Office of River Protection, which is a
congressionally established organization created in December 1998 to
manage tank waste issues. The office has a staff of about 110 DOE
employees and a fiscal year 2004 budget of about $1.1 billion. It manages
Hanford's tank waste cleanup through two main contracts: a construction
contract with Bechtel National for a tank waste treatment plant and a tank
farm operations contract with CH2M Hill Hanford Group (CHG). Through its
contract, CHG manages various activities in support of the waste treatment
project, including planning for
9An acquisition strategy establishes a framework within which detailed
project planning and execution are accomplished. An acquisition strategy
defines an acceptable approach to meeting mission requirements and the
relationships between essential project elements, such as project
management, worker safety, and contract administration.
10In this report, we refer to DOE's order 413.3 and implementing guidance
as its "project management initiative."
11The purpose of the Tri-Party Agreement is to ensure that environmental
impacts associated with past activities are addressed and that
environmental laws are complied with. The agreement covers many other site
activities in addition to the tank wastes. It also outlines a process for
modifying the agreement if needed.
the overall project, developing alternative treatment technologies,
storing the tank waste until treatments are available, and preparing to
retrieve the waste from the tanks for treatment.
DOE Has Revised Its Cleanup Approach to Reduce Costs and Save Time in the
Long Term but Has Increased the Cost of Its Major Construction Project in
the Short Term
By accelerating and otherwise adjusting its previous approach to the tank
waste project-which was unlikely to meet required deadlines-DOE estimates
that it can shorten the project's overall time frame by almost 20 years
and lower the cost from more than $50 billion to less than $30 billion. To
achieve these goals, DOE has expanded the capacity of the reconfigured
treatment facilities and made other changes to the construction phase of
the project. These changes, plus contractor performance problems, have
increased the $4.3 billion construction contract signed in 2000 by 33
percent, or $1.4 billion, bringing the total contract price for the
construction project to $5.7 billion. DOE still expects to begin operating
the waste treatment plant by 2011.
DOE's Previous Cleanup Approach Unlikely to Meet the Required Deadlines
DOE's December 2000 previous approach to the tank waste project did not
define project activities beyond 2018. The department's plan consisted of
(1) a first phase during which facilities would be constructed and about
10 percent of the waste would be processed by 2018 and (2) a second phase
during which treatment capacity would be added and the remaining waste
would be treated. The plan involved constructing three main treatment
facilities-a waste separation facility, a high-level waste vitrification
facility, and a low-activity waste vitrification facility-as well as
various support facilities. Once the plant was commissioned, DOE intended
to separate 10 percent of the waste by volume (and about 25 percent of the
total radioactivity) into high-level and low-activity portions of the
waste12 and then vitrify the separated wastes in two treatment facilities,
one for high-level waste and the other for low-activity waste. DOE plans
to eventually dispose of the high-level waste in a geologic repository.
Although DOE had planned to stabilize the low-activity waste through
vitrification as well, it plans to dispose of this waste at a facility to
be constructed on the Hanford Site. To accomplish treating 10 percent of
the
12There is no statutory or regulatory definition of low-activity waste. At
Hanford, DOE defines it as solidified waste that qualifies as mixed
low-level waste because it is treated to remove radionuclides to below 10
C.F.R. Part 61 Class C concentrations and has been shown to meet
performance objectives equivalent to those in 10 C.F.R. Part 61 Subpart C.
In this report, we refer to this portion of the waste as "low-activity
waste."
waste by 2018, DOE designed treatment facilities with the capacity of 1.5
metric tons per day of high-level waste glass and 30 metric tons per day
of low-activity waste glass. Although DOE reported that this treatment
capacity was sufficient for the project's first phase, it was not
sufficient for treating the remaining 90 percent of the waste. Therefore,
near the end of the first phase of the operations project in 2018, DOE
intended either to expand the existing treatment facilities' capacity or
to adopt another approach. DOE had agreed with its regulators-the state of
Washington and EPA-to more fully define the project's second phase by
2005.13
Under the previous approach, for the second phase of the operations
project, DOE planned to expand the capacity of the treatment facilities in
2018 to 6 metric tons per day of high-level waste glass and 60 metric tons
per day of low-activity waste glass and to extend project activities until
2046-18 years beyond the date agreed upon with regulators. This decision
came from a DOE assessment of the requirements for the project's second
phase, which led the department to conclude that it could not finish waste
treatment by 2028. To meet its regulatory commitment to complete treating
all of the waste by 2028, DOE recognized that it would need considerable
capacity beyond what could be added to the first-phase treatment
facilities. However, DOE also determined that it was not feasible to
obtain the several billions of additional dollars needed to construct
waste treatment facilities beyond what was already under construction. DOE
officials concluded at the time that the proposed expansion was the best
the department could do, and that it would need to renegotiate the current
2028 regulatory deadline for completing waste treatment with EPA and the
Washington State Department of Ecology.
Accelerated Approach Defined a Complete Project That Would Meet Regulatory
Deadlines and Reduce Costs
In August 2002, DOE defined an accelerated approach to address the
processing of Hanford's tank wastes. According to the department, this new
approach would reduce environmental risk more quickly, save billions of
dollars over the previous approach, and allow DOE to meet its regulatory
commitment to complete waste processing by 2028. This accelerated approach
grew out of a DOE-wide effort to reexamine cleanup and to find ways to
accelerate risk reduction and reduce overall cleanup
13In the Tri-Party Agreement, DOE has agreed to retrieve and treat at
least 99 percent of the waste altogether. If the department succeeds,
about 500,000 gallons of waste, known as the "tank heel," will remain when
the tanks are closed. Tank heel waste and DOE's method of tank closure are
outside the scope of our review.
costs. In addition, DOE officials saw the accelerated initiative as a
possible vehicle for completing tank waste treatment by 2028 instead of
2046.
DOE's accelerated approach at Hanford differs substantially from the
department's previous approach and has three main elements:
o Increasing the capacity of treatment facilities now under construction.
Under the accelerated approach, the waste treatment plant was modified to
increase waste treatment capacity. Instead of one "melter" able to produce
1.5 metric tons of waste glass per day in the high-level waste
vitrification facility and three melters able to produce a total of 30
metric tons of waste glass per day in the low-activity waste vitrification
facility, DOE decided to incorporate two higher capacity melters for
treating high-level waste and two higher capacity melters for treating
low-activity waste. DOE expects the expanded facilities to have the
capacity for 6 metric tons of high-level waste glass and 36 metric tons of
low-activity waste glass per day. Figure 1 shows the plant under
construction.
Figure 1: Aerial View of the Waste Treatment Plant, March 2004
o Finding an alternative treatment approach for about 60 percent of the
low-activity waste. Under the accelerated approach, DOE has researched a
number of alternative ways to treat a large portion of the low-activity
waste at facilities other than the low-activity waste plant now under
construction. The main alternative DOE is currently investigating involves
vitrifying about 60 percent of the low-activity waste directly in the
disposal container-a process called "bulk vitrification." In 2004, DOE
plans to begin constructing a pilot facility to further develop and
demonstrate the bulk vitrification process. If the 2year test is
successful, DOE plans to construct a bulk vitrification facility near the
other waste treatment facilities now under construction. In addition, DOE
continues to support development of another potential treatment technology
for this waste, a thermal treatment process called "steam reforming."
o Disposing of waste whose characteristics are consistent with
transuranic waste without vitrifying it.14 As another alternative for
treating part of the waste, DOE has identified 2 to 3 million gallons of
tank waste that it manages as high-level waste but believes the waste does
not require treatment and disposal as high-level waste. DOE has reported
that its analysis of tank records shows that waste in those tanks does not
come from reprocessing of spent fuel and, therefore, does not meet the
legal definition of high-level waste. DOE plans to manage this waste as
transuranic waste and, after drying and packaging the waste, ship it to
the Waste Isolation Pilot Plant in New Mexico for disposal. DOE estimates
that this approach, if approved by regulators, could shorten the operating
project time frame by at least 3 years.
DOE expects that its accelerated approach will enable it to avoid building
another large waste treatment plant and still meet the regulatory deadline
of completing waste processing by 2028.15 DOE also expects that
accelerating treatment to a completion date of 2028 will reduce overall
project costs from about $56 billion to about $27 billion. The difference
14Transuranic waste is defined as waste containing radionuclides with
atomic numbers higher than 92 (the atomic number of uranium) and
half-lives longer than 20 years in concentrations exceeding 100 nanocuries
per gram. 42 U.S.C. S: 2014(e)(e), 40 C.F.R. S: 191.02(i).
15The Tri-Party Agreement requires DOE to report by January 2005 on the
status of its revised approach and submit by January 2006 an updated plan
for completing waste processing.
between the two DOE estimates is $29 billion, but DOE has stated publicly
that it expects to achieve savings of $20 billion by accelerating the
project. However, we believe that DOE's cost-savings estimate is
significantly overstated. (See app. II for a discussion of our concerns
about DOE's costsavings estimate.) Table 2 summarizes the overall
differences between DOE's previous approach and its accelerated approach.
Table 2: Comparison between DOE's Previous Approach (December 2000) and
Accelerated Approach (August 2002) for Processing Hanford's Tank Waste
Previous approach
Capacity from Capacity after Accelerated Element of approach 2011 to 2018
2018 approach
Daily treatment
capacity
of treatment
facilities:
High-level waste 1.5 metric 6 metric 6 metric
glass tons/day tons/day tons/day
30 metric 60 metric 36 metric
Low-activity waste tons/day tons/day tons/day
glass
Role of alternative None None Treat 60 percent of
treatment approaches, low-activity waste
such as bulk vitrification
Overall treatment N/A 2046 2028
completed
Life-cycle cost (2003 to N/A $56 billiona $27 billiona
completion in nominal
dollars)
Source: Compiled by GAO from DOE data.
aThe difference between the two DOE estimates is $29 billion, but DOE has
stated publicly that it expects to achieve a savings of $20 billion by
accelerating the project. Neither savings figure represents savings DOE
can expect to realize, however, because its cost estimates have not been
properly developed. To address these problems, we used a present-value
analysis, in which dollars are discounted to a common year to reflect the
time value of money. See appendix II for a comparison of DOE's estimated
savings with our calculation.
Accelerated Approach and Other Project and Contractor Performance Issues
Have Lengthened the Design and Construction Phases and Increased Contract
Price by $1.4 Billion
DOE renegotiated its contract with Bechtel National to address several
factors, including accelerating the overall project to adjust for the
project's changed scope, revising the project schedule to increase
construction time, resolving contractor claims, and overcoming contractor
performance problems that had occurred to that point. The project's
changed scope required design rework and other equipment and facility
changes to add capacity in the treatment facilities and to separate a
planned analytical laboratory from the waste separation facility. For
example, in the high-level waste treatment facility, the contractor had to
increase individual floor heights, resulting in the building height
increasing 7 feet, to ease the complexity of construction and for added
operational flexibility. DOE has estimated that these changes in scope
will cost about $250 million.
DOE and Bechtel National have also agreed that about $325 million in
increased cost stems from Bechtel's engineering and project management
problems as well as from delays in completing design work. DOE and Bechtel
National have not agreed on the cause of the remaining $850 million in
cost increases, which resulted from estimation errors and omissions and
from various design and construction issues. However, DOE and Bechtel
National resolved the disputed $850 million by changing the contract and
modifying the performance incentives to hold the contractor more
responsible for future cost increases. As a result of the contract
renegotiation, the contract price has risen by $1.4 billion (33 percent),
to $5.7 billion. In addition, the dates for certain interim steps of the
construction project have been modified, although the completion date for
the contract remains the same so that DOE can maintain its schedule to
begin operations of the waste treatment plant by 2011 (see fig. 2)
Figure 2: Key Construction Project Dates: DOE's Previous Approach Compared
with Its Accelerated Approach Previous approach
7 62 5 10 43
12/2000 7/2001 9/2006 2/2007 12/2007 7/2011
Accelerated approach 12/2000 7/2002 1/2008 8/2008 12/2009 7/2011
Actual
Missed milestone
Number of months Source: GAO analysis of DOE data.
aCold commissioning refers to testing facilities using simulated tank
waste. bHot commissioning refers to testing facilities using actual tank
waste.
Contract Performance Reforms Were in Place at Start of Project, Project
Management Reforms Were Added Later
When the contract was originally awarded in 2000, DOE implemented key
elements of its contract performance initiative but chose not to implement
its project management reforms. At the time, DOE had just adopted the
project management reforms resulting from its project management
initiative, but implementing guidance was not in place. As the waste
treatment project got under way and performance problems began to surface,
DOE began to put in place many of the project management initiative's key
elements.
Key Contract Performance Reforms Were in Place When Contract Was Awarded
At the time of the December 2000 contract award, DOE was implementing key
elements of its contract performance initiative throughout the DOE complex
to help improve contractor performance and achieve project cost and
schedule goals. These elements included ensuring that the type of contract
was appropriate for a project's characteristics and risk level, competing
contracts to ensure that the government receives the best offer, and using
performance-based incentives to reward contractors for good performance
and penalize them for poor performance. DOE has worked toward implementing
this contract initiative, which includes these key elements, since 1994.16
The December 2000 Hanford waste treatment project construction contract
incorporated key provisions of DOE's contract performance initiative. For
example:
o
In accordance with a key element of its 1994 contract performance
initiative to select a project-appropriate contract type, DOE decided on
16In 1996, we reported on DOE's implementation of the contract performance
initiative and noted that although the department had begun to apply the
initiative's principles in newly negotiated contracts, full implementation
of the initiative's requirements could take years. See U.S. General
Accounting Office, Department of Energy: Contract Reform Is Progressing,
but Full Implementation Will Take Years, GAO/RCED-97-18 (Washington, D.C.:
Dec. 10, 1996). Again in 2002, we reported that although DOE had made
progress in implementing key requirements on its projects, it needed to
take additional actions to ensure that its projects were achieving the
improved results this initiative was intended to achieve. See U.S. General
Accounting Office, Contract Reform: DOE Has Made Progress, but Actions
Needed to Ensure Initiatives Have Improved Results, GAO-02-798
(Washington, D.C.: Sept. 13, 2002). Further, a 1999 review by the National
Research Council noted that DOE has taken steps to reform its contracting
practices but cautioned that major challenges remained, such as to
consistently negotiate contracts that are favorable to the government. See
National Research Council, Improving Project Management in the Department
of Energy (Washington, D.C.: June 1999).
a cost-reimbursement contract type for the waste treatment plant.
Following our criticisms of DOE's earlier privatization approach,17 DOE
decided that a cost-reimbursement contract with incentive fees would be
more appropriate than a fixed-price contract using a privatization
approach for the Hanford project and would better motivate the contractor
to control costs through incentive fees.
o
DOE decided that a competitive procurement was appropriate for the Hanford
project. After holding a series of meetings with interested vendors to
discuss contract options and how to promote competition, DOE established
an open competition for which two competing teams- Bechtel National and
Fluor Vitrification Group-offered proposals. DOE selected the Bechtel
National team, concluding that it offered the best technical approach and
project management, the highest qualifications of key personnel, and the
best record of past performance.
o
Finally, DOE put in place a performance-based approach to hold the winning
contractor accountable for meeting cost and schedule targets. For example,
DOE provided Bechtel National with an opportunity to earn up to $276
million in cost incentives if it met a target cost of $3.97 billion. In
addition, the contract provided Bechtel National with $20 million more if
it met the schedule date of February 2007 for specified testing of the
facility. Nevertheless, although the contractor could earn the fee for
meeting these goals, the contract also included a provision that retaining
all fees above the minimum guaranteed fee was conditional upon the
successful completion of plant testing once construction was complete. If
the contractor did not successfully test the plant at the end of
construction, it would have to repay all of the incentive fees above the
minimum. A DOE official involved with the project at the time said this
provisional fee concept was intended to follow the contract reform
initiative in holding the contractor accountable for the quality of work
performed.
After the contract was awarded, however, costs began to increase and
delays to occur. In response, DOE further adjusted the contract's
incentive structure to encourage the contractor to perform better. For
example, the initial contract fee structure stipulated that the contractor
would receive a
17See U.S. General Accounting Office, Nuclear Waste: Observations on DOE's
Privatization Initiative for Complex Cleanup Projects, GAO/T-RCED-00-215
(Washington, D.C.: June 22, 2000).
minimum fee of $79 million regardless of performance and set a maximum
total fee of $595 million. In a 2003 contract modification, DOE eliminated
the minimum guaranteed fee and required the contractor to meet a
performance standard to earn any fees. The revised contract sets a maximum
fee of $425 million, as follows: $200 million for meeting a target cost,
$60 million for meeting four construction completion milestones, $54
million for meeting other schedule requirements, and $111 million for
meeting performance-testing criteria demonstrating the facility's ability
to treat the waste. DOE made this change following a 2002 review that
concluded that the original incentives were not functioning effectively,
because the cost had risen to a point where the contractor was no longer
incentivized to contain costs. In internal contracting documents, DOE
noted that the current incentive structure is better balanced and should
increase the contractor's motivation to control costs and, at the same
time, improve on-time performance and operations quality.
Project Management Reforms Were Not in Place at Contract Award
Unlike the contract performance initiative, which has existed since 1994,
DOE's project management initiative had just been issued at the time of
the 2000 contract award. Its provisions were in the early stages of
implementation throughout DOE, and no formal implementation guidance had
been issued.18 Based on good project management practices, this project
management initiative establishes a more rigorous decision-making process
containing specific requirements to be completed at each project decision
point. These requirements include preparing an overall project strategy,
performing up-front planning, and having a thorough design and review of
these plans by headquarters. The initiative also requires conducting risk
evaluations of the projects; comparing budgeted with actual expenditures;
and completing certain requirements, such as a significant portion of
design, before proceeding with construction.
Because DOE had not issued formal guidance under its project management
initiative, reforms that the department did not implement in the December
2000 contract included the following:
o
The project management initiative requires that a project acquisition
strategy be developed during the early planning stages of a project. At
18DOE order 413.3-formally adopted in October 2000-identified the basic
requirements that DOE officials were to follow in managing their projects.
Guidance outlining how the initiative's requirements should be implemented
was not adopted until March 2003.
the time of contract award for the construction project, however, DOE did
not have an approved acquisition strategy describing how it planned to
meet long-term goals for treating all of Hanford's tank waste by
regulatory deadlines. Without such a plan, DOE could not demonstrate
whether the plant being built under the construction contract would meet
the long-term goals.
o
DOE's project management initiative stresses that projects must undergo
thorough up-front planning. This process includes waiting until a design
is at a certain level of completion before setting a firm project price.
For simple, less-complex projects, the project management initiative
recommends that a design be up to 35 percent complete before setting the
project price. Although the initiative does not give a definitive
completion design level for more complex projects, according to the Deputy
Director for Project Management in DOE's Office of Engineering and
Construction Management, the aggregate design of a complex project should
adhere to a similar standard, so that a sufficient level of design is
completed in order to make a reasonably accurate cost estimate. Waiting
until a design is more complete, especially on projects like the Hanford
waste treatment plant that are complex, one-of-a-kind nuclear facilities,
allows DOE to identify many of the design uncertainties and to estimate
costs more accurately. Contrary to guidance adopted in March 2003,
however, DOE set its December 2000 contract price when the design was less
than 15 percent complete.
o
The project management initiative recommends that a project's baseline be
reviewed and validated throughout the life of the project. According to
independent reviews performed in 2002, DOE awarded the Hanford
construction contract with a project cost baseline-$3.97 billion-that had
not been appropriately validated.19 Instead, the contract included a
requirement that the contractor review the previous contractor's design
and cost-or perform a "due diligence" review-after signing the contract to
assess needed changes to ensure that the facility would meet requirements.
The independent reviewers concluded that this due diligence requirement
had left it ambiguous regarding whether changes in the project's scope
proposed by the contractor would be allowed as
19See U.S. Department of Energy, External Independent Review, Independent
Cost Review: CD-3C Review of the Waste Treatment and Immobilization Plant
Project (Washington, D.C.: September 2002), and Report of the Independent
Team for the Hanford Waste Treatment Plant Project Review (Washington,
D.C.: July 1, 2002).
contract cost increases. As a result, the project was more vulnerable to
increasing costs.
o
DOE's project management initiative recommends that cost and schedule
baselines be fully achievable and that, to ensure a "high" confidence of
success, they build in appropriate contingency funding in case of
unforeseen events-that is, additional funding and schedule flexibility
based on a project's degree of risk. In the December 2000 contract,
however, DOE accepted a project cost and schedule baseline that had only a
50 percent chance of succeeding on time and within budget, according to an
internal DOE assessment of the construction contract. A subsequent
independent review of the project found that both the project cost and the
contingency allowance were too low for a project of this complexity, and
the reviewers recommended raising the estimated project cost and the
contingency allowance to provide the project with an 80 percent chance of
success.
DOE proceeded with the December 2000 contract award without putting these
key project management requirements in place for two main reasons. First,
according to the former Deputy Manager of the Hanford waste treatment
project, the department believed that it had already met the intent of its
project management initiative-for example, defining the "mission need," or
what the project was intended to accomplish, and specifying certain design
and engineering requirements in the early planning and design stages under
the previous contractor. DOE also had a system in place to monitor
financial information and report on the project's financial status, as
required under the project management initiative. Furthermore, the
December 2000 contract included requirements for the contractor to develop
project management documents either identical or similar to those under
the project management initiative. For example, the contract required the
contractor to develop a project execution plan, summarizing critical
information necessary to manage a project. The contract also required a
quarterly risk assessment report, which met certain project management
requirements for effective risk management. At the time, the then
Assistant Secretary of Environmental Management said that, with these
steps, the project had met the "equivalent" of many project management
requirements and could move forward, despite not having officially
followed the requirements for the earlier planning stages. Second, these
decisions were made while the department was undergoing regulatory
pressure to keep the cleanup moving forward. For example, after canceling
the earlier (privatization) contract in June 2000, DOE committed to
awarding a new construction contract by January 2001, and
the state of Washington was threatening to take legal action if DOE missed
that deadline. DOE had agreed to begin construction by July 2001, and it
wanted to demonstrate that it could meet this regulatory milestone.
Since 2002, DOE Has Been Implementing Management Practices Aimed at
Strengthening Management of the Construction Contract
Two independent reviews in the midst of performance problems after the
2000 contract award identified significant project management deficiencies
and recommended fuller implementation of project management reforms. These
reviews found that many of the required project-planning documents were
either in draft form or inadequately prepared. For example, the reviewers
found that DOE's draft project execution plan was outdated and incomplete.
In addition, the reviewers noted that the December 2000 contract cost
estimate was understated, and, therefore, the actual cost of the project
had not been communicated to DOE headquarters or to Congress. Furthermore,
the reviewers observed that the aggressive schedule in the baseline was
unrealistic. Maintaining such a schedule would compound the risk of
further unanticipated cost increases. The reviewers recommended that the
baseline not be approved with its existing schedule, but rather that the
cost and schedule targets be increased to a more realistically attainable
goal. In addition to recommendations from these reviews to more fully
implement project management requirements, directions in 2002 from the
Assistant Secretary for Environmental Management strongly emphasized
adhering to project management requirements on all DOE projects.
Following the independent reviews' recommendations, DOE has been
implementing its project management initiative on Hanford's waste
treatment project, putting in place certain requirements lacking in the
December 2000 contract, including the following:
o
Developing a project acquisition strategy. DOE is now developing a plan
(acquisition strategy) for how it will meet overall waste treatment goals
for the project. This strategy addresses how DOE can process all of its
waste by the 2028 regulatory milestone. As of May 2004, this acquisition
strategy was under development.
o
Setting a contract price after design is at a greater level of completion.
When DOE renegotiated the construction contract price of nearly $6 billion
in its April 2003 modification, plant design was about 40 percent
complete. A March 2003 internal DOE review of the proposed change to the
contract price, which is based on a greater level of design completion,
stated that the new price appears to be more reasonable.
o
Proceeding with a baseline that has been reviewed and approved. The
Hanford waste treatment project is now operating under a baseline that has
been validated and approved by DOE headquarters, and additional management
practices have been put into place by the current management to take a
more disciplined approach to keeping the project cost and schedule on
track. When it renegotiated the contract price in April 2003, DOE had
established a "minimum essential" initiative in which it identified areas
of work, such as construction of a container storage building, that it
believed were not essential to successful project operation, thus helping
to reduce the project's overall cost. At the contract renegotiation, DOE
identified about $67 million in work that was unnecessary to the project
and could be eliminated.
o
Increasing contingency funding for unforeseen cost increases. The April
2003 revised project baseline included $550 million in contingency funding
to allow DOE an 80 percent chance of meeting the contract cost and
schedule. To better control the use of this funding, DOE also implemented
a joint DOE-contractor managed board to authorize which unforeseen cost
increases will be funded using contingency funds. DOE established this
board to more tightly ensure that unforeseen project changes do not
increase the current cost baseline. In addition to this contingency, DOE
also set aside another $100 million for unforeseen technical and
programmatic risks.
Even with these steps, further concerns about the project's baseline cost
and schedule exist. For example, out of concern about the reliability of
the project's renegotiated cost and schedule, the conference report
accompanying the Energy and Water Development Appropriations Act for 2004
directed the U.S. Army Corps of Engineers to conduct a detailed,
independent review of the project's cost and schedule baseline. The report
is to be issued to the House and Senate Committees on Appropriations and
may be available to the public by early summer 2004.
Not Implementing Key Project Management Reforms at Start of the
Construction Project Has Added to DOE's Challenges
DOE's decision not to implement the project management initiative at the
beginning of the Hanford waste treatment construction project has
increased the risks and uncertainties the department will face in
completing both the construction project and the operations project to
clean up the Hanford Site. Three main consequences of DOE's failure to
implement project management reforms earlier are heightened risk from
using a fast-track approach for plant construction, missed potential
cost saving opportunities, and lack of a sufficient risk management
strategy for DOE to address legal challenges to its waste treatment
strategy.
Fast-track Project Management Approach Increases the Risk of Encountering
Problems
To minimize the time required to complete construction, DOE has followed a
fast-track approach to the construction project, carrying out plant
design, construction, and technology development activities
simultaneously. Under this approach, design and construction activities
take place in stages, with construction of some sections of a facility
under way before those sections are fully designed. DOE's project
management initiative, however, cautions that such a fast-track,
design-build approach should be used only in limited situations, such as
when work scope requirements are well defined, the project is not complex,
and technical risks are limited. When these conditions do not exist, a
fast-track approach is a risky management strategy. According to DOE's
project management guidance, a fast-track approach to large, complex,
one-of-a-kind projects is inherently high risk.
The construction project at Hanford departs from conditions appropriate
for fast-track management. For example, DOE has acknowledged that the
waste treatment plant is the largest, most complex environmental cleanup
project in the United States. The plant incorporates waste treatment
processes that have never previously been combined into facilities the
size and capacity of those envisioned at Hanford. Furthermore, key
uncertainties still exist regarding the condition of the waste to be
treated and the best technologies to use for certain treatment processes.
Despite these risks, DOE has chosen this approach to keep the construction
project moving forward so that the plant can be completed in time to meet
regulatory milestones. DOE selected a contractor for the December 2000
contract award that, it believed, had experience with the fast-track
approach and would be able to maintain the contract schedule. DOE also
took steps to manage the increased risk associated with a fast track
approach, establishing contingency funding measures in case of
unforeseen occurrences and linking performance fees specifically to cost
and schedule goals.
Despite the risk-mitigation actions taken in 2003, the fast-track approach
may result in continuing problems on the construction project. For
example:
o
Construction is outpacing design, resulting in delays in completing
planned work. Although DOE doubled the funding for engineering and other
professional services to help ensure that design work was completed on
schedule, construction has outpaced design. This lag has delayed
completion of interior walls in the treatment facilities and led to the
need to temporarily reassign construction workers to other tasks.
According to the DOE construction project manager, the project baseline
has not yet been affected, but if the problem persists, both cost and
schedule could be affected.
o
Shortening the facility commissioning period may limit plant reliability.
In its April 2003 contract modification, DOE extended the construction
schedule by 16 months because of delays and design changes. To complete
the construction project by 2011 while increasing the construction phase,
DOE reduced by nearly the same period the commissioning phase, when the
facilities' complex treatment processes will be tested and brought into
sound working order. Because of this shorter commissioning time, DOE
reduced the testing to be conducted on actual wastes. As a result, the
contractor plans to test only two of Hanford's four waste types during
this period. According to experts we contacted, including former DOE and
contractor officials and industry technology development managers who have
been involved in plant start-up operations, this commissioning approach
could overlook significant problems until after the plant becomes fully
operational. These experts have noted that corrective actions at that time
could be more costly and time-consuming than if the problems were found
during the commissioning phase. DOE Hanford managers have countered that
although the period for testing plant operations using actual wastes has
been reduced, the period for testing simulated waste has been expanded by
several months. While they expect operational challenges during the
commissioning period, DOE officials believe that their current
commissioning strategy will be sufficient to test the plant adequately.
o
Resolving key technical challenges for processing the waste has fallen
behind. To keep the construction project on schedule, DOE has
acknowledged that it needs to expedite the resolution of key technical
challenges facing the project. One such challenge involves the generation
of hydrogen gas during waste separation activities. The Defense Nuclear
Facilities Safety Board stated in October 2002 and again in September 2003
that problems with this flammable gas, involving buildup of gas in excess
of safety limits, could result in significant safety and operational
problems. Bechtel National has been addressing this issue through design
changes. However, even though it has been more than 1 year since the Board
first raised this issue, this problem has yet to be fully resolved. An
even more critical technical challenge involves how the wastes are mixed
during treatment processes. To expedite resolution of this technical
challenge, Bechtel National initially decided to rely on computer modeling
of special mixers, called "pulse jet mixers." But because computer
modeling did not provide adequate assurance that the mixers would work,
the contractor decided in April 2003, just 9 months before the design
configuration for the mixers was to be completed, to conduct more
stringent tests. Efforts to resolve the uncertainties associated with the
mixers have delayed the testing schedule by more than 4 months, increased
costs by more than $15 million, and postponed the purchase of several
thousand feet of pipe for the treatment facilities. In March 2004, Bechtel
National reported that modifying the facility design to reflect
improvements to the mixers could require an additional $70 million and
take about 16 months to complete. In its March 2004 monthly contract
status report, Bechtel National stated that such delays have affected the
project's critical path and will increase costs.
o
Depending on a technology not fully tested on Hanford tank wastes to meet
regulatory milestones. To meet its regulatory milestone for completing
treatment of Hanford's tank waste by 2028, DOE is attempting to
incorporate a technology before its effectiveness has been fully
demonstrated on Hanford's tank wastes. For example, DOE is relying heavily
on the assumption that a treatment technology called "bulk vitrification"
will succeed. Although this technology is being managed outside of the
scope of DOE's construction project, it is critical to the cleanup program
because the technology is expected to treat up to 60 percent of the
Hanford's low-activity waste. DOE's current plans depend on this
technology to meet its 2028 cleanup date, even though its effectiveness
has not been fully tested. As a result, to demonstrate that this
technology will work on Hanford's tank wastes, DOE has adopted an
aggressive schedule to begin constructing a pilot facility later in 2004
and to conduct formal testing over about 2 years, beginning in May 2005.
Although initial tests with simulated waste have led DOE to consider this
technology promising, Hanford's waste management director has stated that
the size and treatment capacity of a full-scale plant cannot be confirmed
until testing is completed. Because DOE is relying on bulk vitrification
to process the majority of low-activity waste, any problems in developing
this treatment capability will likely extend the duration of the waste
treatment project and increase its overall cost. DOE's initial schedule
may already be threatened because awarding a contract to build the test
facility is about 6 months behind schedule.
Beyond these potential problems with the construction project, the
fast track approach raises other concerns. First, DOE's construction
project manager has acknowledged that the schedule for constructing and
testing facilities and concurrently resolving technical issues continues
to be high risk. In addition, Bechtel National reported in January 2004
that it was concerned about being able to meet its planned construction
schedule. Second, engineering and construction experts we contacted within
the industry, national research organizations, and academia expressed
concern that proceeding with DOE's fast-track approach could create higher
costs, schedule delays, and facilities not fully capable of treating the
waste. Third, a 2002 study by the National Research Council reported that
DOE should proceed with caution when managing "first-of-a-kind" projects
using a fast track approach.20 The council concluded that developing
technology concurrently with project engineering, design, and construction
activities increases an already high degree of uncertainty. The council
said such projects are risky, costly, and likely to produce facilities
needing significant modifications to ensure that they will work properly.
Fourth, the Defense Nuclear Facilities Safety Board cautioned in June
2002, and again in March 2004, that a fast-track approach could lead to
expensive plant modifications or to the acceptance of increased public
health and safety risks. Finally, DOE's record with the fast-track
approach on complex
20See National Research Council, Progress in Improving Project Management
at the Department of Energy, 2002 Assessment (Washington, D.C.: National
Academies Press, 2003).
nuclear cleanup facilities is not good.21 DOE could not identify for us a
single instance where this approach to constructing a large, complex
nuclear cleanup facility resulted in a project that was completed on time
and within budget.
Delays in Evaluating Risks and Costs Associated with Waste Separations May
Cost $50 Million or More
DOE has delayed evaluating the most desirable approach to take in
separating waste components before further waste treatment can occur.
Separating Hanford's waste into its various components is crucial to DOE's
treatment and disposal plans, and, as of April 2004, DOE was relying on a
single supplier of a costly key substance in the separation process. Waste
separation involves sequential filtering and removing major high-level
constituents, such as cesium-137 and strontium-90, from the waste. The
separation process at Hanford includes a technology called "ion exchange,"
in which the waste flows through columns containing a specially designed
substance, or resin, that chemically collects specific high-level waste
components on its surface as they pass through the columns. The high-level
waste components are then washed from the resin and accumulated as part of
the high-level waste stream. Bechtel National will obtain this resin from
a single, small supplier.
DOE faces two significant risks regarding resin supply: acquiring the
resin in sufficient quantities to operate the separation facility and the
cost of the resin DOE selected. The first risk is that DOE decided to
acquire the resin from a small supplier that may not have the long-term
production capacity to supply the resin in required quantities-about 6,250
gallons per year. In a December 2003 letter to Bechtel National, the
manufacturer stated it can produce 1,600 gallons of resin over a 12-month
period to meet initial commissioning needs. Although the manufacturer
assured Bechtel National that it will eventually be able to produce enough
resin for full scale treatment operations, as of April 2004, contractor
and DOE management officials had not verified that the supplier actually
has this
21We have reported on the cost and schedule problems associated with DOE's
use of a fast track, design-build approach in our reviews of other DOE
projects. For example, see U.S. General Accounting Office, Nuclear Waste:
Hanford Tank Waste Program Needs Cost, Schedule, and Management Changes,
GAO/RCED-93-99 (Washington, D.C.: Mar. 8, 1993); Nuclear Waste: Department
of Energy's Project to Clean Up Pit 9 at Idaho Falls Is Experiencing
Problems, GAO/RCED-97-180 (Washington, D.C.: July 28, 1997); Nuclear
Waste: Process to Remove Radioactive Waste from Savannah River Tanks Fails
to Work, GAO/RCED-99-69 (Washington, D.C.: Apr. 30, 1999); and Nuclear
Waste: Management Problems at the Department of Energy's Hanford Spent
Fuel Storage Project, GAO/T-RCED-98-119 (Washington, D.C.: May 12, 1998).
capacity. These officials stated that they are relying on the
manufacturer's December 2003 assurance that it can produce the quantity of
resin needed for full-scale operations. Bechtel National stated that it is
not responsible for ensuring that the supplier can provide resin beyond
its plant commissioning commitments. The second risk is the potentially
high cost of purchasing the resin from a single source without price
competition. Information obtained in December 2003 indicates that the
resin may cost about $10,000 per gallon. Over a single year of operating
the separation facility, resin costs alone could amount to about $62
million. According to the project's research and technology manager,
however, it may be possible to use an alternative, that may be an equally
effective but significantly less expensive resin, that may be available
from commercial suppliers. This alternative resin may cost about $1,000 to
$2,000 per gallon, or about $6 million to $12 million per year. DOE and
contractor officials, including the research and technology manager and
the project's manager, confirmed that pursuing an alternative resin could
substantially increase the reliability of supply and reduce plant
operating costs.
DOE has taken steps to address these project risks, but delays in
developing a viable risk-mitigation strategy may cost DOE $50 million a
year or more in lost savings opportunities. As early as its December 2000
contract award date, DOE recognized that depending on a single resin
source was high risk. Although Bechtel National began evaluating
alternative resins in 2002, DOE took no formal action on the issue until
2003. After we and others raised this issue with DOE in early 2003, the
department modified the contract in April 2003 to require the contractor
to continue research on alternative resins for use as an option to the
resin specified in the December contract. Initial testing on alternative
resins, completed by the contractor in late 2003, showed promising results
for a resin costing significantly less than the original. Given the
estimated cost of the two resins, the potential cost savings of using the
alternative resin for 1 year of operation could result in a savings of
between $50 million to $56 million, according to the project manager. The
contractor, however, did not aggressively pursue development of an
alternative resin, even though it was authorized to do so, deciding to
defer any significant exploration of alternative resins until fiscal year
2007 or later. The contractor explained that an alternative resin was not
needed to achieve its contractual commitment of delivering an operational
facility. In addition, the contractor was unsure whether it could
successfully develop another resin in time for plant commissioning and
said that spending money on an alternative resin could worsen projected
2005 and 2006 funding constraints.
In late February 2004, DOE directed Bechtel National to test and qualify
an alternative resin in time for commissioning the separation facility.
Although testing an alternative resin is important in properly managing
project risk, with the limited time to certify an alternative resin, DOE
could for several years after treatment operations begin be exposed to the
risk of having an uncertain supply of resin available to support waste
separation operations and of having to pay a higher cost for the resin.
The contractor will have about 2 years to fully test and qualify an
alternative resin if the resin is to be available for procurement by
mid-2006 and for plant commissioning in 2008. Two years may be too little
time because the qualifying process took nearly 10 years for the original
resin. Even if the contractor can successfully qualify the resin within 2
years, it will still have to obtain sufficient quantities for plant
commissioning. The contractor's research and technology manager told us
that this schedule is extremely tight and will be challenging to meet.
Thus, DOE may for considerable time be dependent on the production
capability and pricing of a single supplier for a substance critical to
the waste separation process.
DOE Has Not Adequately Assessed or Mitigated a Legal Challenge to Its
High level Waste Treatment Strategy
Legal Challenge to DOE's High level Waste Treatment and Disposal Strategy
Threatens to Derail Cleanup
A third challenge DOE must address to successfully manage its waste
treatment project is to fully assess the risks of a legal challenge to its
approach for treating and disposing of the tank wastes and to develop an
adequate mitigation strategy for addressing those risks. DOE's project
management initiative requires that the department assess significant
risks, quantify the potential impacts, and develop a mitigation plan for
addressing these risks. DOE, however, has not fully assessed the risks
associated with the legal challenge, has only very rough estimates of the
potential impacts on the waste treatment project's cost and schedule, and
has not developed a comprehensive mitigation strategy.
DOE had developed a process for determining the conditions under which
some of its tank wastes could be considered for disposal as other than
high level waste. This process was defined in DOE order 435.1 and was
generally called a process for determining that certain waste resulting
from reprocessing is "waste incidental to reprocessing."22 To meet
criteria set forth in this order for considering tank wastes as other than
high-level wastes, DOE had to determine that the waste (1) has been or
will be
22The phrase "waste incidental to reprocessing" refers to wastes resulting
from reprocessing spent nuclear fuel that DOE considers not to be
high-level waste.
processed to remove key radioactive components to the maximum extent
technically and economically practical; (2) will be disposed of in
conformance with the safety requirements for low-level waste as
established in NRC regulations; and (3) will be put in a solid physical
form and not exceed radioactivity levels set by NRC for the most
radioactive category of low-level waste, referred to as "Class C
standard."23 Once satisfied that these requirements were met for a
specific segment of the waste, DOE planned to obtain a technical review of
this determination from NRC. NRC's role was to determine whether DOE had
appropriately followed the criteria in order 435.1.24 After a favorable
determination from NRC, DOE would have considered the waste segment to be
"incidental" waste for purposes of treating and disposing of the waste.
The legal basis of DOE's plan to apply different treatment and disposal
approaches to segments of its tank wastes has been challenged in the
courts. In March 2002, a lawsuit filed by the Natural Resources Defense
Council and others challenged DOE's authority to proceed with its planned
approach. The plaintiffs argued that DOE's plan violated the Nuclear Waste
Policy Act by allowing high-level radioactive waste to be classified and
treated as something other than high-level waste. In a brief in support of
their position, plaintiffs expressed the concern that DOE would use its
incidental waste determination process to permanently leave intensely
radioactive waste in the tanks after only minimal treatment. In a July 2,
2003, federal district court ruling, the court agreed with the plaintiffs
that the portion of DOE's order 435.1 setting out its incidental waste
determination process violated the Nuclear Waste Policy Act and found that
that part of the order was invalid. The court concluded that the act
"allows DOE to treat the solids to remove fission products, thereby
permitting reclassification of the waste," but does not permit the same
option for the liquid portion of the waste, which may not be reclassified
and must be treated as high-level waste.25 This ruling poses a significant
barrier to DOE's plan to segment the tank wastes for treatment and
disposal purposes, and it raises the possibility that DOE might need to
find
23As specified in NRC regulations (10 C.F.R. S: 61.55(a)(2)(iii)), Class C
low-level waste must meet the most rigorous requirements for low-level
waste form to ensure stability and requires additional measures at the
disposal site to protect against inadvertent intrusion.
24Although order 435.1 does not require DOE to obtain NRC's concurrence
with its incidental waste determinations, DOE did so to obtain an
independent assessment of its evaluation of waste as incidental to
reprocessing.
25Natural Resources Defense Council v. Abraham, 271 F.Supp.2d 1260 (D.
Idaho 2003).
an alternative strategy to its accelerated approach for cleaning up
Hanford's tank wastes. DOE has appealed this decision.
DOE's Risk Assessment and DOE's project management initiative defines risk
as "the measure of the
Mitigation Planning Are potential inability to achieve overall project
objectives within defined
Inadequate
scope, cost, schedule and technical constraints" and emphasizes that all
projects should have a risk assessment and mitigation strategy. Such an
assessment must identify any technical, cost, or schedule risks to the
project; quantify potential cost and schedule impacts; and develop and
implement a strategy to mitigate or properly manage those risks. According
to the Deputy Director for Project Management in DOE's Office of
Engineering and Construction Management, which oversees implementation of
DOE's project management initiative, managers should analyze both the
consequence to the project if the risk occurs and the likelihood of a
risk's occurring. He noted that if the potential impact to the project is
significant and the likelihood of a risk occurring is highly probable, DOE
managers must develop a formal risk-mitigation plan detailing the
potential risks, impacts, and actions taken and planned to best mitigate
the risk. The project management initiative describes effective risk
management as "an essential element of every project" and makes it clear
that risk management should occur continuously throughout a project.
DOE has not conducted a thorough risk assessment of the potential effects
of the legal challenge on the Hanford tank waste treatment project.
Although a draft internal study concludes that the legal challenge
presents a high risk to DOE's high-level waste program,26 DOE has only
very rough estimates of the consequences for its cleanup program. These
estimates, which are not part of Hanford's formal risk management plan,
cover both near-term impacts on cost and schedule if the lawsuit delays
cleanup but is ultimately resolved in DOE's favor and long-term impacts if
the outcome continues to go against DOE. The department estimated that in
the near term, a 2-year slip in cleanup would cost an additional $350
million at
26See Department of Energy, High-Level Waste-Risk Reduction Project
(HLW-RPP): Managing Waste to Reduce Risk-HLW (Washington, D.C.: May 20,
2003).
Hanford. We found no analysis supporting this estimate.27 DOE officials
said this $350 million near-term estimate was not tied to specific
activities on the waste treatment project but represented the approximate
cost of an extra year of maintaining waste in the tanks. Regarding the
long-term impacts if DOE must treat and dispose of all of its tank wastes
as high-level waste, DOE has provided rough, unvalidated estimates for
both the Hanford site ($19 billion) and its entire high-level waste
program (more than $100 billion). Again, we found little rigorous analysis
to support either estimate. For example, the $19 billion Hanford estimate
is based, in part, on a configuration and capacity of the low-activity
treatment facility that DOE is no longer considering because that
configuration could result in an unacceptably high heat load. The Hanford
estimate also includes a potential cost of $6 billion to remove and
dispose of 149 tanks after waste has been retrieved. Information that DOE
provided to us in April 2004, however, indicates this cost could be closer
to $67 billion. Further, the $19 billion Hanford estimate does not include
the additional costs to DOE for disposing of approximately 86,000 more
canisters of treated waste than currently planned in a deep geological
repository. On the basis of a recent disposal estimate of $650,000 per
canister, this cost alone could add $56 billion to the Hanford estimate.28
The estimate for Hanford is part of DOE's complexwide estimate of "more
than $100 billion" cost impact if outcomes continue to go against DOE.
These revised Hanford estimates and other recent information from DOE
indicate that the department's complexwide estimate of more than $100
billion may also be significantly understated. For example, within its
$100 billion estimate, DOE included $30 billion for disposal of additional
canisters that would be produced complexwide. In response to our request,
however, DOE acknowledged the $30 billion was too low and recalculated the
figure, resulting in a revised estimate of more than $90 billion for this
component of its complexwide estimate.
27During DOE's fiscal year 2003 financial statement audit, DOE provided
its independent auditors with the $350 million estimate for Hanford, based
on a delay in selecting a supplemental technology for treating
low-activity waste. According to the contractor's project manager,
however, as of April 2004, DOE was moving forward with trying to award a
contract to test and develop bulk vitrification, the technology DOE has
tentatively selected to treat a majority of Hanford's low-activity waste.
28DOE originally estimated $30 billion to dispose of approximately 180,000
additional canisters at a geological repository. DOE did not define how
much of the $30 billion applied to each site, but recently the department
recalculated those costs and acknowledged that this cost could be $43
billion at Hanford alone (based on a per canister cost of $500,000). We
used the most recent per canister disposal cost of $650,000 provided to us
by the department in arriving at our estimate of $56 billion.
A main aspect of DOE's risk-mitigation planning-its pursuit of a
legislative remedy-has also been inadequate.29 Regarding the legislative
remedy, DOE has not been effective in developing complete and objective
supporting information and effectively communicating that information to
decision makers or other stakeholders. In addition to using only rough
estimates, DOE has raised stakeholder concerns about its true intentions.
For example, after the court ruling, DOE drafted proposed language to
amend the Nuclear Waste Policy Act of 1982, as amended, to give DOE
specific authority to make incidental waste determinations.30 The state of
Washington and others raised concerns that the proposed language would
leave DOE with too much discretion regarding how to make these
determinations. The state of Washington has been concerned that DOE may
use such authority to permanently and inappropriately leave highly
radioactive waste in the tanks. To date, DOE has not been effective in
countering these concerns or in explaining to its stakeholders or other
decision makers why the incidental waste determination process is in the
public's best interest.
The potential consequences for Hanford's tank waste project, and for DOE's
other high-level waste sites, could be far-reaching and significant. DOE's
own analysis suggests several consequences could occur. In the worst case,
wastes could remain on-site untreated for longer periods of time, posing a
continued threat to workers, the public, and the environment. Complexwide,
DOE would likely need to find a different strategy to treat all of its
tank waste for disposal at a geologic repository. Treated waste, at much
larger volumes than now estimated, could overwhelm the currently planned
geologic repository, hastening the need for additional repository
capacity. For example, DOE estimates that it would need to send an
additional 180,000 canisters of waste from all sites to a repository if it
must dispose of most of its waste as high-level waste. DOE's current plans
call for sending about 20,000 canisters to a repository. Furthermore, the
plant now under construction at Hanford may not be configured to most
effectively treat waste if all of the waste is destined for
29DOE's other main risk-mitigation approach involves appealing the court
ruling. After the July 2003 decision, DOE appealed to the Ninth Circuit
Court of Appeals. As of May 2004, no date had been set for oral arguments
in this case.
30Before the court ruling, we had recommended that DOE seek clarification
of its authority from Congress. See U.S. General Accounting Office,
Nuclear Waste: Challenges to Achieving Potential Savings in DOE's
High-Level Waste Cleanup Program, GAO-03-593 (Washington, D.C.: June 17,
2003).
a repository. For example, all of the tank wastes may need to be treated
through the waste treatment plant, making supplemental technology
development unnecessary. Since the current plant would be unable to
process all of the tank waste by 2028 without using supplemental
technologies, DOE may need to consider alternatives, such as building a
second waste treatment plant. DOE does not currently have estimates of
these scenarios.
Despite these deficiencies in risk assessment and mitigation planning, DOE
believes its current risk assessment and mitigation plan are sufficient.
According to DOE, its cost estimates were intended only to illustrate
potential impacts to its high-level waste program in a worst-case
scenario, and, because the department believes it will ultimately be
successful either in the appeals process or through legislative avenues, a
more detailed risk assessment is not needed. In our view, the
uncertainties associated with this legal challenge to its waste treatment
program are precisely why DOE needs a thorough and objective assessment of
the risks and a mitigation plan that includes complete and effective
communication with decision makers and other stakeholders.
Conclusions
After years of effort and after spending hundreds of millions of dollars
and making several false starts, DOE is now constructing a plant to use in
treating and preparing for disposal of a major portion of Hanford's
high level tank wastes. But DOE began plant construction before it had a
complete acquisition strategy in place, and it is carrying out technology
development, plant design, and construction activities simultaneously. In
going forward, DOE faces significant challenges that will likely continue
to affect cost and completion dates, including the challenge of continuing
to manage concurrent development, design, and construction. DOE will need
to minimize the cost and schedule growth that will likely continue and to
take full advantage of cost-reduction opportunities over the life of the
project. DOE must also carefully assess and manage the implications of
legal challenges to its overall approach to treating and disposing of
high level tank wastes. We believe that full disclosure of the potential
impacts to its high-level waste program is important so that policy makers
and others can undertake a more informed debate.
Recommendations for Executive Action
We recommend that the Secretary of Energy take the following two actions:
o
follow more closely DOE's project management order and implementing
guidance when acquiring complex nuclear waste treatment plants at Hanford
and other DOE sites, especially by avoiding a fast-track, concurrent
approach to the design, technology development, construction, and testing
of such plants, and
o
develop and provide to Congress a plan that includes an estimate of the
costs and time frames needed to treat and dispose of Hanford's and the
rest of DOE's high-level tank wastes if the current court ruling is upheld
and if a majority of DOE's tank wastes must be disposed of in a high level
waste repository.
Agency Comments and Our Evaluation
We provided a draft of this report to DOE for review and comment. In
written comments, DOE's Assistant Secretary for Environmental Management
generally agreed with the report's recommendations. The Assistant
Secretary also provided technical comments as an enclosure to the letter,
which we have incorporated as appropriate. DOE's written comments on our
draft report are included in appendix III.
Regarding the report's two recommendations, DOE agreed to follow more
closely its project management guidance when acquiring complex nuclear
waste treatment plants, especially by avoiding a fast-track, concurrent
approach to designing and constructing such plants. Concerning our
recommendation to develop and provide to Congress a plan that includes an
estimate of the costs and time frames to treat and dispose of DOE's
high level wastes, if a majority of the wastes must be disposed of in a
high-level waste repository, DOE agreed that it should develop more
complete information on the costs to give Congress a better sense of the
magnitude of those costs. However, DOE said it was unwilling to develop an
alternative treatment and disposal plan until the outcome of the legal
appeal has been determined. In our view, to be meaningful, any cost and
schedule estimates that DOE develops should be based on a specific
alternative treatment plan.
In an enclosure to the letter, DOE disagreed with our view that it should
use present values when disclosing the cost savings between alternative
approaches. DOE said that present-value techniques tend to hide the full
value of future costs due to discounting, and that using current dollars
provides for a more direct comparison with a project's life-cycle cost and
DOE's budget. However, although using current dollar estimates may be
appropriate for budget purposes, as we discuss in our report, standard
economic analysis as well as Office of Management and Budget guidelines
(see app. II), state that present-value analysis is the standard
methodology to use for comparing costs of different alternatives that
occur at different times (e.g., DOE's accelerated and baseline
approaches). Present-value analysis reflects the time value of money-that
cost savings are worth more if they are incurred sooner and worth less if
they occur in the future. Contrary to DOE's comment that present-value
techniques "hide" the full value of future costs, present-value analysis
reveals the true costs of projects and is the appropriate method to use to
reliably estimate and compare costs.
DOE also disagreed with our statement that its $20 billion cost-savings
estimate for the Hanford waste treatment project is overstated and
misleading. We believe DOE's $20 billion cost-saving estimate is
overstated because it is not based on a present-value analysis as
discussed above, and the estimate is misleading because it is a point
estimate that does not consider uncertainties inherent in the waste
treatment project and implies a degree of accuracy that is not warranted.
Furthermore, as we noted in the report, DOE's $20 billion cost-saving
estimate could not be derived from DOE's reported costs for its baseline
and accelerated approaches. (See app. II.)
As arranged with your offices, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days
after the date of this report. At that time, we will send copies to other
interested congressional committees and to the Secretary of Energy. 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 on this report, please call me on
(202) 512-3841. Other staff contributing to this report are listed in
appendix
IV.
Robin M. Nazzaro Director, Natural Resources and Environment
Appendix I
Scope and Methodology
To develop information on the Department of Energy's (DOE) cleanup
strategy for Hanford tank waste and how the strategy has evolved since
December 2000, we analyzed information and documents provided by DOE
officials and contractors at the Hanford Site and DOE headquarters. We
also toured the project's construction site. To assess the construction
contract, including its cost and schedule, we reviewed the contract;
contract modifications; and related documents, including the contractor's
monthly progress reports and independent and headquarters reviews, and
analyzed various related studies. We also discussed the progress of the
project with Washington State regulators and Environmental Protection
Agency officials; DOE headquarters officials, including the Assistant
Secretary for Environmental Management; and Bechtel National, Inc.,
(Bechtel National) officials. To assist in evaluating the technical
aspects of DOE's approaches, we obtained assistance from our technical
consultant, Dr. George Hinman, who has a Doctor of Science degree in
physics and is Professor Emeritus at Washington State University. Dr.
Hinman has extensive nuclear energy experience in industry, government,
and academia. To develop information on the impact of DOE's initiative to
accelerate cleanup of the waste, we reviewed DOE reports and studies and
discussed them with key headquarters and field staff. We also reviewed
various studies on the proposed supplemental technologies and other
methods being considered by DOE and discussed them with program officials
at the Hanford Site and DOE headquarters. We also discussed these issues
with Washington State regulators. We relied on dollar figures as provided
by DOE but took various steps-such as reviewing cost validation reports,
analyzing budget formulation documents, documenting cost estimating
assumptions, and obtaining clarifications from DOE budget officials-to
ensure that these data were sufficiently reliable for purposes of this
report. To estimate the potential savings from accelerating the project,
we analyzed DOE's life-cycle cost estimates using a commercially available
risk analysis program called "Crystal Ball" (see app. II). We did not
assess DOE's efforts to address any waste remaining in the tanks after
retrieval is completed or DOE's plans to close the tanks. Both of these
topics were outside the scope of our review.
To examine the extent to which DOE's contract performance initiative and
project management initiative have been implemented on the Hanford
project, we reviewed DOE's contract reform report-Making Contracting Work
Better and Cost Less-and DOE order 413.3 and its implementing guidance. We
interviewed DOE headquarters officials from the Office of Engineering and
Construction Management and the Office of Procurement Assistance. We
reviewed the original and modified contract, related project
Appendix I Scope and Methodology
management documentation, and various reviews of the project. We gathered
documentation on contract reforms and project management reforms at
Hanford and also interviewed DOE officials at Hanford. We discussed
contract management and project management issues with outside experts,
including staff from the National Academy of Sciences, the U.S. Army Corps
of Engineers, experts who took part in the 2002 reviews of the waste
treatment contract, and staff from the Department of the Navy's Nuclear
Naval Program.
To determine the impact of DOE's fast-track, design-build plant
construction approach on contract success, we reviewed various DOE and
contractor studies and interviewed knowledgeable officials. In addition,
we discussed this approach with several independent experts, including a
former senior manager of DOE's environmental management program, an Exxon
operations manager involved in facility start-up operations, and our
technical consultant, to obtain their views on any benefits or potential
risks DOE is facing. To determine the potential cost and schedule impact
associated with the resin needed for DOE's waste separation plant, we
reviewed DOE and contractor studies, risk assessments, and cost and
schedule evaluations. We discussed the implications of these studies with
the officials involved; our consultant, Dr. Hinman; and DOE and contractor
officials.
To address the legal challenge DOE faces, we obtained DOE's summary
estimates of life-cycle cost and schedule impacts to its high-level waste
program at its three high-level waste sites if DOE lost the current court
challenge. We also obtained summary information relating to DOE's estimate
of near-term cost increases to its overall high-level waste program if a
2-year delay in the program occurred. DOE provided a summary table that
showed, by site, such information-including additional years of treatment
processing needed if litigation outcome was unfavorable, additional number
of high-level waste canisters that would need to be produced, and
additional storage costs. For the Hanford Site, we also requested, and DOE
provided, more detailed analysis supporting its summary near term and
longer term estimates. The information DOE provided included a summary
cost comparison of four different waste treatment scenarios, including the
current approach DOE is following. We also obtained from DOE basic
assumptions used in calculating these scenarios and qualifications to
these data. We discussed the Hanford estimates with DOE officials. In
addition, we obtained from key DOE officials responses to a series of
questions focusing on the reliability of the Hanford estimates. These
questions covered issues such as the
Appendix I Scope and Methodology
methodology used to develop the estimates, internal review of the
estimates, and confidence level associated with the estimates. Also,
because these same estimates-for DOE's entire high-level waste program,
not just Hanford-were reported in note 17 in DOE's Fiscal Year 2003
Performance and Accountability Report (financial statements), we
interviewed the auditors who performed the review of the financial
statements for DOE. We obtained the auditors' supporting documentation and
discussed with them the reliability of these estimates. Given our review
of the documentation provided by DOE and our discussions with DOE
officials, we have reservations about the reliability of these data. These
issues are discussed in this report.
For determining DOE's risk-mitigation efforts, we gathered documentation
relevant to DOE's legislative proposal. We discussed DOE's risk-mitigation
strategy with officials from the state of Washington. We also asked DOE to
provide its assessment of its risk-mitigation efforts for the Hanford
waste treatment project as a result of the lawsuit concerning its waste
treatment and disposal strategy. We did not evaluate DOE's litigation
strategy.
We conducted our review from July 2003 through May 2004 in accordance with
generally accepted government auditing standards.
Appendix II
Analysis of DOE's Cost-Saving Estimates
DOE has estimated that by accelerating cleanup of the high-level tank
wastes at its Hanford Site, it can realize savings of as much as $20
billion by shortening the time frame for completing the project by almost
20 years and better aligning waste treatment and disposal approaches with
the wastes' characteristics. Our review of DOE's savings estimate suggests
that the estimate is overstated and misleading. The savings estimate is
overstated because DOE did not consider the time value of money or
properly adjust its estimates for inflation. The savings estimate is
misleading because DOE used point estimates of costs that do not consider
uncertainties inherent in this type of analysis, such as projecting costs
for new or untested technologies or treatment options.
o
DOE's savings estimate is overstated because it did not consider the time
value of money or properly adjust for inflation. According to standard
economic analysis and guidance developed by the Office of Management and
Budget (OMB),1 cost-comparison analyses should be based on life-cycle
costs of competing alternatives with future costs discounted to present
values, that is, adjusted both for inflation and the time value of money.
According to this guidance, DOE should have first converted the annual
expected costs of cleanup for both its baseline estimate and its
accelerated approach estimate to their present value in 2003 and then
compared the two present-value costs. In contrast, DOE's comparison of its
baseline estimate with its accelerated estimate is based on values that
are simply the sum of current dollar values of future costs. As a result,
DOE's methodology does not reflect an accurate comparison of its baseline
estimate with the accelerated approach estimate.
o
DOE's cost-saving estimate is misleading because it did not consider
uncertainties inherent in DOE's waste cleanup program. According to OMB
guidance, agencies should attempt to characterize the nature and source of
uncertainty associated with their data and report data uncertainties or
the full range of values within which their estimates can fall.
Uncertainties inherent in DOE's tank waste program include the future
costs of new technologies and waste treatment options that are not fully
tested and the overall duration of waste treatment activities. For
example, DOE expects to reduce treatment costs significantly based on a
technology called "bulk vitrification." The life-cycle cost estimate
1OMB Circular No. A-94, Guidelines and Discount Rates for Benefit-Cost
Analysis of Federal Programs (Washington, D.C.: Oct. 29, 1992).
Appendix II
Analysis of DOE's Cost-Saving Estimates
developed for this technology, however, is preconceptual and will likely
increase as this technology is developed, according to the bulk
vitrification project manager. DOE has not developed an estimate of the
range of possible costs for this technology and has not reflected this
uncertainty in its cost-savings estimate.
To correct for DOE's methodological errors and illustrate their
potentially significant effect, we obtained DOE's available annual cost
and budget estimates for the project, but we considered the effects of the
time value of money and inflation. In addition, although we were not able
to adjust DOE's cost-savings estimate to demonstrate the effects of all
uncertainties, for illustrative purposes we included the uncertainty
associated with changes in interest rates. Rather than using one interest
rate in the savings estimate, as DOE did, we included in our analysis a
range of interest rates, which resulted in a range of possible project
costs.2
Table 3 summarizes the differences in the cost-savings estimate between
DOE's approach and our present-value analysis that reflects the
uncertainty associated with future interest rates. Our analysis shows that
the estimated savings from DOE's accelerated approach could range from $10
billion to $13 billion, with a mean value of $12 billion. In other words,
the estimated cost savings DOE can expect from implementing its
accelerated cleanup approach is about $12 billion, or about half of the
projected savings the department has reported.
2We assumed that real (adjusted for inflation) annual interest rates would
range from a minimum of 2.92 percent to 5.39 percent with the likeliest
value of 3.84 percent. We used an Excel spreadsheet and a commercially
available risk analysis program, called "Crystal Ball," to perform this
analysis. This program randomly selects values for the interest rate from
the given range and uses each value in the spreadsheet to calculate the
savings.
Appendix II
Analysis of DOE's Cost-Saving Estimates
Table 3: GAO Analysis of DOE's Cost-Savings Estimate for the Hanford Tank
Waste Treatment Project
Dollars in billions
Baseline Accelerated Estimated Description of estimates estimate estimate
savings
DOE's life-cycle cost estimates for
fiscal year 2001 baseline and 2003
accelerated approach (current
dollars) $56a $27b $20c
GAO's present-value analysis
results for the DOE baseline and
accelerated estimate (constant
2003 dollars) $26d $14d $12d
Range of present-value analysis
results (constant 2003 dollars) $23-$29e $13-$15e $10-$13e
Source: GAO analysis of DOE data.
aAccording to the project's budget team leader, this figure is the
approved 2001 baseline (including adjustments).
bThis figure is Hanford's approved 2003 accelerated estimate.
cThe actual difference is $29 billion. Since implementation of its
accelerated initiative, DOE has stated publicly that it expects to achieve
savings of $20 billion.
dThese figures represent the means of our present-value analysis.
eValues represent the maximum and minimum of the range of estimated values
from the present-value analysis when uncertainties are incorporated.
Appendix III
Comments from the Department of Energy
Appendix III
Comments from the Department of Energy
Appendix IV
GAO Contacts and Staff Acknowledgments
GAO Contacts Robin M. Nazzaro (202) 512-3841 William R. Swick (206)
287-4800
Staff In addition to those individuals named above, Chris Abraham, Ellen
Chu, Bob Crystal, Doreen Eng, Doreen Feldman, George Hinman, Nancy
Acknowledgments Kintner-Meyer, Mehrzad Nadji, Tom Perry, Emily Pickrell,
and Stan Stenersen made key contributions to this report.
GAO's Mission 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.
Obtaining Copies of GAO Reports and Testimony
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 e-mail alerts" under the "Order GAO Products"
heading.
Order by Mail or Phone The first copy of each printed report is free.
Additional copies are $2 each. A check or money order should be made out
to the Superintendent of Documents. GAO also accepts VISA and Mastercard.
Orders for 100 or more copies mailed to a single address are discounted 25
percent. Orders should be sent to:
U.S. 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
To Report Fraud, Contact:
Web site: www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in E-mail:
[email protected] Federal Programs Automated answering system: (800)
424-5454 or (202) 512-7470
Public Affairs Jeff Nelligan, Managing Director, [email protected] (202)
512-4800 U.S. General Accounting Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
Presorted Standard
Postage & Fees Paid
GAO
Permit No. GI00
United States General Accounting Office Washington, D.C. 20548-0001
Official Business Penalty for Private Use $300
Address Service Requested
*** End of document. ***