DOD Accounting Systems: Efforts to Improve System for Navy Need Overall
Structure (Letter Report, 09/30/96, GAO/AIMD-96-99).

The Defense Finance and Accounting Service (DFAS) is seeking to reduce
the number of Navy accounting systems and to enhance a Navy-wide system
to account for general fund operations. GAO found that enhancements to
DFAS' Standard Accounting and Reporting System were not guided by a
target systems architecture--a critical step in any systems development
effort. As a result, DFAS' attempts to upgrade the system and correct
its shortcomings have not been adequately planned in conjunction with
Navy, the system's primary user, to minimize technical and economic
risks. As a result, the likelihood has increased that the large
investment already made and planned for this project will not provide
the Navy with a reliable, fully integrated general accounting system.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  AIMD-96-99
     TITLE:  DOD Accounting Systems: Efforts to Improve System for Navy 
             Need Overall Structure
      DATE:  09/30/96
   SUBJECT:  Federal agency accounting systems
             Systems design
             Financial management systems
             Strategic information systems planning
             Funds management
             General fund accounts
             Accounting procedures
             Human resources training
IDENTIFIER:  Navy Standard Accounting and Reporting System
             DFAS Interim Migratory Strategy
             STARS
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed 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.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Department of Defense Officials

September 1996

DOD ACCOUNTING SYSTEMS - EFFORTS
TO IMPROVE SYSTEM FOR NAVY NEED
OVERALL STRUCTURE

GAO/AIMD-96-99

DOD Accounting Systems

(918844)


Abbreviations
=============================================================== ABBREV

  CFO - chief financial officer
  DOD - Department of Defense
  DFAS - Defense Finance and Accounting Service
  FMFIA - Federal Managers' Financial Integrity Act
  FMSO - Fleet Material Support Office
  OMB - Office of Management and Budget
  STARS - Standard Accounting and Reporting System

Letter
=============================================================== LETTER


B-272079

September 30, 1996

The Honorable John J.  Hamre
The Under Secretary of Defense (Comptroller)

The Honorable Deborah P.  Christie
The Navy Assistant Secretary for
 Financial Management and Comptroller

Mr.  Richard F.  Keevey
Director, Defense Finance and Accounting Service

This report presents our analysis of the Defense Finance and
Accounting Service's (DFAS) efforts to reduce the number of Navy
accounting systems and to enhance and implement a Navy-wide system to
account for general fund operations.  We performed this work as part
of our review of the reliability of the Navy's fiscal year 1994
consolidated financial reports, which we reported on in March 1996.\1
In that report, we disclosed serious financial management and
reporting problems that undermine Navy's efforts to ensure proper
accountability and drain resources needed for military readiness. 


--------------------
\1 CFO Act Financial Audits:  Increased Attention Must Be Given to
Preparing Navy's Financial Reports (GAO/AIMD-96-7, March 27, 1996). 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

DFAS selected the Standard Accounting and Reporting System (STARS) to
serve as Navy's system for general fund accounting.  In the process,
STARS has taken over the functions of some outdated field-level
systems and is expected to produce some net cost savings, primarily
from not having to maintain and operate those older Navy systems. 

DFAS selected STARS because it was believed to be the least deficient
of Navy's 25 existing accounting systems.  Although the least
deficient system, STARS still had serious shortcomings.  For example,
it did not comply with Department of Defense (DOD) basic financial
requirements, such as DOD's standard general ledger and key
accounting requirements.  DFAS plans to enhance STARS to overcome
these serious deficiencies, but its efforts are hampered by the lack
of a target systems architecture that would define the systems'
expected functions, features, and attributes, including internal and
external interfaces, and data flows for all STARS components.  The
lack of an architecture--or blueprint--also has contributed to
problems with the enhancement efforts, such as incomplete planning
and date-driven or missing milestones, and makes it difficult to
evaluate alternatives that may be more effective or less expensive. 
DFAS's piecemeal approach to the STARS improvement project could
result in costly and time-consuming redesign efforts in the future. 

In addition, implementation of STARS at field-level activities we
reviewed was not completely successful.  For example, Navy activities
we visited said that Navy field staff received limited training and
DFAS provided insufficient technical support.  The Director of the
STARS Project Office has acknowledged these problems and has taken or
planned actions to correct them. 


   BACKGROUND
------------------------------------------------------------ Letter :2

The Navy operates thousands of shore activities worldwide and employs
over 1 million military and civilian personnel.  Hundreds of Navy
Fund Administering Activities at the field level provide budgeting
and accounting for more than 9,000 Navy cost centers.  A cost center
is the lowest level in the Navy's financial management chain of
command and may be a ship, an aircraft squadron, a staff office, or a
department or division of a shore activity where identification of
costs is required. 

DFAS, which organizationally reports to the Under Secretary of
Defense (Comptroller), was created in January 1991 to streamline and
standardize DOD's finance and accounting procedures, systems, and
operations while reducing the costs of such services.  Ownership of
many DOD financial management systems, including some Navy systems,
was transferred to DFAS.  At the same time, many mixed financial
management systems,\2 such as logistics systems which provide
inventory and property financial data, remained under the control of
the services or other DOD components.  Policy guidance and direction
related to accounting system development and operations is performed
by the DOD Comptroller's Office. 

In the past, Navy activities reported financial information to
Defense Accounting Offices maintained by DFAS which, in turn,
generated monthly reports to DFAS Cleveland Center.  Beginning in
fiscal year 1995, DFAS began consolidating its Navy Defense
Accounting Offices into six sites called operating locations. 
Information from these operating locations will continue to be
reported to DFAS Cleveland Center.\3

Navy financial management systems have many long-standing problems
that prior system development efforts have not fixed.  For example,
according to the Naval Audit Service, during the 1980s, Navy spent
over $260 million on two failed attempts to consolidate and
standardize its accounting systems.  Both the Standard Automated
Financial System and the Integrated Disbursing and Accounting
Financial Management System were terminated in 1989 due to design
problems, excessive cost escalations, and slippages in completion
time. 

Navy's financial management system problems continue even though many
of Navy's systems have been transferred to DFAS.  For example, DOD's
fiscal year 1995 Federal Managers' Financial Integrity Act (FMFIA)
report stated that a majority of DOD's financial systems did not
comply with Office of Management and Budget (OMB) Circular A-127 and
that many perform similar functions, resulting in inefficiencies and
disparate business practices.  Navy's Statement of Assurance, used to
prepare the DOD Federal Managers' Financial Integrity Act report,
also stated that its systems did not comply with Circular A-127's
integration, accounting classification codes, and general ledger
requirements. 

In August 1991, DFAS began to review DOD's accounting systems and to
develop a plan to (1) decrease the number of financial systems and
(2) correct systems deficiencies.  In December 1993, DFAS developed
an Interim Migratory Strategy, consisting of the following two
phases. 

  -- First, DFAS planned to reduce DOD's 91 general fund systems to
     11 interim migratory systems.  The interim migratory systems
     were to be selected from the military services' existing
     financial systems that were considered to be the least
     deficient.  In addition, DFAS planned to identify and correct
     the interim migratory systems deficiencies.  In particular, DFAS
     plans called for enhancing the systems to comply with DOD's
     standard general ledger, key accounting requirements,\4 and
     standard budget and accounting classification code\5 by October
     1, 1997. 

  -- Second, DFAS planned to eventually select the best interim
     migratory system(s) to implement DOD-wide as its target
     system(s).\6 DFAS has not established time frames for this
     effort. 

In June 1994, DFAS Cleveland Center selected STARS Field Level to
serve as Navy's interim migratory system for field-level general fund
accounting.  The Assistant Secretary of the Navy for Financial
Management concurred with STARS' selection.  DFAS Cleveland and Navy
personnel deemed STARS the newest, least deficient, and most advanced
of Navy's 25 existing general fund accounting systems.  Through April
30, 1996, the largest of the former field-level systems had been
converted to STARS Field Level at 923 activities, and 122 activities
remained to be converted to STARS Field Level.  In addition to STARS
Field Level, the STARS umbrella includes other components--an on-line
bill paying component and components that provide financial data to
Navy's major claimants.\7 DFAS Cleveland Center is also in the
process of developing a STARS financial reporting module at the
departmental level.  The STARS enhancement effort is intended to
ensure that all STARS components comply with DOD's key accounting
requirements and standard general ledger as well as to implement
other improvements. 

In April 1996, DFAS created the Defense Accounting System Program
Management Office to centrally manage the consolidation and
modernization of DFAS accounting systems.  After the period of our
review, DFAS selected STARS as one of three DFAS target general fund
accounting systems,\8 under the oversight of this Program Management
Office. 


--------------------
\2 A mixed system supports both financial and nonfinancial functions. 

\3 The following GAO reports contain our assessment of this
consolidation effort:  DOD Infrastructure:  DOD Is Opening Unneeded
Finance and Accounting Offices (GAO/NSIAD-96-113, April 16, 1996) and
DOD Infrastructure:  DOD's Planned Finance and Accounting Structure
Is Not Well Justified (GAO/NSIAD-95-127, September 18, 1995). 

\4 The 13 key accounting requirements are (1) general ledger control
and financial reporting, (2) property and inventory accounting, (3)
accounting for receivables, including advances, (4) cost accounting,
(5) accrual accounting, (6) military and civilian payroll procedures,
(7) system controls, including funds control and internal control,
(8) audit trails, (9) cash procedures and accounts payable, (10)
system documentation, (11) system operations, (12) user information
needs, and (13) budgetary accounting.  We did not evaluate these
requirements to determine whether they comply with federal financial
system requirements. 

\5 DFAS is developing the standard budget and accounting
classification code with the objective of providing a structure for
financial data to ensure comparable and consistent financial
information reporting.  Because the code is not finalized, we did not
examine the proposed budget and account classification code's
adequacy in meeting this objective. 

\6 The DOD Inspector General has called into question the strategy's
overall goal of enhancing seriously deficient systems--which entails
a significant investment of time and money--for later consolidation
into yet another DOD-wide system.  In a recent report, the Office of
the Inspector General recommended that the DFAS interim migratory
strategy be cancelled and that a single DOD-wide system approach be
adopted for general fund accounting.  The General Fund Interim
Migratory Accounting Strategy (DOD Office of the Inspector General,
report no.  96-180, June 26, 1996). 

\7 Major claimants are Navy bureaus, offices, commands, or
headquarters designated to receive operating budgets for operation
and maintenance appropriations and procurement appropriations
directly from the Office of the Chief of Naval Operations. 

\8 The other two target general fund accounting systems selected were
the Standard Accounting, Budgeting, and Reporting System and a
modified Corps of Engineers Financial Management System. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :3

To assess Navy's efforts to reduce the number of accounting systems
and implement and enhance STARS, we examined DOD, DFAS, and Navy
documents and conducted interviews with appropriate officials.  We
also reviewed STARS system documentation and compared it to the
financial management system architecture guidance in the Joint
Financial Management Improvement Program's Framework for Federal
Financial Management Systems.  We also interviewed the Director of
the STARS Project Office on this issue. 

To evaluate DFAS Cleveland Center's plans to enhance STARS, we
examined STARS project planning documentation, such as its April 18,
1996, Plan of Action and Milestones and software project plans.  We
also reviewed analyses related to implementing various key accounting
requirements in STARS prepared by an outside contractor.  We
interviewed officials from this contractor, the STARS Project Office,
DFAS Cleveland Center, and Navy's Fleet Material Support Office
(FMSO)--which serves as the primary STARS Central Design Agency.\9

To review the implementation of STARS at the field level, we
judgmentally selected 18 Navy shore activities that converted to
STARS Field Level between July 1993 and March 1995, from a universe
of east and west coast activities.  We visited each of these sites,
examined documents related to various aspects of Navy's financial
management operations, and identified areas applicable to or
interfacing with field-level accounting and reporting.  We obtained
sample financial reports with explanations of their purpose and use,
conducted interviews with field-level financial managers, and
reviewed supporting accounting and reporting documentation.  During
this review, we noted any problems concerning accuracy, timeliness,
and usefulness and examined their cause and resultant effect on
field-level financial operations. 

We performed our work at the Office of the DOD Comptroller, DFAS
Headquarters, DFAS Cleveland Center's STARS Project Office, Navy's
FMSO (Mechanicsburg, PA), and the 18 Navy shore activities listed in
appendix I. 

The Department of Defense provided written comments on a draft of
this report.  These comments are presented and evaluated in the
"Agency Comments and Our Evaluation" section of this report and are
reprinted in appendix II.  Our work was performed from April 1995
through early August 1996 in accordance with generally accepted
government auditing standards. 


--------------------
\9 A Central Design Agency is responsible for system analysis,
program development, and system maintenance/enhancements. 


   SOME COST SAVINGS ARE EXPECTED
------------------------------------------------------------ Letter :4

We believe that savings will accrue as a result of eliminating the
duplication and inefficiencies of supporting and maintaining Navy's
25 existing accounting systems, although we did not attempt to
quantify such savings.  In October 1994, a contractor to the STARS
Project Office completed an economic analysis that compared the costs
of developing and implementing STARS to continuing with the existing
systems.  This economic analysis estimated that converting five of
Navy's existing general fund accounting systems to STARS (primarily
at the field level) would save $162 million in the first 5 years. 
This initial estimated savings, primarily in maintenance and
operating costs, was reduced by the estimated STARS Field Level
system development, maintenance, operating, and training costs of
$145 million, for a net savings of $17 million.  Over 15 years, net
savings were projected to total $293 million. 

We did not assess the reliability of these estimates.  However, we
note that achieving the projected level of net savings could be
diminished by the need to provide additional training and technical
support to field activities in using STARS Field Level.  These issues
are discussed later in this report.  In addition, the October 1994
economic analysis was not a total life-cycle economic analysis of all
STARS components.  For example, it did not include costs to enhance
all of the STARS components to bring them into compliance with DOD's
standard general ledger, key accounting requirements, and the
standard budget and accounting classification code.  The Major
Automated Information Systems Review Council,\10 which is reviewing
the STARS project, directed the STARS Project Office to conduct a
total life-cycle economic analysis of all STARS components.  This
analysis is expected to be completed by December 31, 1996.  In
addition, as discussed in the next section, DFAS has not developed a
complete system architecture for STARS--basically a blueprint for
what the system will do and how it will operate.  As a result, any
estimate of STARS total costs will be incomplete. 

We also found that, in less than 2 years, actual obligations to
enhance STARS in certain areas were significantly higher than
budgeted.  The STARS Project Office estimated that in fiscal years
1995 and 1996, STARS software development costs would total $35.6
million.  Of this amount, $18.8 million was budgeted for projects
pertaining to the key accounting requirements, the budget and
accounting classification code, and the consolidation efforts.\11 As
of July 23, 1996, actual obligations for the software projects
related to these three areas were $24.5 million, or 30 percent, above
the budget estimate, although total STARS software development
obligations were close to what was estimated.  In addition, some
software development projects have already exceeded their total
budget.  For example, the STARS Project Office estimated that for
fiscal years 1995-1997 (1) property and inventory accounting and (2)
cash and accounts payable key accounting requirement enhancements
would each cost $500,000.  However, as of July 23, 1996, obligations
for these enhancement efforts were $1.1 million and $1.2 million,
respectively.  Neither of these projects is scheduled to be completed
in fiscal year 1996, although, according to DFAS, an accounts payable
function was implemented for one STARS component--STARS Field Level
(but not for the STARS claimant module). 

In addition, STARS budget estimates were incomplete and lacked
supporting documentation.  For example, these budget estimates did
not include DFAS' internal costs, such as the STARS Project Office. 
In fiscal year 1996, the STARS Project Office personnel costs alone
were estimated at $1.4 million.  Moreover, the STARS Project Office
could not find documentation for much of the July 1995 budget
estimate and instead provided us with a written rationale on the
methodology it used to estimate the STARS software development costs. 
According to this rationale, the STARS Project Office consulted with
FMSO and, based on these discussions, used prior projects of similar
scope and size as a basis for the estimates.  However, the STARS
Project Office used projects related to only one STARS component to
estimate the cost of enhancing all of the STARS components.  Each
STARS component would require different levels of effort to modify
since the components do not have the same program attributes. 


--------------------
\10 The Major Automated Information Systems Review Council is
convened to review system development projects' compliance with DOD
life-cycle management policy, procedures, and standards when (1)
program costs for any single year exceed $30 million, (2) total
program costs exceed $120 million, or (3) total life-cycle costs
exceed $360 million.  STARS is undergoing such a review because it
met these criteria. 

\11 The STARS budget estimate did not include any costs to implement
the key accounting requirements and the budget and accounting
classification codes past fiscal year 1997.  The total fiscal year
1995-1997 STARS software development budget estimate was $50 million,
of which $21.7 million pertained to the key accounting requirements,
budget and accounting classification code, and consolidation efforts. 


   ENHANCEMENT EFFORTS NOT GUIDED
   BY SYSTEMS ARCHITECTURE
------------------------------------------------------------ Letter :5

The STARS enhancement effort is not guided by a target system
architecture.  A target systems architecture is a composite of all
interrelated functions, information, data, and applications
associated with a system.  Specifically, such a systems architecture
is an evolving description of an approach to achieving a desired
mission.  It describes (1) all functional activities to be performed
to achieve the desired mission, (2) the system elements needed to
perform the functions, (3) the designation of performance levels of
those system elements, and (4) the technologies, interfaces, and
locations of functions.  The lack of a target STARS architecture
increases the likelihood of project failure, additional development
and maintenance costs, and a system that does not operate efficiently
or effectively. 

Moreover, the information contained in a complete architecture
provides the opportunity to perform a thorough alternatives analysis
for the selection of the most effective system at the least cost. 
According to a February 1994 DFAS memorandum, the decision to choose
STARS Field Level was "an intuitive one based on the collective
experience of the capitalized Navy field general fund accounting
network...." We have found that successful organizations manage
information systems projects as investments and use a disciplined
process--based on explicit decision criteria and quantifiable
measures assessing mission benefits, risks, and cost--to select
information system projects.\12 In addition, recent OMB guidance
recommends that agencies select information technology projects based
on rigorous technical evaluations in conjunction with executive
management business knowledge, direction, and priorities.\13 Further,
the Congress and the administration recognized the value of treating
information system projects as investments by enacting the
Information Technology Management Reform Act of 1996 (Public Law
104-106, Division E), which calls for agency heads, under the
supervision of the Director of OMB, to design and implement a process
for maximizing the value of their information technology acquisitions
and assessing and managing the associated risks, including
establishing minimum criteria on whether to undertake an investment
in information systems.  Managers can use the detailed information
found in the systems architecture to enhance their analysis of these
critical issues. 

Although STARS was selected without the benefit of an established
architecture, such an architecture can provide needed structure and
discipline as the STARS enhancement projects move forward.  For
example, it is unlikely that Navy and DFAS could ever achieve the
requirement set forth in the Chief Financial Officers (CFO) Act and
OMB Circular A-127 that agencies implement an integrated financial
management system\14 without the structure provided by a STARS
financial management systems architecture.  In order to implement a
single, integrated financial management system, Circular A-127
specifies that agencies should plan and manage their financial
management systems in a unified manner with common data elements and
transaction processing.  A critical step in accomplishing this is the
development of a financial management systems architecture. 

According to the Joint Financial Management Improvement Program
Framework for Federal Financial Management Systems, a financial
management systems architecture provides a blueprint for the logical
combination of financial and mixed systems to provide the budgetary
and financial management support for program and financial managers. 
Preparing a financial management system architecture is also
consistent with the best practices we found\15 in leading
organizations, which established and managed a comprehensive
architecture to ensure the integration of mission- critical systems
through common standards.  In addition, DOD Directive 7740.2,
Automated Information System Strategic Planning, states that
automated information system strategic processes shall be supported
by information architectures that address the information
requirements, flows, and system interfaces throughout the
organization, including headquarters, major commands, and separate
operating agencies. 

Although the decision to enhance STARS was made over 2 years ago,
DFAS Cleveland Center has not yet developed a target STARS system
architecture which would include a definition of the systems'
expected functions, features, and attributes, including internal and
external interfaces, and data flows.  An architecture is particularly
critical since several of the STARS enhancements are not only to
correct existing system problems but are expected to add new
functions to STARS (either programmed as part of STARS or through
interfaces with other systems) at considerable cost.  For example,
the STARS Project Office plans to add an accounts payable function to
the STARS claimant module and plans to interface STARS with a
property system.  (STARS does not currently collect property data.)

In addition, without an architecture, DFAS is not in a position to
reasonably estimate the total cost to enhance STARS.  For example, as
previously mentioned, actual obligations to enhance STARS to comply
with the cash and accounts payable and property and inventory key
accounting requirements were already double the total $500,000 budget
estimate for fiscal years 1995-1997, even though these projects were
still in the planning stage.  Further, complex development efforts
such as these pose a greater technical risk of failure which can be
mitigated by developing a target architecture. 

The Director of the STARS Project Office agreed that a STARS
architecture should be developed.  He stated that he plans to develop
a STARS architecture that would include identifying the data sources
of systems that interface with STARS.  Although STARS is a DFAS
system, many of the systems that interface with STARS are controlled
by Navy.  As a result, a STARS architecture cannot be developed
without the direct involvement of Navy and the identification of all
feeder systems, interfaces, and supportive detailed data elements. 
Therefore, it is imperative that DFAS and Navy's Assistant Secretary
of the Navy (Financial Management and Comptroller) work cooperatively
to develop a STARS target architecture to ensure that STARS will meet
the needs of its primary user in an effective manner. 


--------------------
\12 Executive Guide:  Improving Mission Performance Through Strategic
Information Management and Technology (GAO/AIMD-94-115, May 1994). 

\13 Office of Management and Budget, Evaluating Information
Technology Investments:  A Practical Guide, November 1, 1995. 

\14 To be most effective, a Navy financial management systems
architecture should be developed within the context of a DOD-wide
financial management systems architecture.  However, DOD does not
have a financial management system architecture. 

\15 Executive Guide:  Improving Mission Performance Through Strategic
Information Management and Technology (GAO/AIMD-94-115, May 1994). 


   INCOMPLETE PLANNING AND
   UNREALISTIC MILESTONES HAMPER
   ENHANCEMENT EFFORTS
------------------------------------------------------------ Letter :6

Our analysis noted instances of incomplete planning and missing or
slipped milestones that strongly suggest that STARS enhancements will
not meet DOD requirements in the near future.  We believe that these
problems are symptomatic of the lack of a STARS system architecture. 
As one of the first steps in any systems development effort, the
development of an architecture would guide the enhancement efforts
and set the appropriate time frames for the completion of major
tasks. 

One example that highlights STARS architecture and planning issues is
DFAS' evaluation of how another system could provide property
accounting data to STARS.  A June 1996 contractor analysis of this
property system found that differences between the STARS and the
property system's lines of accounting would have to be resolved
before an interface is developed.  Moreover, as of July 1, 1996, the
property system had been implemented at only one Navy site and only
eight additional sites have been scheduled to implement the property
system.  As a result, even if the interface issues between STARS and
the property system were resolved, only a very limited amount of Navy
property data could be transmitted.  Additionally, a STARS contractor
was directed not to work on certain problem areas related to the cash
procedures and accounts payable key accounting requirement.\16

According to the STARS Project Office Director, the necessary
analysis that the contractor was to complete will be done internally,
although no specific plans existed as of early August, 1996. 

We also found several instances of milestones that were date-driven
rather than based on an analysis of the tasks to be completed.  DFAS'
September 1995 Chief Financial Officers Financial Management 5-Year
Plan stated that STARS key accounting requirement deficiencies would
be corrected by September 30, 1997.  However, the April 18, 1996,
STARS Plan of Action and Milestones stated that STARS enhancements
would comply with DOD's key accounting requirements by October 1,
1996.  The plan included no reason for the accelerated time frame. 
The STARS Project Office Director stated that the October 1, 1996,
milestone for completing the programming, testing, and data
conversion for modifying STARS to comply with the key accounting
requirements was not derived from an assessment of the scope of these
projects.  Rather, the implementation date was established to
coincide with the Navy's requirement to prepare and have audited
financial statements.\17

In addition, we reviewed the April 18, 1996, STARS Plan of Action and
Milestones and found that it did not include several key analysis
tasks that are needed to successfully implement the DOD standard
general ledger and some of the key accounting requirements.  The Plan
of Action and Milestones also indicates missing and slipped
milestones.  For example, we found that the plan

  -- did not address how one of the STARS modules, which currently
     has its own general ledger account structure, will be brought
     into compliance with DOD's standard general ledger;

  -- did not address how STARS will be enhanced to comply with the
     audit trail key accounting requirement which states that all
     transactions be traceable to individual source records
     maintained in the system;

  -- did not specify how the systems analysis for enhancing STARS
     field-level and headquarters claimant modules to meet the key
     accounting requirement for budgetary accounting will be
     performed and by whom, and did not provide for analyzing and
     documenting the current environment and identifying needed
     changes, which is the approach planned in making most key
     accounting requirement analyses;

  -- did not provide for identifying needed changes and solutions to
     control weaknesses as part of the analysis of the current
     environment related to the system control function of STARS
     field-level and headquarters claimant modules; and

  -- provided, in several cases, milestones for completing the
     analyses of the current STARS environment and planning for
     future STARS enhancements that were dated several months before
     a contractor was scheduled to provide them. 

We also found that the lack of an overall plan or architecture
contributed to the lack of participation of one of Navy's key systems
development offices.  Specifically, although Navy's FMSO is the
primary Central Design Agency for STARS, it has had a limited role in
the STARS enhancement project.  For example, in December 1994, the
STARS Project Office tasked FMSO with completing, by December 1995,
functional descriptions and/or system specifications to enhance STARS
to comply with six key accounting requirements, including those
related to accounts receivable and accounts payable.  On September
14, 1995, FMSO was also tasked with completing, by March 31, 1996, an
expanded functional requirement analysis and detailed system
specifications for the key accounting requirements related to general
ledger control and financial reporting for the STARS claimant module. 

A project status report dated May 30, 1996, showed that FMSO had not
completed these tasks and had (1) spent little time on the analyses
required for the accounts receivable, cash procedures and accounts
payable, and general ledger and (2) spent no time on the other key
accounting requirements analyses.  According to FMSO officials, the
STARS Project Office Director instructed them to work on other
priorities.  Additionally, FMSO officials stated that they did not
know whether the milestones and costs for modifying the STARS
components to comply with the key accounting requirements were
reasonable because the scope of the modifications to be made were not
known. 


--------------------
\16 These problem areas relate to (1) suspense transactions, (2)
successor checks, (3) military and civilian payroll interface
procedures, and (4) accounting for travel. 

\17 The Chief Financial Officers Act of 1990, as expanded by the
Government Management Reform Act of 1994, requires major agencies,
including DOD and the services, to prepare financial statements and
have them audited, beginning with those for fiscal year 1996. 


   FIELD-LEVEL IMPLEMENTATION
   COULD BE IMPROVED
------------------------------------------------------------ Letter :7

Our review of STARS Field Level implementation at 18 Navy shore
activities disclosed problems related to training and technical
support.  Specifically, field staff received limited training. 
Representatives of over half of the activities told us that the STARS
Field Level training (1) did not focus on areas specifically related
to their daily jobs, (2) was provided by instructors, often
contractors, that had STARS Field Level knowledge but did not have a
working knowledge of Navy accounting and/or the activity's existing
accounting system, and (3) did not include follow-up training in most
cases.  Further, only about one-half of these activities had received
training in using available software that would allow them to use the
system more effectively and efficiently.\18 After we brought these
training deficiencies to the attention of the Director of the STARS
Project Office, he agreed that training needed to be improved. 
According to the Director, the STARS Project Office has collected
information on the field activities' training needs and plans to
develop a set of training requirements.  However, he stated that
additional STARS training will be contingent on available funding. 

We also found that DFAS provided insufficient STARS Field Level
technical support.  For example, representatives at six activities
cited DFAS' failure to provide a central focal point or people with
sufficient knowledge to provide timely answers to questions and
responses to problems.  The Director of the STARS Project Office
agreed that STARS technical support was a concern.  He stated that he
planned to consider options to address this concern and that better
training would also reduce the number of user problems. 


--------------------
\18 These include the on-line query management function, a "help"
function to assist users in utilizing the system more efficiently,
and the capabilities of the MONARCH software, which enables users to
download data from STARS Field Level to extract, reformat, and
analyze information. 


   CONCLUSIONS
------------------------------------------------------------ Letter :8

Because the DFAS STARS enhancement project was not guided by a target
systems architecture--a critical step in any systems development
effort--DFAS' efforts to enhance STARS and correct numerous
shortcomings have not been adequately planned, in conjunction with
Navy, the system's primary user, to mitigate technical and economic
risks.  This is particularly true for planned new STARS functions,
such as property, which would entail the greatest risk.  As a result,
the likelihood that the large investment already made and planned for
this project will not yield a reliable, fully integrated Navy general
fund accounting system is increased.  In addition, STARS
implementation has been hampered by limited training and insufficient
technical support, which will have to be addressed as the enhancement
project moves forward. 


   RECOMMENDATIONS
------------------------------------------------------------ Letter :9

To increase the likelihood that the STARS enhancement project will
result in an efficient, effective, and integrated Navy general fund
accounting system, we recommend that the Under Secretary for Defense
(Comptroller), in conjunction with the Assistant Secretary of the
Navy (Financial Management and Comptroller), expeditiously develop a
target STARS architecture.  As part of this process, the Comptroller
should (1) identify the economic and technical risks associated with
the implementation of STARS enhancements, (2) develop a plan to avoid
or mitigate these risks, and (3) obtain the Major Automated
Information Systems Review Council's assessment and approval.  Until
this architecture is complete, the Comptroller should cease the
funding of enhancements to STARS components that add new functions to
STARS. 

Also, once a target STARS architecture has been developed and
approved, we recommend that the Director, DFAS, enhance its Plan of
Action and Milestones to ensure that it contains (1) the steps that
will have to be taken to achieve this architecture, including key
analysis tasks which relate to how STARS modules will meet the key
accounting requirements, (2) the parties responsible for performing
these steps, and (3) realistic milestones. 

In addition, to improve STARS Field Level's day-to-day operations at
the field level, we recommend that the Director, DFAS,

  -- provide additional user training, particularly in functions that
     allow users to use the system more effectively and efficiently
     and

  -- provide a central focal point for enhanced technical support
     through such means as establishing a "hot line" staffed by
     knowledgeable personnel. 


   AGENCY COMMENTS AND OUR
   EVALUATION
----------------------------------------------------------- Letter :10

In providing written comments on a draft of this report, DOD
generally agreed with our findings but did not concur with our
overall recommendation that it cease funding of STARS enhancements
until the target architecture is completed.  The full text of DOD's
comments is provided in appendix II. 

DOD's response stated that since 1991, DOD has made substantial
functional and technical improvements, compliance improvements, and
significant financial reporting refinements to STARS.  While DFAS has
implemented some STARS improvements, STARS does not yet fully comply
with DOD's key accounting requirements or standard general ledger,
which is why the enhancement effort was started. 

With respect to our recommendations, DOD agreed that a STARS target
architecture must be completed which includes the identification of
source data in the target system for all interfaced systems. 
However, DOD stated that STARS is a fully operational system with a
documented architecture of current interfaces, processes, and
procedures except for Navy-owned logistic systems.  According to DOD,
as new enhancements are added to STARS, they will be added to the
target architecture.  We disagree that STARS has a current documented
architecture.  DFAS was unable to provide us with an architecture. 

In addition, DOD did not concur with our recommendation to stop
funding enhancements that add functions to STARS until the target
architecture is complete.  DOD's comments indicated that STARS
enhancements must continue so that the migratory strategy can be
completed as soon as possible because (1) Navy's funding has been
either curtailed or terminated beginning in fiscal year 1997 in
anticipation of completing the enhancements, (2) key accounting
provisions are needed in the current system to establish needed
controls and meet CFO reporting requirements, and (3) a "learning
curve" situation would be created because personnel resources would
have to be removed and later returned to the initial staffing level. 

Continuing to develop STARS enhancements without the benefit of a
completed target architecture runs counter to the basic purpose of
developing such an architecture--to provide structure and discipline
to a system enhancement effort before changes are made to ensure that
the best decisions are made in terms of operational effectiveness,
flexibility, maintenance, and cost.  Although Navy has funds
available now to work on the enhancements, to spend them without a
proper planning effort has not proven in the past to be an effective
use of resources.  Without a target architecture, DOD runs a high
risk of spending millions of dollars enhancing STARS and implementing
a system that still will not meet the CFO Act financial reporting
requirements nor be developed in a timely and cost-effective manner. 
Indeed, as we discussed in the report, the STARS enhancement project
has already experienced incomplete planning, missed milestones, and
budget overruns. 

In regard to DOD's point that personnel resources would have to be
removed and later returned to the initial staffing level, creating a
"learning curve" situation, we believe that any personnel currently
assigned to the enhancement efforts could be reassigned to the
architecture development effort.  This would allow them to use the
expertise they have gained from working on the enhancements to
efficiently produce an accurate and complete target architecture. 
Once the architecture is completed, these personnel could then
continue to use their expertise on the systems development efforts
that DFAS and Navy decide to pursue in light of the architecture
results. 

DOD concurred with our remaining recommendations.  In regard to the
establishment of a "hot line," DOD's response noted that it had
established a "hot line" to address technical system problems at DFAS
Cleveland Center and the Defense Mega Center in Mechanicsburg.  DOD's
response also stated that, by December 31, 1996, DFAS will perform a
follow-on review to determine the feasibility of expanding the "hot
line" service to DFAS operating locations. 


--------------------------------------------------------- Letter :10.1

We are sending copies of this report to the Chairmen and Ranking
Minority Members of the Senate Committee on Armed Services, the House
Committee on National Security, the Senate Committee on Governmental
Affairs, the House Committee on Government Reform and Oversight, and
the House and Senate Committees on Appropriations.  We are also
sending copies to the Secretary of Defense, the Secretary of the
Navy, and the Director of the Office of Management and Budget.  We
will also make copies available to others on request. 

The head of a federal agency is required by 31 U.S.C.  720 to submit
a written statement on actions taken on these recommendations to the
Senate Committee on Governmental Affairs and the House Committee on
Government Reform and Oversight within 60 days of the date of this
report.  You must also send a written statement to the House and
Senate Committees on Appropriations with the agency's first request
for appropriations made more than 60 days after the date of this
report. 

If you or your staffs have any questions concerning this report,
please contact me at (202) 512-9095.  Major contributors to this
report are listed in appendix III. 

Lisa G.  Jacobson
Director, Defense Financial Audits


LIST OF ACTIVITIES VISITED
=========================================================== Appendix I

Headquarters, Naval Audit Service, Arlington, VA

Navy Reserve Readiness Command, Region Six, Washington, DC

Naval Hospital, Pensacola, FL

Naval Air Station, Pensacola, FL

Naval Technical Training Center, Pensacola, FL

Naval Education and Training, Program Management Support Activity,
Pensacola, FL

Fleet Combat Training Center-Dam Neck, Virginia Beach, VA

Naval Amphibious Base-Little Creek, Norfolk, VA

Naval Air Station-Oceana, Virginia Beach, VA

Naval Hospital, Newport, RI

Naval War College, Newport, RI

Naval Reserve Readiness Command, Region One, Newport, RI

Naval Station, San Diego, CA

Naval Air Station, North Island, San Diego, CA

Naval Air Station, Whidbey Island, Oak Harbor, WA

Trident Refit Facility, Bangor Bremerton, WA

Naval Submarine Base, Bangor Silverdale, WA

Strategic Weapon Facility, Pacific, Silverdale, WA




(See figure in printed edition.)Appendix II
COMMENTS FROM THE DEPARTMENT OF
DEFENSE
=========================================================== Appendix I



(See figure in printed edition.)



(See figure in printed edition.)


The following are GAO's comments on the Department of Defense's
letter dated September 26, 1996. 

GAO COMMENTS

1.  See the "Agency Comments and Our Evaluation" section of this
report. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix III

ACCOUNTING AND INFORMATION
MANAGEMENT DIVISION, WASHINGTON,
D.C. 

Roger R.  Stoltz, Assistant Director
Janett P.  Smith, Assistant Director
Linda J.  Lambert, Senior Auditor
James L.  Ariail, Jr., Senior Auditor
Francine M.  DelVecchio, Communications Analyst

NORFOLK REGIONAL OFFICE

Christopher M.  Rice, Senior Evaluator
Rebecca S.  Beale, Senior Evaluator

SEATTLE REGIONAL OFFICE

Pat L.  Seaton, Senior Evaluator
Julianne H.  Hartman, Senior Evaluator


*** End of document. ***