_.-. .-- .._.-...._____. . -__--..--_.-.--. --y-1_ - .JuIy I t I’M) INFORMATION RESOURCES Army Corporate Data Base Disregards Congressional and DOD Direction RESTRICTED --Not to be released outside the General Accounting Of&e unless speclflcally approved by the Of&e of Congressional l&e-iatious. GAO/‘1 M’IW ‘-Wfi4 Information Management and Technology Division B-239633 July 19, 1990 The Honorable John P. Murtha Chairman, Subcommittee on Defense Committee on Appropriations House of Representatives Dear Mr. Chairman: This report responds to your September 5,1989, request that we review the Department of the Army’s efforts to develop a corporate data base. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies to interested parties and make copies available to others upon request. The report was prepared under the direction of Samuel W. Bowlin, Director, Defense and Security Information Systems, who can be reached at (202) 275-4649. Other major contributors are listed in the appendix. Sincerely yours, lbdzAJ+~ Ralph V. Carlone Assistant Comptroller General c Ekecutive Summq In fiscal year 1988, the Army canceled its Corporate Data Base Project Purpose because it could not address congressional and Secretary of Defense questions about mission need and potential economic benefit. The Army estimated that the project would cost approximately $130 million, excluding some development and all operations and maintenance costs. The Chairman of the Subcommittee on Defense, House Committee on Appropriations, asked GAO: . to determine whether the Army is pursuing the objectives of the can- celed project through other efforts, and if so, . to compare the merits of the Army’s current efforts to those of a cen- trally directed program subject to Army and Defense oversight. The Army began the Corporate Data Base Project in 1984 to improve the Background quality of information used to make personnel, equipment, financial, and other decisions. The project called for the development of an Army- wide data base that would be accessible to decision-makers at the head- quarters, major command, and installation levels. Although portions of the data base would be located at these three organizational levels, it would ultimately operate as a single data base. The Congress denied the Army’s fiscal year 1987 funding request for the project because the Army had not adequately defined requirements, performed a cost/benefit analysis, or determined the total project cost. The Congress also directed the Army to submit the project to the Office of the Secretary of Defense for an oversight review. After the September 1986 oversight review, the Office of the Secretary of Defense directed the Army to justify the need for the project, deter- mine its total costs and benefits, and establish a management approach for controlling its development and operation. In 1988, the Army can- celed the project because it could not provide the required justification or determine the project’s potential economic value. Without regard for congressional and Department of Defense direction, Results in Brief the Army is continuing to develop an ad hoc corporate data base. GAO identified eight initiatives at the headquarters, major command and ” installation levels that are intended to provide capabilities identical or very similar to the canceled data base project. Because these initiatives are not centrally-directed or in compliance with pertinent federal and Page2 GAO/IMTEG90~4Army CorporateDataBase ExecutiveSummary Defense acquisition policies, the Army cannot be sure that these systems will meet valid mission needs, will work together and not duplicate one another, or that they are cost effective. In addition, the Army has no idea how much money is being spent on these eight initiatives or what it will cost to develop, maintain, and operate the systems. Principal Findings The Army Is Pursuing GAO'Swork at Army headquarters, two major commands, and selected Corporate Data Base installations identified eight system development initiatives which will provide capabilities identical or similar to the canceled Corporate Data Objectives Base Project. The headquarters-level initiatives involve the development of a standard data base, an unknown number of decision support sys- tems, and a long-distance communications system. The initiatives at the major command and installation levels primarily involve the develop- ment of standard data bases. While GAOidentified eight ongoing initia- tives, there may be other Army organizations with initiatives also pursuing the objectives of the canceled project. Army Management of GAOalso found that the Army’s management of the initiatives lacks the Initiatives Lacks Meri ts of merits of a centrally-directed program. Specifically, the initiatives lack a coordinated implementation plan and guidance because the Army has a Centrally-Directed not established a program office or completed its information architec- Program ture (i.e., framework for how the systems would fit together). Conse- quently, the Army is not sure that the initiatives will work with and not duplicate others. The Army also is not sure that the initiatives will be able to exchange data useful to Army decision-makers because it has not fully implemented its data standardization program, which is intended to eliminate conflicting and inaccurate data. Additionally, although federal and Defense policies call for require- ments determinations and economic analyses prior to system develop- ment, the organizations controlling seven of the eight initiatives had begun system development but had not completed these requirements. The organization responsible for the eighth initiative plans to comply with the policies. Therefore, the Army lacks assurance that seven of the eight initiatives are based on valid requirements. The Army also does not know whether the most economical approach has been selected for Page3 GAO/IMTEG90-54 Army CorporateDataBase ExecutiveSummary the seven initiatives or how much it may cost to develop any of the ini- tiatives. In addition, the Army did not know how much had been spent on the initiatives because the organizations controlling them have not fully tracked costs. GAOrecommends that the Secretary of Defense direct the Secretary of Recommendations the Army to suspend development activity on the eight initiatives GAO identified, and not to resume any of them unless all required federal and Defense acquisition requirements are met. This would include preparing mission needs statements and requirements and economic analyses. In addition, GAOrecommends that the Secretary of Defense direct the Sec- retary of the Army to prohibit funding for any other initiative whose purpose is to achieve a corporate data base capability until the Army adequately justifies to the Office of the Secretary of Defense and the Congress the mission need and potential economic benefit of an Army- wide corporate data base. At the Chairman’s request, GAOdid not obtain official agency comments Agency Comments on a draft of this report. However, GAOdid discuss the results of this review with agency officials and they agreed with the facts presented. Page4 GAO/IMTEG9084Army CorporateDataBase Page5 GAO/IMTJ3G9044 Army CorporateDataBase Contents , Executive Summary 2 Chapter 1 8 Introduction History of the Army’s Corporate Data Base Congressional and Defense Reaction to the Army 8 10 Corporate Data Base Project Objectives, Scope, and Methodology 10 Chapter 2 13 The Army Is Pursuing Headquarters Initiatives Are Pursuing Canceled Data Base Objectives 14 Corporate Data Base Two Major Command Initiatives Are Following Corporate 15 Project Objectives Data Base Objectives Installation Initiatives Are Pursuing the Objectives of the 15 Canceled Project Chapter 3 17 Army Management of Initiatives Lack Central Direction Initiatives Are Not Based on Analysis of Need or 17 18 Initiatives Lacks Consideration of Economic Benefit Merits of a Centrally- Directed Program Chapter 4 24 Conclusions and Conclusions Recommendations 24 25 Recommendations Agency Comments 25 Appendix Major Contributors to This Report 26 Tables Table 2.1 Comparison Between the Initiatives and 13 Corporate Data Base Project Objectives Table 3.1 Project Compliance With Federal and Defense 20 Acquisition Policies Page6 GAO/IMTEG90-64 Army CorporateDataBase \ Contents Abbreviations ACSIM Assistant Chief of Staff for Information Management DOD Department of Defense IQRSCOM Forces Command Y GAO General Accounting Office IMTEC Information Management and Technology Division TRADOC Training and Doctrine Command Page7 GAO/iMTEG!M84Army CorporateDataBase Chapter 1 Introduction , The Army began the Corporate Data Base Project to improve the quality of information used to make personnel, equipment, organizational struc- ture, and budget decisions affecting the entire Army. The project was intended to correct the adverse impact that inconsistent, duplicate, and conflicting data was having on these decisions by providing a repository of consistent, up-to-date information, Army organizations at the headquarters, major command, and installa- tion levels manage data needed for personnel, logistics, acquisition, operations, facilities, and budget and finance decisions. Headquarters is the focal point for pulling together data for all Army decisions. Below the headquarters level are several major commands that manage data within functional areas corresponding with headquarters. Two of the Army’s major commands are the Training and Doctrine Command (TRADW) and the Forces Command (FORSCOM).TRADOC is responsible for training all soldiers and establishing doctrine on how the Army will be organized and equipped. FORSCOMis responsible for the operations and readiness of all active and reserve Army units in the continental United States, Alaska, Puerto Rico, the Virgin Islands, and Panama. The organizational level below the major commands consists of the installations. Worldwide the Army has 179 installations, with TRADOC and FORSCOMcontrolling 16 and 17 of them, respectively. Since the bulk of the Army personnel, equipment, facilities, and other assets are at installations, this level creates or processes much of the data needed for Army-wide decisions. In September 1983, the Vice Chief of Staff announced the need for a History of the Army’s corporate data base, highlighting the adverse effect that conflicting and Corporate Data Base inconsistent data was having on Army-wide decisions. In 1983, the Army contracted with the American Management Systems and the Sys- tems Research Applications Corporations for a joint study of the problem. In March 1984, the two firms issued a report on the Army’s corporate data base. According to the 1984 report, the Army’s systems did not effectively support the information needs of the Army-wide decision process. Many of the systems managed data from the organizational or functional per- spectives of their developers, without regard for overall Army informa- tion needs. While some systems operated at one or more of the Army’s three organizational levels, many of the systems actually manipulated Page8 GAO/IMTEG90-64 Army CorporateDataBase chapter 1 Introduction the same data or subsets of the same data for different purposes. In addition, the systems collectively contained inconsistent or incomplete data. In June 1984, the Assistant Chief of Staff for Information Management (ACSIM) directed the United States Army Management Systems Support Agency to develop a prototype to help evaluate the technical feasibility of a corporate data base. The Support Agency developed the prototype and determined that a corporate data base was technically achievable. In December 1985, on the basis of the results of the prototype, the ACSIM issued a corporate data base concept paper and a plan of action. The ACSIMpaper indicated that conceptually the corporate data base would be available to anyone with the appropriate clearance. Portions of the data base would be located at each of the three organization levels, but would operate together as a central data base. The ACSIM plan established guidelines for development of the Army Cor- porate Data Base and assigned the Information Systems Command responsibility for the project. The Command delegated program manage- ment responsibility to its Information Systems Engineering Command, which designated a project manager for the Army Corporate Data Base Project in August 1986. In September 1986, the Army Corporate Data Base Project Office issued an implementation strategy, which called for a three-phased develop- ment of the proposed data base. Phase 1 called for the development of separate data bases with a common data structure that would be con- nected via a communications network to meet the information require- ments of the headquarters, major commands, and installations. The common data structure was to be defined in an Army-wide data dic- tionary. In this phase, users would be able to obtain data from the data bases on the network if they knew its location. Phase 2 specified the development of a distributed data base to provide shared access to data. With this system capability, users would be able to obtain data from any of the separate data bases developed under phase 1 without knowing its location because the data bases would begin to operate as one. Phase 3 called for the incorporation of artificial intelligence, or enhanced deci- sion aides, into the distributed data base. Artificial intelligence capabili- ties would be made available to help users quickly analyze multiple alternatives, select and locate data, or formulate questions for decision- making purposes. The Army estimated that the project would cost approximately $130 million, excluding some development and all opera- tions and maintenance costs. Page9 GAO/IMTEC9O-64 Army CorporateDataBase Chapter1 r Introduction While reacting favorably to the data management disciplines associated Congressional and with the Corporate Data Base Project, the Congress denied the Army’s Defense Reaction to fiscal year 1987 appropriations request to buy computers for the project the Army Corporate because of concerns about the approach. In its August 14,1986, report on the Department of Defense Appropriations Bill, 1987, the House Data Base Project Committee on Appropriations stated The Committee is excited about the Army’s plans to develop. . . a central approach to management of data. However, a significant financial investment will be neces- sary to accomplish these objectives. The amount required has not been disclosed to the Congress. Further, it appears that the formal requirements determination, eco- nomic analysis, and other prerequisites dictated by the Defense Department’s life cycle management policy for acquisition of computers are not in place for the Corpo- rate Data Base.... Therefore, hardware acquisition at this time is premature. The Committee would be receptive to a reprogramming request during fiscal 1987 if the Comptroller of the Defense Department determines it is warranted after a formal oversight review of Army plans. The Appropriations conferees agreed with the House Committee and expressed concern that the acquisition had not been approved by the Major Automated Information Systems Review Committee. Further, the conferees directed the Army to limit its obligation of funds to those needed for planning and defining requirements. On September 23,1986, the Defense Major Automated Information Sys- tems Review Committee reviewed the Army’s Corporate Data Base Pro- ject. In a December 1986 memorandum, the Defense Committee noted that the Army’s justification for the project was based on generic problems and did not demonstrate that the corporate data base would satisfy specific mission needs or result in economic benefit. Further, the memorandum noted that the Army had not adequately addressed tech- nical, managerial, financial, or operational issues. The Committee directed the Army to demonstrate all aspects of developing and oper- ating the data base before proceeding with the project. Although the Army developed a demonstration, it canceled the Corporate Data Base Project in fiscal year 1988 because it could not define the mission criti- cality or economic benefit of the project. On September 5,1989, the Chairman, Subcommittee on Defense, House Objectives, Scope,and Committee on Appropriations, asked GAO whether the Army is contin- Methodology uing to pursue the objectives of the canceled Corporate Data Base Pro- ject. Our objectives were to (1) determine whether the Army is still pursuing the objectives of the Corporate Data Base Project through Page10 GAO/lMTEC-9044 Army CorporateDataBase . chapter 1 Introduction various headquarters, major command, and installation-level initiatives; and (2) if so, compare the merits of the Army’s current approach with that of a centrally-directed program subjected to Defense and Army oversight. To accomplish our objectives, we identified and reviewed selected Army data base development efforts. We performed our work at the Depart- ment of Defense and Army headquarters in the Washington, D.C. area; the Information Systems Command at Fort Huachuca, Arizona; the Information Systems Engineering Command at Fort Belvoir, Virginia; TRUKMJheadquarters at Fort Monroe, Virginia; FORSCOM headquarters at Fort McPherson, Georgia; and at Army installations in Fort Sill, Oklahoma, and Fort Hood, Texas. For background on the Army’s original project, we interviewed knowl- edgeable Defense and Army officials, reviewed pertinent Army studies and analyses, and Corporate Data Base Project documents. We also reviewed Department of Defense Inspector General, Department of the Army Inspector General, and Army Audit Agency reports. To determine whether current Army projects are pursuing corporate data base objectives, we reviewed Army planning documents for the canceled project to identify key project objectives and organizations responsible for accomplishing them. We contacted the responsible orga- nizations to determine their progress toward accomplishing the identi- fied objectives. On the basis of this work, we identified eight initiatives and compared the objectives of the identified initiatives to those for the canceled Corporate Data Base Project. Given the time constraints of our review and the original scope of the canceled project, we did not attempt to identify all initiatives following the objectives of the canceled project. Additionally, we traced the origins of the current initiatives to deter- mine whether they were part of the original Corporate Data Base Project. To compare the merits of the Army’s current approach to that of a cen- trally directed program subjected to Defense oversight, we reviewed the Army’s Corporate Data Base Project planning documents to identify the expected benefits of central project management. Also, we assessed the current initiatives to determine whether they were being developed in accordance with federal, Defense, and Army policies and regulations for automated information systems. We also spoke with officials from the National Institute of Standards and Technology on their guidelines for prototype development with data base management systems. Page11 GAO/IMTEGQOdM Anuy CorporateDataBase Chapter1 Introduction We conducted our review from September 1989 to May 1990 in accor- dance with generally accepted auditing standards. As the Chairman requested, we did not obtain formal written comments from the Depart- ment of the Army on a draft of this report. We did, however, discuss the issues in this report with officials from the Office of the Secretary of Defense and the Department of the Army and they agreed with the facts presented. Page12 GAO/IMTEG90-64 Army CorporateDataBase Chanter 2 The Army Is Pursuing Corporate Data Base Project Objectives We identified eight Army information system development initiatives that are pursuing the objectives of the canceled Corporate Data Base Project. Specifically, we identified three headquarters, two major com- mand, and three installation-level initiatives following various objec- tives of the canceled project. The three headquarters-level initiatives involve the development of a standard data base, an unknown number of decision support systems, and a long-distance communications system, The five initiatives at the major command and installation levels primarily involve the development of standard data bases, There may be other initiatives at Army headquarters organizations, major com- mands, and installations that we did not visit which are also pursuing objectives of the canceled project. As noted in chapter 1, the Army planned to implement the corporate data base in three phases. Phase 1 involved development of standard data bases at the headquarters, major command, and installation levels and use of telecommunication systems to permit remote users to access them. Phases 2 and 3 called for the use of distributed data bases and artificial intelligence, respectively. The following table illustrates how the objectives of the eight initiatives compare with those of the canceled Corporate Data Base Project. Table 2.1 Comparison Between the initiatives and Corporate Data Base Corporate Data Base Objectives Project Objectives Phase 1 Phase 2 Phase 3 (Standard (Telecom- (Distributed (Artificial Initiatives data base) municationsl data base) intellbence1 Headquarters Decision resources data base ---- X Decision -____- support systems X X X ..__ X $nno;kanagement support a X -__- Major Command ~_____---. TRADOC decision support systems X X X X 63SCOM command data base X X X installation AmrzdLw;de installation support X TRAD~C installation support modules X X Fort Hood integrated data base X An X indicates that the initiative is pursuing the corporate data base objective. “This network permits users to access data bases at the headquarters, major command, and installation levels. Page13 GAO/IMTEG90-64 hy CorporateDataBase Chapter2 I TheArmy Is PursuingCbrporateDataBase ProjectObjectives The three headquarters initiatives involve the development of a stan- Headquarters dard data base called the Decision Resources Data Base, decision sup- Initiatives Are port systems under the Headquarters Decision Support Systems effort, Pursuing Canceled and a long-distance communication system called the Army Management Support Network. The Information Systems Command at the Pentagon Data Base Objectives is developing the Decision Resources Data Base, which is intended to provide a repository of data to improve headquarters decision-making. A December 1988 Army information paper on the data base called it the headquarters component of the Army’s Corporate Data Base. Two sys- tems that will initially use the data base process organizational, per- sonnel, and equipment data on all Army units from around the world. Army officials familiar with both the canceled project and the data base agreed it is pursuing the headquarters objectives of the canceled project. The headquarters Decision Management Agency is pursuing another ini- tiative, the Headquarters Decision Support Systems, which involves the development of an unknown number of computer applications and data bases that follow objectives of the canceled corporate data base project strategy. Similar to objectives in the Army’s 1986 Corporate Data Base Concept Paper and Plan, the systems are being developed to improve Army decision-making by allowing users access to data in standard and other data bases at the headquarters, major command, and installation levels. Like the canceled project, the Headquarters Decision Support Systems call for the use of artificial intelligence and distributed data bases. Under this initiative, the Army also developed a telecommunica- tions system called the Headquarters Decision Support Systems Net- work, which permits over 2,000 users to access the systems. The 1987 Decision Support Systems Master Plan calls for implementing these sys- tems Army-wide. Headquarters Decision Systems Management Agency officials agreed that the initiative is pursuing many objectives estab- lished for the canceled Corporate Data Base Project. The Information Systems Command at the Pentagon has also installed a network like that called for in the Army’s Corporate Data Base Project strategy to provide long-distance communications capability. The Army Management Support Network provides long-distance communications among headquarters, TRADOC,FORSCOM, Depot Systems Command, and Fort Sill-all key sites called for in the corporate data base plans. In June 1985, the Army installed the first circuit in the network to increase the data available to the headquarters decision support systems. Army officials responsible for the network indicated that the fundamental objectives of the network are similar to those of the Corporate Data Base Project. Page14 GAO/IMTEGSO84 Army Caporate.DataBase Chapter2 TheArmy Is PursuingCorporateDataBase ProjectObJectives The two major command initiatives are TFiADoC’sDecision Support Sys- Two Major Command terns effort and FORSCOM’S development of a command data base. Started Initiatives Are in July 1987, TRAWC’Sinitiative is a prototype of the headquarters deci- Following Corporate sion systems effort. As called for in the canceled Corporate Data Base Project, the TRAKXXeffort involves development of a corporate data base Data Base Objectives capability available to decision-makers throughout the Command. Addi- tionally, the Command has developed an automated data dictionary to document standard Command data definitions and established a policy that Command software developers use the data dictionary. TRAIXX?has installed a command-wide telecommunications network, called the TRAWL Decision Support System Network, to permit users to exchange data among the 16 Command installations and with Army headquarters via the Army Management Support Network. The Com- mand’s automation plans also call for the development of a distributed data base capability and the potential use of artificial intelligence capa- bilities like those called for in the canceled project, Officials responsible for the TRADoCDecision Support System agree that the objectives of this effort are the same as those for the Army’s canceled corporate data base initiative. Although not as extensive as the TRADoCprogram, IWWOM has informa- tion system initiatives underway that are pursuing corporate data base objectives. The FORSCOM initiatives, which started in 1986, primarily focus on developing a standard Command data base for its headquarters decision-makers. The Command also is promoting common data struc- tures and documenting data standards in a Command dictionary. Com- mand plans call for linking Command headquarters and installation- level systems and data bases via a telecommunications network. FORSCOM officials agreed that the Command is pursuing objectives similar to those of the Army Corporate Data Base. The three installation-level initiatives pursuing objectives of the can- Installation Initiatives celed project are the Army-wide Installation Support Modules Project, Are Pursuing the the TR.ADoCInstallation Support Modules initiative, and the Fort Hood Objectives of the Integrated Data Base initiative. The Army-wide Installation Support Modules Project and the TRADoCInstallation Support Modules initiative Canceled Project are different attempts to expand the Army’s use of support systems developed at Fort Sill, a TRADOCinstallation. Fort Sill developed the Installation Support Modules to permit its users interactive access to information processed in the Army’s standard automated information systems, which are centrally designed and maintained systems used by Page15 GAO/IMTEG!W34Army CorporateDataBase Chapter2 TheAnmyIs FkrsulngCorporateDataBase Pro&et Objectives more than one of the Army’s major commands. The Fort Sill systems were also developed to permit data sharing at the installation. The export of the Fort Sill systems to other installations was an objective of the canceled Corporate Data Base Project. The objective of the Army-wide Installation Support Modules Project is to install an upgraded version of the Fort Sill systems at all 179 Army installations. Similarly, TRADOC’SInstallation Support Modules effort involves enhancing the existing Fort Sill systems and installing them at its 16 installations. Army Officials responsible for both the Army-wide and TRADOCInstallation Support Modules projects agreed that their ini- tiatives are pursuing the objectives of the canceled Corporate Data Base Project. The third installation-level initiative is being developed at Fort Hood and is called the Integrated Data Base. This project evolved from the Army Corporate Data Base Project, In the September 1986 Corporate Data Base Plan, the Army Corporate Data Base Project Manager desig- nated Fort Hood as the site for the installation-level prototype of the Army Corporate Data Base. After the Army Corporate Data Base Pro- ject was canceled in late 1987, Fort Hood continued developing the pro- totype and changed the name of the project to the Fort Hood Corporate Data Base. In 1988, the project name was changed to the Integrated Data Base. While the project’s scope has been reduced, it is still intended to be the standard data base for the installation, Officials from the installa- tion’s Directorate of Information Management agreed that the current data base project evolved from the canceled project and that the current project is pursuing the similar objectives. Page16 GAO/IMTEG90-64 Army CorporateDataBase Chapter 3 Army Managementof Initiatives Lacks Merits of a Centrally Directed Program The Army’s management of the eight initiatives (which are pursuing objectives similar to those of the canceled project) does not provide assurance that the systems will work together, not duplicate one another, and are based on valid mission needs. Further, the Army is not sure that a cost effective capability will be achieved. For example, we found that the initiatives lack central direction and guidance because the Army has not established a program office, completed its informa- tion architecture, or fully implemented its data standardization pro- gram. Additionally, organizations controlling seven of the eight initiatives have not completed requirements determinations to validate mission needs or economic analyses to ensure that the most cost-effec- tive development approach has been used. The organization controlling the eighth initiative plans to comply with these requirements. Because these organizations have not completed economic analyses, the Army also does not know how much it may cost to develop the initiatives. Fur- ther, the Army does not know how much has been spent on the initia- tives because the responsible organizations have not tracked their costs. Without central project management and an information architecture, Initiatives Lack the Army is not sure that the initiatives will be able to work together Central Direction and not duplicate other systems. Additionally, in the absence of an information architecture, the Army lacks a basis on which to plan a transition from its current to the future information systems environ- ment. The Service also does not know whether the initiatives will be able to exchange data useful to Army decision-makers because it has not fully implemented a data standards program, which is intended to elimi- nate conflicting and duplicative data. When the Army canceled the Corporate Data Base Project, it eliminated the program office that had been established to prepare an overall implementation strategy and oversee integration and coordination. The Corporate Data Base Implementation Strategy Plan stated that without central coordination and control the Army could not adequately inte- grate and manage the project. Although individual organizations are developing aspects of the corporate data base, the Army has not estab- lished a program office to manage and oversee the individual initiatives to ensure that they will work together and not duplicate others. According to Army headquarters officials, a program office was not established because the Service could not reach a consensus that a cen- trally-directed program was the best way to implement a corporate data base. Page17 GAO/lMTEG90-64 Army CorporateDataBase Chapter8 Army Managementof I&iativea Lacks IMerItsof a Centrally-Direct4 Program Army regulations require the development of an information architec- ture to ensure that automated systems will work with and not duplicate others. The information architecture is supposed to be the basic frame of reference for information management decisions and provides the basis for planning the logical transition from the current to the future information systems environment. However, as we reported in June 1990,’ the Army has not developed an Army-wide information architec- ture. We also reported that architecture development had been hindered at all levels because the Army has neither exercised effective manage- ment nor provided adequate guidance to facilitate architecture develop- ment. Without an information architecture, the Army cannot ensure that information systems initiatives will fit into the architectural frame- work once it’s developed. Even with an Army-wide information architecture, meaningful data exchange among the systems supporting the Army’s decision-making process could be jeopardized because the Army has made minimal pro- gress in establishing data standards. In 1986, the Army Audit Agency reported2 that the Service’s lack of progress was attributable to ineffec- tive policy and procedures, and that the data standardization program needed redirection, Although the Army revised its policy on data stand- ardization in September 1989, it does not plan to publish guidance on how to implement the policy until November 1990. To ensure that automated systems meet mission needs at the lowest Initiatives Are Not overall cost, federal and Defense policies require that system develop- Based on Analysis of ment efforts be justified by preliminary analyses. Specifically, federal Need or Consideration information resources management regulations state that the acquisition of new or additional information processing resources shall be based on of Economic Benefit mission needs and supported by a requirements analysis prior to begin- ning system development. Army regulations require all organizations to develop an information architecture to identify all requirements for their information systems. After the requirements analysis is completed, federal regulations require an economic analysis to identify the most cost effective I Information Resources: Army Should Limit New Initiatives Until Management Program Is Imple- mented (GAO--68, June 29,lQQO). 2Army Data Element Standardization Program, Assistant Chief of Staff for Information Management, United States Army Audit Agency, (Report No. SW 86-203, March 28,1986). Page18 GAO/IMTJ3G9O84 Army CorporateDataBase Chapter3 Army Managementof Initiatives Lacks Merit4 of a Centra&Birect45dProgram approach for developing an automated information system. DODInstruc- tion 7041.3, Economic Analysis and Program Evaluation for Resource Management, requires that proposals involving a choice between two or more alternatives include (1) a comparison of the total estimated cost/ benefit of each alternative over its useful life and (2) all of the resources required to meet the stated objectives. Further, the instruction requires periodic comparisons between estimated and actual costs and benefits to determine the cost effectiveness of the system. Army regulations also require that the analysis be updated when assumptions of the original study change. In developing the initiatives, the Army has failed to follow these required acquisition procedures and policies. Specifically, organizations controlling seven of the eight initiatives have not completed require- ments analyses, prepared economic analyses, or tracked their total costs. The organization controlling the eighth initiative project plans to prepare requirements and economic analyses, but does not plan to track all of the initiative’s equipment costs. Officials responsible for the initia- tives generally told us that requirements analyses were not a priority. Further they did not believe that they were required to prepare eco- nomic analyses or track total costs. However, without such analyses, the Army cannot be sure that the initiatives are based on valid mission needs and represent the most cost effective solution or how much it may ultimately cost to develop the initiatives. Moreover, the Army does not know how much has been expended on the projects because it has not tracked their costs. The following table lists the initiatives and notes their compliance with the pertinent information systems policies and regulations. Page19 GAO/IMTEG9044Army CorporateDataBase Chapter3 . Army Managementof Initiatives Lacks Merits of a Centrally-DirectedProgrsm Table 3.1 Project Compliance With Federal and Defense Acquisition Requirements Economic Policies Project analysis analysis Headauatters Decision support systems No No Decision resources data base No No No Army management support network No No No Maior Command TRADOC decision support systems No No No FORSCOM command data base No No Noa - Installation Army-wide installation support modules Yesb Yesb No TRADOC installation support modules No No No Fort Hood intearated data base No No No aAlthough FORSCOM had not tracked total costs, the Command agreed to do so for the remainder of the project. bThe Army-wide Installation Support Modules Project Office plans to complete the requirements and economic analyses prior to beginning system development. Headquarters Initiatives The organizations controlling the three headquarters initiatives had not completed requirements determinations, prepared economic analyses, or tracked their costs. Although headquarters officials recognize that requirements analyses are needed, they told us that they have not been completed because the headquarters information architecture has not been completed. They told us that waiting for its development would have delayed individual systems development. Further, while responsible headquarters officials told us that they did not believe an economic analysis was required for the systems, we did find a September 1988 cost/benefit analysis of the Headquarters Deci- sion Support Systems. The analysis disclosed that headquarters had not prepared an overall economic analysis, and that organizations control- ling the individual systems had not maintained reliable costs or deter- mined whether tangible benefits had accrued. The Headquarters Decision Systems Management Agency agreed that an overall economic analysis was needed and that cost and benefit information on the indi- vidual systems was not available. Additionally, Army headquarters officials did not know how much had been expended on the systems because the initiatives have been funded Page20 GAO/IMTEG90-64 Army CorporateDataBase Chapter3 Army Managementof Initiatives Lack13 MerItaof a Centrally-DirectedProgram by many different organizations and their total costs have not been cen- trally tracked. However, the September 1988 cost/benefit analysis esti- mated that for fiscal years 1988 through 1996 the headquarters decision support systems alone would cost over $600 million. Major Command Initiatives TRADOC and FORSCOMhave not completed requirements analyses or pre- pared economic analyses for their initiatives. Further, neither Command tracked the total costs for their initiatives. TRADOC officials stated that completing its information architecture or requirements analysis was not a priority until recently. In response to Department of Defense Inspector General and Army Audit Agency reports, the Command started to prepare the requirements analysis and plans to complete it in June 1990. TRADOC officials also stated they did not believe they were required to prepare an economic analysis or track actual costs because the effort is a prototype. However, the Command’s decision support systems effort goes well beyond the Army’s definition of a prototype. Army regulations state that a prototype is used to evaluate a design or test aspects of a proposed production system involving high-risk technology. According to National Institute of Standards and Technology officials responsible for the Institute’s guide on prototyping,3 TRADOC is developing a produc- tion system and not a prototype. In their opinion, TRADOC’S prototype involves too many installations and is taking more time than needed to further define requirements or test aspects of a proposed system. Fur- ther, Command officials have not tracked all the costs for the effort because they have omitted some of the personnel and equipment costs needed to implement the proposed system. Similarly, FORSCOM has not completed its requirements analysis, pre- pared an economic analysis, or tracked all the costs for its Command Data Base initiative. Command officials told us that they have not com- pleted their information architecture or requirements analysis because they lack needed headquarters guidance. As we reported in June 1990,4 the Command has done significant work toward developing its architec- ture, but an official stated that they could not afford further investment “NBS Special Publication 500-148, Application Software Prototyping and Fourth Generation Lan- guages, National Bureau of Standards (now called the National Institute of Standards and Tech- nology), May 1987. 41nformation Resources: Army Should Limit New Initiatives IJntil Management Program Is Imple- mented (GAO/IMm90_ _b8 , June 29,lQQO). Page21 GAO/IMTEC-9084 Army CorporateDataBase Chapter3 . Army Managementof Initiativee Lacks Merits of a Centrally-DirectedProgram without the overall architecture to guide them. FORSCOMofficials also told us that they did not believe it was necessary to prepare an economic analysis or track total costs because the initiative was not a major pro- gram and had been justified in the Information Management Plan approval process. However, we determined that Army regulations require the analysis for projects of all sizes. Further, Army regulations do not permit organizations to substitute the Information Management Plan approval process for the analysis. During the course of our review, Command officials decided to prepare an economic analysis and track total costs for the remainder of the effort because they wanted to comply with the pertinent Army policies. The Command had not com- pleted its analysis prior to the conclusion of our review. Installation-level Only one organization controlling the three installation-level initiatives Initiatives plans to comply with regulations for preparing requirements and eco- nomic analysis. Further, none of the organizations have fully complied with Defense policies on cost tracking-two have incomplete costs and the other one has not tracked costs at all. The Army-wide Installation Support Modules Project office has initiated its requirements analysis and plans to conduct an economic analysis prior to system development. Although the Army-wide project office is tracking costs, it does not plan to include all of the costs associated with implementing the system. For example, the Army plans to install the modules at all 179 Army installations, but an undetermined number of installations may require upgrades to their computers to operate them. The Army plans to provide the needed upgrades through other pro- grams and does not plan to identify the costs as part of the Installation Support Modules project. TRADOC has not completed a requirements analysis, prepared an eco- nomic analysis, or fully tracked the costs for its Installation Support Modules initiative. TFtADOCofficials told us that the Command has not completed information architectures or requirements analysis for all of its installations because they were not priorities until recently. The Command plans to complete the analyses in June 1990. Although TRADOC Decision Support Systems officials initially stated that the initiative did not require an economic analysis because it was a part of their prototype effort, they did provide us a 1987 cost/benefit anal- ysis prepared by the Information Systems Command. However, the anal- ysis did not include all equipment costs for the system. For example, Page22 GAO/IMTEC-9084 Army CorporateDataBase . Chapter3 Army Men8gementof Initiativee IAWlcs Merits of a Centrally-DirectedProgram while the Command’s installations have procured new data base man- agement systems and operating systems for the Installation Support Modules system, the 1987 analysis did not address these costs, Addition- ally, the Command did not know the total cost of the system because portions of its development have been funded by many different organi- zations and not centrally tracked. For example, Fort Sill developed a major portion of the system but did not track its costs. Fort Hood has not completed a requirements analysis, prepared an eco- nomic analysis, or tracked total costs for its Integrated Data Base. Fort Hood officials told us that the installation has not completed its infor- mation architecture or requirements analysis because it had not been a priority until recently. Fort Hood officials also said they had not prepared an economic analysis because the type of data base management system being used has been proven in private industry. We do not agree with the installation offi- cials’ assertions because federal and Defense policies for automated information systems require that economic analyses address the activity to be automated-in this case, the installation. The regulations also require organizations to complete the analysis prior to beginning system development and conduct periodic reviews to determine the cost effec- tiveness of the approach selected. Additionally, installation officials stated that they did not track the total cost of the data base develop- ment because portions of it have been funded by the users. Page23 GAO/IMTEC-90-84 Army CorporateDataBase Chapter 4 Conclusionsand Recommendations Without regard for the 1986 congressional and Defense direction, the Conclusions Army is continuing to develop an ad hoc corporate data base. Because this effort is not centrally directed or in compliance with pertinent fed- eral and Defense policies, the Army cannot be sure that the systems it is developing are based on valid mission needs, will work together and not duplicate one another, or that a cost effective capability will be achieved. We identified eight initiatives at the headquarters, major command, and installation levels that are intended to provide capabilities planned for the canceled project. The headquarters-level initiatives involve the development of a standard data base, an unknown number of decision support systems, and a long-distance communications system. The ini- tiatives at the major command and installation levels primarily involve the development of standard data bases. Our work was limited to initia- tives at Army headquarters, two major commands and selected installa- tions; other organizations also may have projects pursuing the objectives of the canceled project. We also found that the initiatives lack the merits of a centrally-directed program. Specifically, we found that the initiatives lack central direc- tion and guidance because the Army has not established a program office, completed its information architecture, or fully implemented its data standardization program. Without a program office or information architecture, the Army is not sure that the initiatives will work with and not duplicate others. The Army also is not sure that the initiatives will be able to exchange data useful to Army decision-makers because it has not fully implemented a data standards program, which is intended to eliminate conflicting and inaccurate data. Additionally, although federal and Defense policies for automated infor- mation systems call for formal requirements determinations and eco- nomic analyses prior to beginning system development, organizations controlling seven of the eight initiatives had not complied with these policies. The organization responsible for the eighth initiative plans to comply. Therefore, the Army lacks assurance that the initiatives are based on valid requirements. The Service also does not know whether the most economical approach has been selected or how much it may cost to develop the initiatives. In addition, the Army did not know how much has been spent on the initiatives because the organizations con- trolling them have not fully tracked costs. Page24 GAO/LMTEG90+34 Army CorporateDataBase . Chapter4 ConcludonfsandRecommendations We recommend that the Secretary of Defense direct the Secretary of the Recommendations Army to suspend development activity on the eight initiatives we identi- fied, and not to resume any of them unless all required federal and Defense acquisition requirements are met. This would include preparing mission needs statements and requirements and economic analyses. In addition, we recommend that the Secretary of Defense direct the Sec- retary of the Army to prohibit funding for any other initiative the pur- pose of which is to achieve a corporate data base capability, until the Army adequately justifies to the Office of the Secretary of Defense and the Congress the mission need and potential economic benefit of an Army-wide corporate data base. At the Chairman’s request, we did not obtain official agency comments Agency Comments on a draft of this report. However, we did discuss the results of this review with agency officials and they agreed with the facts presented. Page26 GAO/IMTEG9084Army CorporateDataBase Appendix Major Contributors to This Report Thomas J. Howard, Assistant Director Information Wiley E. Poindexter, Evaluator-in-Charge Management and M. Scott Laemmle, Evaluator Technology Division, Washington, DC. Carl L. Higginbotham, Regional Assignment Manager Atlanta Regional Christopher T. Brannon, Evaluator Office Joseph J. Watkins, Regional Management Representative Norfolk Regional Suzanne K. Wren, Evaluator Office (510470) Page26 GAO/lMTEGQO44 Army CorporateDataBase -- --
Information Resources: Army Corporate Data Base Disregards Congressional and DOD Direction
Published by the Government Accountability Office on 1990-07-19.
Below is a raw (and likely hideous) rendition of the original report. (PDF)