Inspection Report I-97-05

APPENDIX II - INVENTORY OF JPATS AIRCRAFT - NOT INCLUDED IN THIS HYPERTEXT VERSION.

APPENDIX III - UNITED STATES MARSHALS SERVICE RESPONSE TO DRAFT REPORT - NOT INCLUDED IN THIS HYPERTEXT VERSION.

APPENDIX IV - OFFICE OF THE INSPECTOR GENERAL'S ANALYSIS OF MANAGEMENT'S RESPONSE - NOT INCLUDED IN THIS HYPERTEXT VERSION.

SUBJECT: Inspection of the Justice Prisoner and Alien Transportation System, Report Number I-97-05

In your first status report to the Attorney General on the Justice Prisoner and Alien Transportation System (JPATS), dated October 6, 1996, you wrote that:

The Inspections Division has completed a review of JPATS implementation, and based upon this review, we can report that JPATS is moving increasing numbers of prisoners and aliens. However, we have been unable to draw conclusions about the efficiency of JPATS operations because of the lack of usable operational and cost data related to the movement of prisoners and aliens. This lack of adequate data has been a long-standing problem in the USMS.

USMS efforts are currently underway to implement an Automated Prisoner Scheduling System (APSS), designed to capture operational information needed to make future JPATS management decisions (e.g., about aircraft, routes, and hubs). However, as designed, the APSS will not capture all the costs incurred by the Department components involved in transporting prisoners and aliens. Although capturing complete cost information presents a challenge, it must be done if JPATS management is to have the information it needs to make cost-effective use of JPATS resources. We therefore recommend that JPATS management and the Chief Information Officer (CIO) work together to develop a systematic approach for collecting

and analyzing USMS, Immigration and Naturalization Service (INS), BOP, and Federal Bureau of Investigation (FBI) prisoner and alien transportation cost information needed to make cost-effective decisions about the use of JPATS resources.

Complicating efforts to coordinate data collection and analysis has been poor communications between the CIO (developer of APSS) and JPATS users, especially the primary user, the USMS Prisoner Services Division (PSD). When the CIO assumed responsibility for APSS development in August 1995, his first action was to select a new contractor. He did so because he had serious reservations about the performance and design phase proposal submitted by the requirements phase contractor, and because he believed the system could be developed at a lower cost than the requirements phase contractor was projecting. The CIO subsequently made changes in approach and time frames for implementing the APSS that were not effectively communicated to PSD managers at headquarters and the Kansas City scheduling center.

We reviewed the actions taken by the CIO and, while it is too early to fully assess the merits of the CIO's decisions, the poor communication between the CIO and PSD puts the APSS project at risk. PSD managers are responsible for the movement of prisoners and aliens, and ultimately for APSS. They must clearly understand and be kept apprised of changes in approach and implementation time frames as these may affect future operational decisions and resource requirements. Accordingly, three of the five recommendations in our report are intended to require more effective communication and coordination.

We also note that computer security requirements have not been defined for the APSS system. We found that well into the design phase the USMS has not yet performed a systematic review of the security requirements for the APSS, and user (USMS, INS and BOP) security requirements have not been established. Department of Justice policy requires that security specifications be defined prior to initiating formal development of application systems, and that appropriate technical, administrative, physical, and personnel security requirements be included in procurement specifications for the acquisition or operation of automated information systems.

Given that security requirements were not defined, we recommend that the CIO promptly initiate a formal review of APSS computer security requirements. Specifically, steps should be taken to identify APSS security risks and appropriate security measures to address those risks. Specific security safeguards and controls should, at a minimum, be designed and implemented to prevent and detect unauthorized system access and access to data, and misuse of the system by authorized users. In addition,

contingency planning is needed to ensure that appropriate arrangements are made for backup systems and data recovery.

In summary, we found that JPATS is moving increasing numbers of prisoners and aliens, but that more needs to be done to provide JPATS management with the data it needs to make the most efficient use of JPATS resources. We hope the comments, suggestions and recommendations contained in the attached report will be useful in your efforts to improve and enhance JPATS operations.

We sent copies of the draft report to your office February 3, 1997, and requested written comments on the findings and recommendations. Your March 12, 1997 response addressed each of the five recommendations. We have attached your response as Appendix III.

On the basis of your written comments, we consider all recommendations resolved but have kept two of them open pending further action. Appendix IV explains why the recommendations were not closed and what action is needed.

Please respond to the open recommendations by June 17, 1997. Your response should provide the additional information requested. If actions have not been completed, please provide proposed completion dates. Guidance on report follow-up and resolution can be found in Department of Justice Order 2900.10.

We congratulate you and the JPATS staff for your recent achievement in receiving the Vice President's Hammer Award, and appreciate the cooperation we received from managers and staff throughout the USMS as we carried out our review. Please let me know if we can provide you with any additional information. I can be reached on (202) 514-3435.

Transporting prisoners and aliens is a rapidly growing activity in the Department of Justice (Department). The United States Marshals Service (USMS), Prisoner Services Division (PSD), is responsible for managing the Justice Prisoner and Alien Transportation System (JPATS) and is being challenged to schedule and move the growing numbers of prisoners and aliens. The existing system for scheduling prisoner and alien movements is antiquated. The USMS is attempting to put in place the infrastructure needed to automate the scheduling system and provide for the much needed analysis of transportation workloads and resource requirements. In the interim, Department managers are forced to make multi-million dollar decisions regarding aircraft and detention facilities without the benefit of reliable historical information and analysis of air and ground operations. As a result, current spending and decision making may not best support future JPATS requirements.

JPATS conducted 189,071 prisoner and alien movements for Fiscal Year (FY) 1996, and this number is expected to grow to approximately 221,000 in FY 1998. A movement is defined as each scheduled segment of a prisoner trip. Complete end-to-end trips may include more than one segment. Using its growing fleet of aircraft, and with buses owned and operated by the Bureau of Prisons (BOP) and the Immigration and Naturalization Service (INS), JPATS is moving increasing numbers of pretrial and sentenced prisoners and illegal aliens. JPATS also assists state and local law enforcement agencies by transporting non-Federal prisoners. This is consistent with the President's commitment to provide Federal assistance to local law enforcement agencies whenever possible. In FY 1996, JPATS transported 5,286 state and local prisoners.

Since its inception, the USMS has been responsible for detaining and delivering pretrial prisoners to courts and detention facilities. Added to those responsibilities is the movement of post-conviction (sentenced) prisoners who are in the custody of the BOP. As the pretrial and sentenced prisoner populations increased, the need to have a more centralized scheduling process became necessary. In 1979, the USMS created the National Prisoner Transportation System (NPTS). The NPTS ended the practice of the 94 districts independently making long distance travel arrangements for prisoners. Located in Kansas City, Missouri, the NPTS scheduled and coordinated air and ground transportation for prisoners nationwide.

Rather than relying solely on commercial flights to move prisoners over long distances, the USMS established its own air transport capability in 1985 with the purchase of a Boeing 727 (B-727) aircraft, capable of moving about 100 prisoners at a time. The hub of the newly created Air Operations Division was located at the airport in Oklahoma City, Oklahoma.

The INS transportation system had a parallel evolution. Initial transportation methods to move aliens primarily involved the use of vans and buses that were independently operated in each of INS' regions. Through the years, increasing numbers of aliens required detention and removal over longer distances, giving rise to the need for air transportation. INS air operations began in 1954. The Air Transport Branch (ATB) mission has through the years focused primarily on the removal of criminal aliens. It also arranged for alien transfers, deportation and removal flights, and relocating aliens to lower-cost detention facilities until their removal. ATB air resources were limited to a few small aircraft and one medium sized airplane capable of moving about 50 aliens at a time. In 1987, the INS moved ATB operations to Pineville, Louisiana. The ATB is now part of JPATS operations as a result of the merger in October 1995.

Through the years the USMS and INS obtained additional aircraft and changed the mix of their aircraft. The two organizations did not routinely coordinate prisoner and alien movements or share transportation resources to effect these moves. The result was under utilization of aircraft capacity, especially by INS on return trips from the southwest border. Alien transportation movements, by the very nature of the deportation process, tend to be one-way, while incarcerated inmate and pretrial detainee transportation movements tend to be round-trip in nature. Therefore, the Department maintained two separate air fleets to transport prisoners and aliens. This duplicative air transport capability strongly suggested cost and administrative inefficiencies, especially because one of the fleets was flying empty half the time.

Previous Studies Cited Lack of Basic Operational Data To Support Prisoner Transportation Management Decisions

As early as 1989, the Audit Division, Office of the Inspector General, highlighted the fact that the USMS did not have an adequate system to record and report NPTS program costs. The October 1989 audit report stated that the USMS needed to improve the management of the NPTS for directing, controlling, recording, and reviewing its operation. Specifically, the NPTS only reported selected costs and omitted those incurred by the service providers. Because costs were understated, the NPTS presented an inaccurate picture of prisoner movement costs. In addition, the failure to present total costs made it difficult to identify areas where economy and efficiency could be achieved.

The John A. Volpe National Transportation Systems Center, Department of Transportation, conducted a study of the NPTS at the request of the USMS. In its March 1992 report, "National Prisoner Transportation System: Preliminary Assessment," the Volpe Center concluded that the paper-based, manual USMS prisoner scheduling system used by the NPTS operated effectively and reliably, i.e., scheduling coordinators were able to keep up with the scheduling and routing of aircraft and passengers. However, the report also noted that because the prisoner scheduling system was not automated, management did not have access to the operational and cost information needed to support decision making about the use of JPATS resources. The report stated that the most basic information, such as patterns of movements and unit costs of movement, was not readily accessible or available. The study also pointed out that it was not possible to determine whether the NPTS was running efficiently, and that the needs for additional or different types of aircraft, staffing levels, facilities, and vehicles could not be adequately justified.

The Volpe Center conducted another study in 1993 on the possible merger of USMS and INS transportation services. The March 1993 report, "Feasibility of Consolidating INS and USMS Transportation Services," stated that: "Neither the INS nor the USMS collect data or conduct analyses that would permit the evaluation of the cost-effectiveness of their services." Without information on the patterns of origins and destinations of their passengers, the report stated that it was ". . . impossible to judge performance or determine capital equipment requirements." To illustrate this point, the Volpe report provided examples of questions that could not be addressed without data analysis, including:

The 1993 Volpe report concluded that "substantial cost savings in prisoner transportation are almost certainly possible, but . . . , analyzed evaluation of the cost-effectiveness of INS and USMS prisoner transportation is impossible at present because usable data are not collected or tabulated." The report went on to note that whether or not the INS and USMS air assets were merged, the potential savings "cannot be achieved without improved management and planning functions."

The Justice Management Division (JMD) report, "A Management Review of Air Fleet Operations within the Department of Justice," issued in May 1994, recommended that the Attorney General consolidate INS and USMS air operations. JMD recognized and confirmed the weaknesses pointed out in the previous Volpe reports, but stated that the merger would nonetheless generate savings due to economies of scale, reduce costs for commercial flights, and avoid duplicative investments in aircraft upgrades. The Deputy Attorney General decided in August 1994 to consolidate the air transport operations of the USMS and INS.

The merger of USMS and INS air operations to form JPATS began in August 1994, and became operational as of October 1, 1995. The objective of the merger was to take advantage of centralized management, scheduling, maintenance, and associated economies of scale for both the USMS and INS. Combining the two operations was expected to result in increased passenger loads and lower per passenger costs. Responsibility for management of JPATS was delegated to the USMS. During the transition period of the merger, the NPTS assisted in the aircraft movements of aliens.

The JPATS implementation strategy stated that for the long-distance moves the INS would select cities along existing USMS air routes where aliens could be picked up and discharged. INS would use existing ground transportation resources to transport criminal aliens to these cities to be placed on the larger JPATS B-727s for transport. If required, criminal aliens as well as USMS and BOP prisoners could be temporarily housed at the 1,800-bed BOP Federal Transfer Center located at the Oklahoma City Airport to await another flight to their final destination.

The Federal Transfer Center is integral to the success of JPATS because it provides a hold-over facility for prisoners and criminal aliens at the same location where the larger JPATS aircraft are based. The two B-727s currently in the JPATS inventory each operate about 4-5 days each week and make trips to the northeast, northwest, southwest, and southeast areas of the country. These B-727s are used to travel the longer routes and return to the Oklahoma City Airport each evening. Also based at the USMS facility at the Oklahoma City Airport are four Sabreliners that carry only six to seven prisoners at a time. Because of their limited passenger capacity, the Sabreliners are primarily used to move prisoners in time-critical situations or emergencies.

JPATS also keeps some of its aircraft (one DC-9 and two Convair-580s), some of which were obtained from the merger with INS, at its satellite aircraft facilities located in El Paso, Texas, and Alexandria, Louisiana. These aircraft are used primarily to shuttle criminal aliens from the Federal Transfer Center to the BOP criminal alien facilities at Oakdale, Louisiana, and Eloy, Arizona, or to assist in the transportation of noncriminal aliens.

Although JPATS owns and operates aircraft that it can route and use as needed, it does not have direct control over all the transportation assets it uses to transport prisoners and aliens. Specifically JPATS:

The Federal Bureau of Investigation (FBI) also furnishes a 35-passenger aircraft and pilots on a reimbursable basis. This FBI assistance helps JPATS transport prisoners and criminal aliens in the under-serviced Mid-Atlantic/Northeast corridor.

The number of prisoners and aliens needing to be moved continues to grow each year. The Federal inmate population grew from an average of 55,407 in FY 1991 to 92,157 in FY 1996. The number of BOP facilities has grown from 65 as of September 30, 1990, to 86 as of May 31, 1996. Likewise, the illegal alien population continues to rise with substantial increases in the numbers of criminal aliens in Federal custody. Non-citizens in Federal prisons increased from 4,088 in 1984 to 18,929 in1994, an increase of 363 percent in 10 years, and INS is devoting more and more resources to its efforts to apprehend individuals in the United States illegally.

In its first year, JPATS moved more prisoners and aliens than had been moved in the previous four years. As the following chart shows, the total number of JPATS prisoner and alien movements for FY 1996 was 189,071. This FY 1996 total represents at least 24,000 more movements than was accomplished in any of the previous four years. Almost all of this increase was due to the large number of illegal aliens moved. In FY 1996, JPATS moved three times as many aliens as INS and USMS together moved in FY 1995. JPATS management expects the growth in prisoner and alien movements to continue increasing at least through FY 1998.

	FY 1992	FY 1993	FY 1994	FY 1995	FY 1996
USMS Moves	157,089	160,046	152,240	151,387
INS Air Moves	3,554	5,017	6,380	10,334*
JPATS Moves USMS INS					157,843 31,228
TOTAL	160,643	165,063	158,620	161,721	189,071

* April - September 1995 served as a transition period for JPATS. USMS and INS began merging their air transportation services and assets during this time. Many of these movements were carried out jointly by the USMS and INS.

JPATS MANAGEMENT MUST MAKE IMPORTANT RESOURCE DECISIONS WITH LIMITED INFORMATION

Lacking an automated system to capture and analyze operational and cost data, JPATS management must make important resource decisions based upon limited information. Prisoner movement data currently exists only on tens of thousands of individual movement request forms. Under current conditions, it would be an onerous task to assess whether JPATS is making efficient use of its resources. As JPATS grows and evolves, management will need operational and cost data to support decisions such as:

JPATS needs to demonstrate that transportation routes are optimized to ensure that air and ground resources are used in the most efficient manner, i.e., not just scheduling passengers on prearranged modes of transportation and routes. In its March 1992 report, Volpe presented a sample of prisoner movements that represented what appeared to be inefficient routing. One movement cited in the report was from the Federal Correctional Institution at Milan, Michigan, to the Metropolitan Correctional Center in New York City (about 600 miles). A prisoner was picked up by van at Milan and transported to Detroit for air transport. After flying the prisoner to the Federal Correctional Institution at El Reno, Oklahoma (used prior to the opening of the Federal Transfer Center), the prisoner was confined there for 6 nights, and then was put on another flight to Atlanta. The prisoner stayed overnight in the Atlanta Penitentiary, and the next day was airlifted to the Federal Correctional Institution at Otisville, New York. A USMS van picked him up in Otisville, completing the final segment of the trip to the Metropolitan Correctional Center. The complete trip covered about 2,650 miles over 8 days.

On the surface, it seems that an alternate form of transportation, such as bus or van, could have been a more economical way to move this prisoner. However, the lack of complete operational and cost data makes it difficult to establish whether this was the best way of moving the prisoner between these two points. Complicating the analysis is the fact that in most cases, the time it takes to move prisoners is not considered an important issue, except for those in an expedited or priority status, such as prisoners mandated to appear in court. The following map illustrates the circuitous route taken for this movement.

In March 1996, a JPATS report, "Analysis of the JPATS Air Program," included what appeared to be examples of inefficient use of air resources. For example, a B-727 flight on October 18, 1995, was to transport 47 prisoners from Kansas City to the Federal Transfer Center (about 325 miles). Although a direct flight would have taken about 1 hour, the aircraft made numerous stops along its scheduled route to the northeast corridor and the 47 prisoners spent 6 1/2 hours aboard the aircraft before arriving at the Federal Transfer Center. The following map reflects the much longer route taken by these prisoners.

A second example was a flight on November 15, 1995, on which 36 prisoners picked up at the Federal Transfer Center were destined for the Federal Correctional Institution at Oakdale (about 450 miles). Instead of a 1-hour flight, they spent 6 hours and 20 minutes on the aircraft making stops along its scheduled route to the northeast corridor before arriving in Oakdale. This circuitous trip is shown in the following map.

The lack of operational data for analysis means that JPATS will continue to route prisoners and aliens as previously described and not know whether those movements have been completed in the most efficient and economical manner. While speed, security, and safety are prime reasons for transporting prisoners and aliens by aircraft, JPATS still needs to ensure that they are transported in the most cost-effective manner. Collecting sufficient data, and then analyzing it, is a necessary first step in meeting this challenge. Merely analyzing one prisoner and alien movement for cost-effectiveness will not provide a sound basis for establishing a complete routing system for aircraft, buses, and vans.

Future JPATS plans include a second, and possibly a third hub (in addition to the USMS Air Operations Center in Oklahoma City) for aircraft operations. According to the Chief of JPATS scheduling operations, use by JPATS of a portion of an existing 300-bed BOP facility in Atlanta is being negotiated. The facility would serve as a holding facility comparable to the Federal Transfer Center. In addition, Seattle, Washington, is being considered as a western hub for JPATS because of a 500-bed BOP facility nearing completion.

At present, JPATS management lacks the necessary data to identify the most cost-effective or logistically-sound locations for new hubs. The initial benefit of using BOP facilities in Atlanta or Seattle could be offset by other factors not fully examined or known, e.g., the majority of prisoner and alien movements may originate/end in a different city. An in-depth analysis of air routes and patterns of prisoner movements needs to be conducted to fully justify new hub selections.

Additional hubs could also impact the usage of the Federal Transfer Center. This 1,800-bed facility was primarily established because of JPATS centralized air operations in Oklahoma City. With more hubs and decentralized air operations, the Federal Transfer Center may not be fully utilized. In addition, this facility was built as a transfer facility and not as a long-term detention facility as are other BOP institutions.

If additional hubs are selected and become operational, typical flight distances could be significantly reduced. The current inventory of aircraft would then have to be re-evaluated, and this could bring into question the need for the transcontinental capability of the B-727s currently in use. In this scenario, use of intermediate size aircraft instead of B-727s for JPATS flights might be more advantageous.

Despite this possibility, the USMS said they recently had to purchase a third B-727 through the Working Capital Fund to keep up with the growing workload. However, if two additional hubs become operational in the near future, the need for this recently purchased third B-727 and its capabilities could be diminished.

AUTOMATED PRISONER SCHEDULING: AN IMPORTANT STEP TO IMPROVE JPATS DECISION MAKING

The scheduling center in Kansas City coordinates and schedules all prisoner movements through a labor-intensive, manual process. Currently, there are 18 coordinators directly responsible for scheduling all air and ground movements. Ground movements are processed by coordinators assigned to six geographic areas, whereas air movements are processed by coordinators assigned to either the smaller (e.g., Sabreliner) or larger (e.g., B-727) aircraft. Consequently, a long-distance move across the entire country often involves multiple coordinators processing the prisoner's ground and air movements.

Each day the scheduling center receives about 330 requests for prisoner transportation. The INS, the BOP, and the USMS district offices are the originators of these requests, and Kansas City receives these requests through a variety of methods including facsimile, teletype, telephone, or e-mail. USMS district offices submit requests using a USMS Form 106. The BOP submits these requests in a format similar to a Form 106 that is readily usable by the coordinators. The INS requests, however, have to be transcribed into the Form 106 format; a time-consuming process if multiple moves are involved. All requests for prisoner and alien movements are then processed and routed physically through the scheduling process (ground and air coordinators) until the prisoner or alien reaches the final destination.

Capturing and analyzing data cannot be accomplished effectively because of the manner in which the original information is received, processed, and stored in the scheduling center. First, the process for scheduling and recording data by the coordinators in Kansas City is not automated. It is an entirely manual process in which the request for movement and associated data is initially received from the originating source in a variety of ways (facsimile, e-mail, telephone, or teletype). The request is then either used in its original format (a hard copy Form 106) or transcribed into a usable Form 106 format. The Form 106 is physically handled by each coordinator involved with the trip. The Form 106, with movement data recorded on it in a variety of ways by the various coordinators, is filed and retained for two years. Tens of thousands of these Form 106's are filed at the scheduling center in Kansas City. Due to past and current staffing levels at the scheduling center and the volume of work involved in this labor-intensive scheduling process, analysis of the data has not been part of the coordinators' daily operational procedures. Therefore, comprehensive analysis of operational trends and patterns has not been accomplished.

Much of the data entered on the Forms 106 by the coordinators is not standard because of special circumstances involved with specific movements. Scheduling prisoner movements requires flexibility and innovation on the part of the coordinators. Coordinators are continually confronted with movement priorities and special situations. Prisoners in an expedited or priority status must be accorded special consideration. Non-routine arrangements must also be made in many medical and security situations. Coordinator verification of certain special requests and seeking corrections concerning clerical errors or absence of necessary data can also disrupt the process.

Evidence of the special circumstances encountered by the coordinators may not always be recorded on the Form 106 because there are no requirements as to what coordinators are to record on it. This practice allows each coordinator to have a unique method of deciding and annotating what they see as necessary, required, or helpful data on the Form 106. This current situation does not, in our opinion, provide for the gathering and analyzing of reliable and usable data on which to assess the efficiency or effectiveness of this operation.

The USMS has tried unsuccessfully since 1983 to automate its prisoner transportation request system in Kansas City. According to JPATS managers, some of the reasons for the unsuccessful attempts were insufficient funds, a change in management support for the project, and a system that did not meet the objectives of the users. The 1992 Volpe study pointed out that information needed for analysis was not consistently available for each movement request (i.e., origin-destination information, actual routes and schedules, sources of movements, and preferred schedules). The study stressed that collecting such data daily and reviewing it frequently would greatly enhance management's ability to forecast trends and make appropriate adjustments.

The Automated Prisoner Scheduling System (APSS) is currently being developed by the USMS to automate many of the manual, time-consuming scheduling processes in Kansas City and to capture the operational data needed to support analysis of JPATS operations. The APSS is an important step in streamlining JPATS operations and allowing managers to make cost-effective use of JPATS resources.

Because the APSS represents an important step toward improved JPATS management decision making, we reviewed the ongoing APSS project to assess its progress to date. The project is being developed in three phases: requirements, design, and implementation. It is currently entering the implementation phase. Based upon our discussions with USMS and other Department officials, and our review of APSS progress to date, we have several concerns about current APSS systems development activities.

In August 1995, the USMS appointed an individual to a newly created Chief Information Officer (CIO) position. The CIO was responsible for the development of all information systems within the USMS. Responsibility for APSS development was subsequently moved by the USMS Director from the Assistant Director, PSD, to the CIO. This change in responsibility occurred during the requirements phase of APSS. One of the first actions by the CIO was to select a different contractor for the design phase than JPATS program managers had wanted. The CIO had serious reservations about the performance and proposal of the requirements phase contractor. The CIO also wanted more competition among contractors during the life cycle of the project. Although the CIO's decision to select a new contractor was within his authority, it created apprehension among the users responsible for the JPATS program (PSD managers and staff) about whether the complete system, as defined by the previous contractor, would be operational within established time frames.

The CIO subsequently made changes in the approach and time frames for implementing the APSS that were not effectively communicated to all levels of PSD management. Specifically, the CIO stated that given the history of failed attempts to implement an automated prisoner scheduling system, APSS development would be carried out incrementally, i.e., USMS requirements for an automated scheduling system would be addressed first and the needs of other JPATS users would be addressed later. While the CIO's incremental approach delayed the completion date of the project beyond that projected by the original contractor, the CIO wanted to ensure that each phase produced a complete and usable product before moving to the next phase. In our opinion, there is no clear evidence as to whether the original contractor's target date for completion was realistic. The lack of communication to the primary users about these changes led to differences in their expectations as to when the complete system would be operational and what would be the capability of the system.

Dealing with the complexity of the system and the many Department users means that future development and implementation of an APSS will require extensive coordination, not only within the USMS, but also with INS and BOP. INS and BOP users of JPATS have not been brought fully into the development process. We discussed this lack of participation by INS and BOP with the CIO. He explained that coordination with INS and BOP had not been a priority early on in the process. Preliminary meetings have since taken place, but INS and BOP requirements have not yet been fully defined.

Success of the system depends on continual involvement by all those responsible for transporting prisoners and aliens. We are therefore recommending that the Assistant Director, PSD, promptly convene a senior-level JPATS user steering committee to ensure senior-level management involvement in all aspects of APSS development. Senior managers from all JPATS user organizations (i.e., USMS districts, INS, and BOP) should be included. We also recommend that the CIO provide detailed monthly status reports to all steering committee members.

Because coordination and involvement between developers and users is critical to the success of the program, we further recommend that the Assistant Director, PSD, select and appoint an individual to serve as a liaison between PSD and the CIO. The individual serving in this liaison position should be assigned to work directly with the CIO and his staff to ensure that JPATS user requirements are fully defined and communicated to the APSS developers.

The USMS has not yet performed a systematic review of the security requirements for the APSS. Well into the design phase, we found that user (USMS, INS and BOP) security requirements had not been fully explored nor documented. According to Department Order 2830.1D, Automated Information Systems Policies:

Given that this did not occur, we recommend the USMS begin a formal review of JPATS security requirements. Specifically, steps need to be taken to identify APSS security risks, and appropriate security measures to address those risks. Specific security safeguards and controls should at a minimum be designed and implemented to prevent and detect unauthorized system access and access to data, and misuse of the system by authorized users. In addition, contingency planning is needed to ensure that appropriate arrangements are made for systems redundancy and data back-up and recovery.

We also noted that while the APSS should, as designed, capture operational information needed to make future management decisions (e.g., about aircraft, routes, hubs, etc.), the current system has not been designed to capture or integrate the wide range of cost information available in a variety of forms and locations that is needed to ensure that resources are allocated and utilized efficiently. Specifically, the system is not currently designed to capture all costs incurred by the Department components involved in transporting prisoners and aliens. As explained in a September 18, 1996 memorandum from the CIO to members of the Marshal's Advisory Subcommittee on Technology:

We therefore recommend that the Assistant Director, PSD, and the CIO work together to identify and plan for the capture of USMS, INS, BOP, and FBI prisoner and alien transportation cost data needed to make the most cost-effective use of JPATS resources.

NEXT STEPS: WORK PROCESS REDESIGN AND INTEGRATION OF APSS WITH DEPARTMENT PRISONER AND ALIEN TRACKING SYSTEMS

The APSS, when completed, should provide the much needed automation of the transportation scheduling operation in Kansas City. However, the approach being used in developing this system is to automate the existing process used by coordinators in Kansas City to schedule the movement of prisoners and aliens. Work process redesign that will determine how coordinators should best perform their tasks in an automated environment is not contemplated at present. We believe work process redesign could provide additional productivity improvements in the scheduling process.

In addition, initial implementation of the APSS may not include integration of existing USMS, INS, and BOP tracking systems to make the scheduling process more efficient. The current scheduling system requires some duplicative and wasteful re-entry of information (i.e., re-entry of Form 106 information for all INS movement requests) and will not provide all USMS users, or any INS or BOP users, with the capability to directly query the system to confirm flight manifests, prisoner schedules, or completed prisoner movements.

To fully support the JPATS mission, we believe the APSS must be integrated with existing USMS, INS, and BOP prisoner and alien tracking systems. Each of these separate systems is undergoing changes, but these changes have not as yet been effectively coordinated so as to facilitate the fully electronic submission of movement requests, or ad hoc user query capabilities. Until these systems are integrated, JPATS will not operate as efficiently or effectively as it could.

Automation is critical to the long-range success of JPATS operations. Successful implementation of the APSS, and subsequent JPATS automation efforts, will require the full cooperation and efforts of the CIO and the Assistant Director, PSD. We therefore recommend that the Director, United States Marshals Service, direct: