I United States General Accounting Office Report to Congressidnal Requesters &A0 September 1990 . AIR TRAFFIC CONTROL The Interim Support Plan Does Not Meet FAA’s Needs United States General Accounting Office Washington, D.C. 20548 Resources, Community, and Economic Development Division B-240352 September l&l990 The Honorable Frank R. Lautenberg Chairman, Subcommittee on Transportation and Related Agencies Committee on Appropriations United States Senate The Honorable William Le man Chairman, Subcommittee on‘\, ransportation and Related Agencies Committee on Appropriations House of Representatives As requested in your letter of December 15,1989, we evaluated the planning and analysis supporting the Federal Aviation Administration’s (FAA) Interim Support Plan (ISP). More specifically, we (1) determined whether FAA adequately identified its requirements when developing the ISP and (2) assessed whether the program is progressing in a manner that reflects its stated urgent nature. FAA developed the ISPin 1987 to sustain existing air traffic control (ATC) equipment and increase computer capacity, primarily at terminal radar approach control facilities.’ The program was intended to bridge the gap between current and future ATC systems caused by delays in long-term modernization programs. The ISPis a conglomeration of 15 projects that ‘, will upgrade or provide additional air traffic controller positions, air- port radars, weather sensors, landing systems, and communications equipment. It has an estimated cost of about $416 million. The ISP projects are described in appendix I. FAA inadequately identified its requirements for the ISP.FAA did not con- Results in Brief duct a requirements analysis as called for by federal regulations and its own procedures. Moreover, FAA did not complete its assessment and approval process for the BP until over a year after seeking funding from the Congress. FAA now views the ISPas insufficient to meet its interim needs and is initiating programs costing at least another $126 million to further expand capacity. ’ Terminalradarapproachcontrolfacilitiesarelocatedat or nearairportsto directaircraftarrivals anddeparturesinto andoutof thejurisdictionof theairportcontmltowers. Page1 GAO/RCRDW213 Interim Support Plan B-240362 be determined through a requirements analysis. The analysis is to include the present and projected work load and a performance evalua- tion of the currently installed automated data processing systems. These regulations apply to the most costly ISPprojects, which upgrade the information processing equipment for air traffic controllers. FAAnow intends to supplement the ISPwith programs that provide addi- FAA Did Not tional interim capacity. We believe FAA might not have underestimated Adequately Plan and its capacity needs if it had adequately analyzed its requirements when Assessthe ISP designing the ISPand assessed it in a timely and responsive manner. FAA Did Not Adequately As a large expenditure of Airport and Airway Trust Funds, the ISP Analyze Its Requirements deserves a strong justification.” Such a justification should have been provided by a requirements analysis, as mandated by the Federal Infor- mation Resources Management Regulation and configuration manage- ment. A requirements analysis should assess the performance of current systems. For the ISP, we believe data such as current capacity usage levels, equipment outages, and spare parts shortages would have been appropriate indicators of performance. A requirements analysis should also assess the present and projected work load. We believe that a com- parison of current and projected capacity, reliability, and maintain- ability levels would have accomplished this assessment for the ISP. However, FAA did not perform such an analysis in designing the ISPor before requesting funds for the ISPfrom the Congress. In July 1987, FAA commissioned a working group of its headquarters, regional, and con- tractor staff to devise a plan to meet FAA’s interim needs. Teams on capacity, reliability and maintainability, and air-to-ground equipment : were established. The teams defined problems, received briefings from FAA and contractor staff, surveyed the regions, and identified projects they believed were necessary. The teams then reconvened, combined their various projects, and proposed them as the ISP.Instead of devel- oping projects on the basis of a rigorous requirements analysis, the working group based the projects primarily on the general knowledge of experienced regional and headquarters staff. According to the working “Establishedin 1970,theAirport andAirwayTrustFundfinancescapitalandothercostsassociated with theATCsystemfacilitiesandequipment.TrustFundrevenuesaregenerated mostlyby an8 percenttax onairlinetickets Page 3 GAO/‘EcED9o213 Interim Support plan B-340362 requirements for the Voice Switching and Control, Advanced Automa- tion, Automated Weather Observing, Microwave Landing, and Air Route Surveillance Radar Systems.7 FAA Did Not Assessthe The working group finished developing the ISPin August 1987. Support ISP in a Timely or and approval of the program were soon gained through briefings to FAA’SAssociate Administrator for Development and Logistics, the Asso- ResponsiveManner ciate Administrator for Air Traffic, and the Department of Transporta- tion’s Assistant Secretary for Budget and Programs. Such high-level approval allowed the ISPto be included in the fiscal year 1989 budget submission to the Congress in January 1988. FAA officials wanted to include the ISPin the first budget possible because they considered it urgent. In June 1988, FAA added the ISPto the NASPlan, and in Sep- tember 1988, the Congress appropriated the total $150 million that was requested for the first year. Only after these events did FAA begin activities associated with assessing and approving the program through the configuration management pro- cess. For example, in March 1989, after including the ISPin its budget for the second year, FAA started the required assessment and approval pro- cess for the ISPto ensure that it was technically sound and properly coordinated with other programs, This process involved developing an ISPproposal and having the 32 relevant FAA offices and the SEICevaluate and comment on the proposal. The ISPwas approved through the process 3 months later, in June 1989, almost 2 years after the program was developed, and 9 months after it was initially funded. FAA’s guidance for conducting the formal assessment and approval pro- cess does not clearly state when it should take place in relationship to a funding request for a project. The FAA official responsible for developing the guidance believes it is preferable for projects to be formally assessed and approved before being included in the budget. We agree because the process provides a thorough review by key officials at FAA and the SEIC. As it did for the ISP,this review can raise concerns that necessitate important changes in the program. During the assessment and approval process for the ISP,FAA and the SEIC raised serious concerns in their written comments. Some of the SEIC’S comments were related to its February 1988 cost-benefit study on the ‘sir TrafficControl:ContinuedImprovements Neededin FAA’sManagement of theNASPlan(GAO/ RcEb-89-7,Nov.10,1988). Page6 GAO/RCEB-90-213 Interim SupportPhn R-240362 it planned to procure and provided support for the locations where it planned to install the equipment. FAA Is Developing FAA now plans additional interim projects for more computer capacity at Additional Interim terminals. FAAhas proposed one new project and is considering others to address unforeseen needs at terminals. These projects will cost at least Projects $126 million. One project will modernize and enhance FAA’Sterminal software at a cost of $59 million. The initial year of funding for this program was fiscal year 1990. Among other things, this project will standardize the software changes made locally at terminals to gain capacity. FAAis developing another project to provide 150 controller dis- plays for large terminal facilities, in addition to the 100 provided by ISP, at a cost of $67 million. FAAis also considering providing additional com- puter capacity for 5 to 15 “high density” terminal facilities, probably through systems like that at the New York terminal facility. On the basis of the cost of equipment and related items for the New York system, we believe upgrading each terminal will cost at least $8 million. Although our review did not assess the justification provided for these additional projects, we believe FAA should conduct a requirements analysis in con- formity with the Federal Information Resources Management Regulation for each to avoid the same problems that characterized the ISP’Splan- ning and assessment. FAA’Sprocurement and installation of IsP equipment are behind schedule, FAA’s Procurement although FAA considers the program urgently needed. When FAAplanned and Installation of ISP the ISP,it set unrealistic procurement and installation schedules. Equipment Are Behind At the April 1988 hearings to consider the first FAAbudget submission Schedule that included the ISP,the then-FAA Administrator said of the program: “there is probably no more serious operational requirement.” FAA’Sorig- inal schedules and plans to expedite procurements through sole source and existing contracts whenever possible reflected this urgency. By the end of fiscal year 1989, FAAhad planned to have virtually every ISPpro- ject under contract and the first equipment installed for several projects. However, only 5 of the 15 projects were under contract by the end of fiscal year 1989. At the time of our review, 8 of the 15 projects were under contract, but no ISPprojects have been completed and only 2 projects have had any equipment installed. On average, FAA is about 1 year behind its original contract schedule and 3 years behind its original installation schedule (see app. I). According to FAA’sestimates, it will be Page7 GAO/RCED-30.213 Interim SupportPlan R-240362 We believe that before proposing the ISP, FAA should have given this Conclusions expenditure from the Airport and Airway Trust Fund more careful con- sideration. FM should have better defined its interim requirements for sustaining terminal operations until long-term modernization programs are completed. Also, FAA should have conducted the formal assessment and approval process and requirements analysis before asking the Con- gress to fund the ISPand should have better addressed the major con- cerns raised during the process once it was conducted. If FAA had adequately planned and assessed the ISP,it might not have underesti- mated its interim capacity needs. As a result, FAA now intends to supple- ment the ISPwith additional projects. We also believe that, in planning the ISP,FAA should have set more real- istic schedules for procuring and installing ISPequipment. Despite the purported urgency of the program, FAA now estimates that it will be 1998 before the traveling public can benefit from FAA’s using all the ISP equipment at its terminal facilities. We have previously reported on problems with planning and imple- menting major NASPlan initiatives. The ISPexhibits the same problems. We are recommending changes to ensure that future interim projects are planned and assessed more effectively. Because FAA has been developing additional interim projects, we recom- Recommendations mend that the Secretary of Transportation direct the FAA Administrator to . ensure that FAA properly applies its assessment and approval process and the Federal Information Management Resources Regulation, including conducting requirements analyses, to future projects before submitting them to the Congress for funding; and . develop specific capacity, reliability, and maintainability requirements and goals for planning and assessing interim programs. To determine how effectively FAA planned and assessed the ISP,we reviewed studies on equipment upgrades, interviewed officials who were part of the 1987 team that devised the ISP,and reviewed documen- tation of the configuration management process for the ISP.We reviewed the SEX’S cost-benefit study and discussed the ISPand FAA’Splanning and approval process with SEICofficials. We interviewed the BP program manager as well as the facility and equipment program managers for the Page9 GAO/RCJ?&9O-213 Interim SupportPlan Page11 GAO/RCED-90213 Interim SupportPlan Abbreviations ARTS Automated Radar Terminal System ASR airport surveillance radar EAR% En Route Automated Radar Terminal System FAA Federal Aviation Administration ISP Interim Support Plan NAS National Airspace System ORD first operational readiness demonstration SEIC System Engineering and Integration Contractor TRACQN Terminal Radar Approach Control Facility Page13 GAO/RCEW90.213 Interim SupportPlan AppendixI ISPProjects Table 1.1:Status of ISP Contracts, as of April 1990 Dollars in millions Original Current contract award contract award cost of Projects milestone milestone8 project ARTS IllA -. Solid state memory 1989 1989 $47.5 Keyboards, trackballs, software, and - ancillary equipment -.~ 1989 1989 3.2 Additional radar oositions 1989 1990 83.4 - Disk drive refurbishment 1989 1989 7.9 ARTS II/HA - Separate flight plan 1989 1990 .3 Multi-sensor processor 1988 1990 4.8 Tape drive replacement 1989 1990 5.4 Additional radar positions 1989 1990 40.8 Uninterruptable power system 1989 1988b 75 Refurbishment of En Route Automated Radar Terminal System (EARTS) 1989 1989 3.1 Airport surveillance radars 1989 1990 135.6 Long-range radar upgrade 43.5 Receivers 1989 1990 c Transmitters 1990 1992 c integrated communicatrons swrtching systems bypass 1989 1990 17.0 Instrument landing system glade slope antennas 1989 1990 7.3 Airport weather sensors 1989 1990 5.1 Total estimated project cost 412.4 ISP program support 3.3 Total Estimated Cost $415.7 Note The years prowded are fiscal years “FAA has already awarded some contracts for projects wth current m&tones in 1990 Including the long-range radar receivers upgrade, airport weather sensors, and alrport swelllance radar projects “Funds were borrowed from other programs to make use of an existing contract, no ISP funds have been obligated yet on this Item ‘For the purposes of our rewew, we obtained only the total cost for the long-range radars and not how that total was broken down between receivers and transm!tters Page15 GAO/R4XIHW213 Interim SupportPIan Appendix1 ISPPmjeets FAA awarded contracts for upgrading both systems so that now all of the ARTSIII systems have been converted to ARTSIIIA, and FAA plans to con- vert all the ARTS11sto ARTSIIAs by August 1991. ARTSIIIA facilities have been experiencing limited track capacity and ARTS IIIA Solid State low memory reliability largely because of the outdated core memory Memory currently used. To increase capacity and improve reliability, FAA plans to replace the core memory at each ARTSIIIA facility with solid state memory. The solid state unit processes data 20- to 30-percent more effi- ciently than the core memory. Since the core memory is no longer manu- factured, FAA believes that replacing it with solid state units will facilitate maintenance and supply support of the existing ARTSIIIA equipment. FAA contracted for solid state memory in 1989. FAA claims that ARTSIIIA facilities have a high failure rate with their ARTS IIIA Keyboards, keyboards and trackballs, which together make up part of the data Trackballs, Software, entry subsystem. The keyboard is arranged in alphabetical and numer- and Ancillary ical sequence and is used by the controllers to enter flight data and to make modifications to the displayed information. To enter display coor- Equipment dinate information, the controllers rotate the black plastic trackball, which causes the small symbol on the display to move in a corre- sponding motion. FAA plans to procure 126 keyboards and trackballs and the necessary software and ancillary equipment (communications, clocks, etc.) to accommodate the hardware requirements. The planned procurement will provide two keyboard and trackball spares for each TRACON.FAA contracted for this equipment in fiscal year 1989 and expects that it will improve system performance and facilitate easier maintenance. Because of the increase in air traffic and training needs, FAA has deter- ARTS IIIA Additional mined that it needs 100 additional ARTSIIIA radar positions at a limited Radar Positions number of terminal facilities. FAA plans to replace the older ARTSIIIA data entry and display subsystems at 5 facilities with the 100 new fully digital ARTSdisplays and then use the displaced data entry and display subsystems for expansion positions at less busy ARTSIIIA locations. The installation of these additional positions will accommodate continued air traffic growth and eliminate the output display overload problems now encountered at some ARTSIIIA facilities. FAA expects to sign the contract for these new positions in fiscal year 1990. Page17 GAO/RCED9@213InterhSupportPlm Appendix1 ISPPmjecta occur with the magnetic tape unit. FAA plans to contract for disk drive subsystems to replace the tape drives at all ARTSII locations in fiscal year 1990. As with ARTSIIIA, FAAhas determined that because of projected air ARTS IIA Additional traffic growth and training needs at various ARTSIIA facilities, an addi- Radar Positions tional 168 radar positions are needed. FAA plans to procure the neces- sary additional radar displays: 30 to be used as operational positions and the remaining 128 to be used for training. Procuring these new posi- tions allows some radars to be designated specifically for training instead of training controllers on the operational displays that are needed to control traffic. FAA plans to contract for these new positions in fiscal year 1990. Some ARTSIIA facilities have been reporting a high incidence of outages ARTS IIA because of electrical power fluctuations and interruptions. These out- Uninterruptable ages occur because commercial power fluctuates beyond what FAA's Power System solid state systems can tolerate. FAA plans to resolve this problem by procuring uninterruptable power systems for those sites showing a high incidence of power-induced failures. These power systems will “rectify” the commercial power so that it is more easily accepted by solid state systems and thus less likely to fluctuate or stop. They will also provide a backup within 15 seconds if the commercial power fails. FAA procured these systems through an existing contract in fiscal year 1988. EARTSis essentially an expanded ARTSposition in an en route environ- Upgrades to En Route ment instead of in a terminal facility. EXWScan accept either short- or Automated Radar long-term radar information and can therefore provide both en route Terminal System and terminal air traffic control services, depending on what is needed at a given location. The ISPproject will upgrade or provide spare parts for (EARTS) three different parts of the system: tape drives, generators, and commu- nications It will also enable EIARTSto place lost or downed aircraft more accurately. Two of the four EARTSprojects have been completed, one is in operational testing, and FAA plans to complete contracting for the other project in 1991. ASR-9radars provide controllers with air traffic and weather informa- Airport Surveillance tion within about 60 miles of terminals, FAA uses such criteria as the cost Radars (ASR-9) of the establishment, the instrument operations at the terminal, and the page19 GAO/ECEDgO213InterimSopportPlan Appendix I JSP Projecta additional landing guidance to pilots, helping them to achieve the optimal glide slope of 3 degrees. FAA intends to contract for the conver- sion kits in fiscal year 1990. Airport weather sensors provide critical information to controllers who Airport Weather use the information to assist pilots in takeoff and landing procedures. Sensors The current weather sensors at terminals are failing because of their age. This has prompted supply support problems. FAA plans to procure 109 cloud height indicators and 80 remote readout hygrothermometers to replace the old equipment. The cloud height indicators determine the height of clouds with laser beams, replacing the older equipment that used the light reflected off the clouds to measure cloud height. The remote readout hygrothermometers give the controllers the temperature and the dew point, from which they determine the likelihood of fog. This new equipment will provide more accurate weather information and facilitate easier maintenance by ensuring manufacturer-supportable equipment. FAA contracted for the new sensors in fiscal year 1990, via an existing contract. Page 21 GAO/RCED90213 Interim Support Plan The first five copies of each GAO report are free. Addi&nal copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made ant to the Superintendent of Documents, when necessary. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. e U.S. General Accounting Offrice :;; P. 0. Box 0015 i ., Gaithersburg, MD 20877 &.’ Orders may also be placed by calling (202) 275-6241. ‘. i f-. Appendix II Major Contributors to This Report John W. Hill, Associate Director, Transportation Issues Resources, Allen Li, Assistant Director Community, and Robert E. Levin, Assistant Director Robert D. Wurster, Assignment Manager Economic Sara Magoulick, Evaluator-in-Charge Development Division, Kathleen M. Johnson, Evaluator Washington, D.C. (341247) Page22 GAO/RCELHJ&213 Interim SupportPlan - Appendix I JSP Projects weather at the terminal to determine whether a facility qualifies for an Ma-9 radar. FAA plans to procure six ASR-9radars with ISPfunds. Two of these will be the second radars at their respective sites; the remaining four are new establishments. These six sites qualified for radars after the original NASPlan Ma-9 radar contract was signed. These additional radars are being purchased through an option to the original contract. FAA executed this option in late 1989. Long-RangeRadar 200-mile radius. These radars are having supportability and maintain- Upgrade ability problems, largely because they are equipped with outdated vacuum tube circuitry. FAA plans to procure 76 solid state transmitter modification kits and 12 solid state receiver modification kits to replace the vacuum tubes. The radars will eventually be replaced under the NAS Plan, but FAA believes that upgrading the existing radars will allow these radars to be maintained until the replacement program is com- plete. FAA contracted for the receivers in late 1989 by modifying an existing contract and will establish a new contract for the transmitters in fiscal year 1992. Integrated cation between pilots and briefers in TRACONS and towers that handle Communications aircraft operating under instrument flight rules.’ As its name suggests, Switching Systems the bypass will make it possible to circumvent the integrated communi- cations switching systems and continue communications on a limited BYPa= number of frequencies when the switching system fails. This equipment will be installed at 80 facilities. FAA plans to contract for this equipment in fiscal year 1990. Obstructions around runways, either naturally occurring or resulting Instrument Landing from new construction, can adversely affect landings by interfering System Glide Slope with signals being transmitted to the planes by existing instrument Antennas landing systems. As a result, pilots can misgauge the angle of descent. To avoid this danger, FAA plans to procure 55 glide slope conversion kits, which, by adding a third antenna to the glide slope facilities, will give ‘Aircraft fly undereitherinstrumentor visualflight rules.Instrumentflight rulesaircraftmustbe controlledandin contactwith anair traffic controller,whilevisualflight rulesaircraftareonly monitoredby wntrollers Visualflight rulesaircraft,however,mustfollowtherulesandflight prow duresgoverningthe specificairspacein whichtheychoose to fly andarerestrictedfromsomeair- spacewoundmajorairportsfor safetyreasons. Page 20 GAO/RCEB90213 Interim Support Plan Appendix1 ISPRojects The disk drive units of the ARTSIIIA system store air traffic control ARTS IIIA Disk Drive software, load the software into computer memory, download the Refurbishment software from computer memory, and record air traffic control activity information. This recorded information is used to identify software glitches and investigate accidents and incidents. The units have been failing at an increasing rate. FAA plans to refurbish 154 disk drives, upgrade 86 disk control units, and procure disk test equipment. This project will involve pulling disk drives out of the TRACONSand replacing them temporarily with spares while they are refurbished. Once the refurbishment is complete, the drives will be reinstalled at the TRACKINS. FAA contracted for the refurbishment and spares in fiscal year 1989. To cope with insufficient flight plan capacity at ARTSIIA facilities, FAA ARTS IIA Separate plans to develop a separate flight plan data file for all ARTSIIA Flight Plan software. Thus far, all aircraft data have been contained in a single table in memory that holds a total of 256 aircraft. FAA has determined that a separate arrival/departure tabular list of flight plans will hold data for more aircraft and will in this way meet the continued demands of growing air traffic. FAA plans to contract for this new software in fiscal year 1990. The procurement of dual-sensor, dual-processor ARTSIIA systems is pro- ARTS IIA Multi-Sensor posed for the largest ARTSIIA locations and those ARTSIIA locations Processor requiring dual-radar coverage. Currently, ARTSIIA facilities are unable to accept more than 11 radar displays or to support dual-radar sensors. In addition to enabling the ARTSIIA system to accept more than 11 dis- plays or another radar sensor, this second processor will also provide a back-up level of operations in the event of a processor failure. The ISP will provide funding for the first site at Pensacola, Florida, and the con- tract will have options for the remaining sites. FAA is scheduled to con- tract for the processor in fiscal year 1990. Like the ARTSIIIA disk drives, the ARTSII magnetic tape drive units store ARTS II Tape Drive air traffic control software, load the software into computer memory, Replacement download the software from computer memory, and record air traffic control information. The problems with these tape drives include both a high failure rate and inadequate supply support because one of their necessary components, the drive motor, is no longer manufactured. This situation is especially critical because the facilities have no back-up method to permit a reload of the operation software if a failure should Page13 GAO/RCED-90-213InterimSupportPLan Appendix1 JSPPmjects Table 1.2:Status of the Implementation of ISP Projects, as of April 1990 Origina\:;: Current or a actual ORD Equipment Projects milestone date milestone installed ARTS IllA SolId state memory 1990 1990 yes Keyboards, trackballs, software, and ancillary equipment 1989 1992 “0 Additional radar positions 1990 1993 no Disk drive refurbishment c 1992 “0 ARTS IIA Separate flight plan 1989 1992 “0 Multi-sensor processor 1989 1992 no Tape drive replacement 1989 1992 -~~ “0 Additional radar oosltions 1989 1992 “0 Unlnterruptable power systems 1989 "0 Refurbishment of En Route Automated Radar Terminal System (EARTS)b c yes AIrport Surveillance Radars 1993 “0 Long-range radar upgrade Receivers 1989 1991 “0 Transmitters 1990 1995 “0 Integrated communications switching systems bypass 1989 1993 --~- “0 Instrument landing system glide slope antennas 1989 1992 no Airport weather sensors 1989 1990 “0 Note: Years provided are fiscal years. aORD means first operational madness demonstration ‘This ISP project refurbishes or provides spare parts for three different parts of the EARTS system The equipment mvolved in these efforts is all either fully or partially installed In 1991, FAA hopes to complete contracting for the fourth part of this project. which will enable EARTS to place lost or downed alrcraft CORD dates not available for these projects Air traffic controllers at FAA'S Terminal Radar Approach Control Automated Radar (TRACON) facilities sequence and separate aircraft arriving at or Terminal System departing from airports under their control. These TRACONS are each (ARTS) equipped with Automated Radar Terminal Systems (ARTS), a computer system that receives input from radar, identifies and tracks aircraft, associates the aircraft with flight plans, provides safety warnings, and displays aircraft identification and position location to controllers. The ARTSsystem is divided into two groups: ARTS III, which supports the 63 largest TRACONS, and ARTSII, which supports the 119 smaller TRACONS. Page16 GAO/RCEB9@213InterimSupport~h Appendix I ISP Projects The Interim Support Plan (ISP) is divided into 15 projects, which are intended to forestall system support problems, increase capacity, and provide for expansion of the air traffic system until the Advanced Auto- mation System is implemented under the National Airspace System (NAS) Plan. Four of these projects were contracted in fiscal year 1989, four have been contracted in fiscal year 1990, and the remaining projects are scheduled to be contracted in late fiscal year 1990 or in 1991. Table I.1 gives the dollar value of each project and the original and recently revised schedule for contracting. Table I.2 gives the original and recently revised schedule for the first installation under each project and shows whether any of the upgrades have been installed. This appendix also briefly describes the ISP projects, the inadequacies of the existing system that the Federal Aviation Administration (FAA) believes made the projects necessary, and what FM hopes to accomplish through their installation. Page 14 GAO/RCRB9@213 Interim Support Plan Contents Letter 1 Appendix I 14 ISP Projects Automated Radar Terminal System (ARTS) ARTS IIIA Solid State Memory 16 17 ARTS IIIA Keyboards, Trackballs, Software, and 17 Ancillary Equipment ARTS IIIA Additional Radar Positions 17 ARTS IIIA Disk Drive Refurbishment 18 ARTS IIA Separate Flight Plan 18 ARTS IIA Multi-Sensor Processor 18 ARTS II Tape Drive Replacement 18 ARTS IIA Additional Radar Positions 19 ARTS IIA Uninterruptable Power System 19 Upgrades to En Route Automated Radar Terminal System 19 (EARTS) Airport Surveillance Radars (ASR-9) 19 Long-Range Radar Upgrade 20 Integrated Communications Switching Systems Bypass 20 Instrument Landing System Glide Slope Antennas 20 Airport Weather Sensors 21 Appendix II 22 Major Contributors to This Report Tables Table 1.1: Status of ISP Contracts, as of April 1990 15 Table 1.2: Status of the Implementation of ISP Projects, as 16 of April 1990 Page 12 GAO/RCF.D-30.213 Interim Snpport Plan E-240362 - individual ISPprojects. We discussed the need for the ISPwith FAA’s Air Traffic Requirements staff, officials at a terminal radar approach con- trol facility, and an official at the FAA Supply Depot. To determine the status of FAA's implementation of the ISP,we discussed its progress with the ISPprogram manager, Air Traffic Requirements officials, and the facility and equipment program managers. We reviewed project performance schedules maintained by the SEICfor cer- tain ISPprojects. We also obtained funding information from FAA’s budget office. Our review was conducted between August 1989 and July 1990 in accordance with generally accepted government auditing standards. As agreed with your office, we did not obtain official agency comments. However, FAA officials reviewed a draft of this report for accuracy. They generally agreed that the information in the report was accurate. In a few instances, they suggested some technical changes, which we made as appropriate. Unless you publicly announce its contents earlier, we plan no further distribution of this report until 14 days from the date of this letter. At that time, we will send copies to interested con- gressional committees; the Secretary of Transportation; and the Admin- istrator, FAA. We will also make copies available to other interested parties upon request. This work was performed under the direction of Kenneth M. Mead, Director, Transportation Issues, who may be reached at (202) 275-1000. Major contributors are listed in appendix II. I/ J. Dexter’Peach V Assistant Comptroller General Page 10 GAO/‘lWED3C-213 Interim Support Plan R-240362 1998 before the traveling public can benefit from all the ISPequipment at terminal facilities. Obligation rates for the ISPare not as high as planned, which also indi- cates the program is behind schedule. FAA planned to obligate ISPfunds in the year they were appropriated, although the funds are available by law for 5 years. We agree that urgency would be best served by obli- gating the funds as soon as FAA can arrange to do so. However, despite reprogramming $11.5 million of ISPfunds to other NAS Plan projects, FAA had obligated only 34 percent of the remaining fiscal year 1989 funds as of March 1990, leaving about $91 million to be obligated. As of the same date, FAA had obligated about 31 percent of the fiscal year 1990 funds, leaving about $50 million to be obligated. By the end of fiscal year 1990, FAA estimates that only 57 percent of 1989 and 1990 funds will be obli- gated. In addition, FAA has extended its funding plans for the ISPinto fiscal year 1992, a year beyond the original plan. Clearly, FAA’s initial contracting, installation, and obligation schedules for the ISPwere optimistic. The FAA official responsible for the original schedules based them on information provided by FAA program man- agers who were to procure the equipment. He believes the program man- agers were aware of the urgency FAA attributed to the ISPand provided optimistic estimates that reflected that urgency. Because the official had recently transferred to headquarters from an FAA field office and had limited procurement experience, he said that he did not recognize that the information was unrealistic. Given the history of ATC system procurements, we believe that FAA could have foreseen that its original schedules were optimistic. We reported in November 1988 that NAS Plan projects were behind their 1983 schedules by an average of 3 years. We recently testified that pro- ject delays persist.” A comparison of FAA’s estimates as of January 1990 with its prior year estimates showed that 8 of 12 major projects experi- enced an additional delay of 200 or more dayslo In view of these past trends, FAA needs to guard against setting unrealistic schedules for future interim projects. %ues Relatedto FAA’sFiscal Year 1991Budget Request (GAO/T-RCED-90-66, Apr. l&1990). ‘“Air TrafficControl:Statusof FAA’sEffortto Modernize theSystem(GAOIRCED-SO-146FS, Apr. 17,1990). Page 8 GAO/RClXNO-213 Interim Support Plan R-240352 ISP,which is required by the configuration management process. The study analyzed 10 of the 15 ISPprojects and determined that benefits outweighed costs for only 2 projects, which upgrade existing air traffic controller positions, and a portion of a third project on airport radars. However, FAA stated in its March 1989 proposal for the ISPthat a cost- benefit study was not conducted because the projects are necessary and the benefits of capacity improvement and system maintainability have no generally accepted standards for quantification. Although FAA now recognizes that a cost-benefit study was conducted, it believes the study underestimates ISPbenefits. In its comments on the ISPproposal, the SEIC stated that more than a dozen studies have demonstrated valid and FAA- approved methods of assessing benefits like those projected in the ISP. We agree with the SEICthat valid analyses of benefits are possible and believe that cost-benefit analysis is a useful tool for planning the NAS modernization.* The SEICand FAA offices had other concerns about the documentation provided for the ISP.The SEICcommented that “there is no supporting documentation to define how the requested quantities were determined, or at what locations the improvements will be made.” In addition, the SEICrecommended that “a line-by-line approval of the ISPinitiatives be undertaken since there is no integrated functional relationship among the various ISPinitiatives.” By this, the SEICmeant that since the ISP projects upgrade such different types of equipment, they should be con- sidered separately. One FAA office, the Management Control Service, commented that the justification of urgency for the ISPwas “vague and contains generalities.” Another, the Office of System Engineering and Program Management, was concerned that some of the proposed airport radars did not meet FAA’Sown criteria for establishing which airports need radars. In response to these concerns, in April 1989, FAA cancelled its plans to procure the airport radars that did not meet its own establishment cri- teria. The change reduced the total projected cost of this ISPproject by about $88 million. FAA did not change any of the other seven procure- ments that were found not to be cost-beneficial, formally refute the SEIC’Scost-benefit study, provide additional supporting documentation for quantities or locations, or conduct a line-by-line approval. We believe FAA should have been more responsive to the concerns about the ISP.FAA should have documented its explanation for the quantities of equipment ‘Air TrafficControl:ContinuedImprovementa Needed in FAA’sManagement of theNASPlan(GAO/ mBS9-7, Nov.10,19&J). Page 6 GAO/RCEDw)-213 Interim Support Plan E240352 group, the ISPwas required because of capacity, reliability, and main- tainability shortfalls with current equipment, but we could find no evi- dence that it documented actual or projected shortfalls in these areas. Without a requirements analysis, FAA had no sound basis for claiming such shortfalls. FAA would have been in a better position to perform a requirements analysis for the ISPif it had a capacity management and performance monitoring program. In July 1989, we reported that FAA had not effec- tively defined its capacity needs and did not monitor the performance of its computer systems.” In December 1989-Z years after initiating the ISP-FAA responded that it was implementing a performance monitoring and management tool at large terminals. In June 1990, we recommended that the Secretary of Transportation direct the Administrator, FAA, to set up computer capacity and performance management programs at smaller terminals as well.” FAA would also have been in a better position to assess its requirements for interim equipment if it had capacity, reliability, and maintainability goals. Currently, FAA only has such goals for future equipment. FAA uses the NASSystem Specification to assess long-term modernization pro- grams.6 For example, the NASSystem Specification calls for the area con- trol computer complex, a major modernization project, to have a system availability of 99.9995 percent and a mean time between failures of 3,340 hours. The FAA official responsible for developing and imple- menting configuration management policy, procedures, and standards acknowledges that the NASSystem Specification does not provide suffi- cient goals for interim projects and that such goals are needed. The inadequate definition of requirements for ISPfollows a familiar pat- tern. FAA has had this problem with 5 of the 12 major NASPlan systems. We reported in November 1988 that FAA had inadequately defined 4S (GAO/IMTEG89-63, July 6, 1989). “Air TrafficControl:SmallerTermmalSystemsCapacityRequirements Needto BeDefined(GAO/ IMTE7J _90_60, June26,19901. “TheNASSystemSpecification, Volume1,FunctionalandPerformance Requirements for the NationalAirspaceSystem, Generalestablishesthefunctional,performance,design,manufacture/con- struction,logistics,personnelandtraining,documentation, verification,andinterfacerequirements for theNationalAirspaceSystem. Page4 GAO/‘RCEB9@213 Interim SupportPlan 5240362 The ISPis not progressing in a manner that reflects its stated urgent nature. Because FAA is about 1 year behind its contracting schedule and about 3 years behind its installation schedule, it will be 1998 before the traveling public can benefit from all the ISPequipment. FAA’Sproblems with the ISPare familiar. We have previously reported on inadequately defined requirements and schedule slippages that delay the benefits of ATC modernization to the traveling public. Changes are needed to ensure that future projects are planned and assessed more effectively. In 1981 FAA embarked on a long-term program, the National Airspace Background System (NAS) Plan, to modernize the air traffic control system. The NAS Plan is funded by the Airport and Airway Trust Fund. The NASPlan’s purposes are to achieve safer airspace and a more efficient ATC system at an affordable cost to the government and the traveling public. FAA designed the ISPto meet field operational and maintenance needs pending the implementation of NASPlan programs. To expedite the interim program, FAA elected to use existing contracts, and sole source and off-the-shelf procurements when possible. FAA considered replacing the systems with improvements that needed developing and testing or just providing additional maintenance for the ATC systems, but it rejected these alternatives as either too slow or too expensive to meet its needs. As described in draft FAA Order 1800.8F, National Airspace System Con- figuration Management, configuration management is FAA’s means of formally assessing and approving changes to the NASPlan. Among other things, the order calls for FAA to conduct a thorough requirements anal- ysis so that changes are traceable to operational requirements and sup portive of the FAA’s goals and objectives. Several FAA offices and the FAA’SSystem Engineering and Integration Contractor (SEIC) were responsible for conducting the configuration management process for the BP. Further, the Federal Information Resources Management Regulation, Part 201-30, requires agencies to base acquisitions of new or additional information processing resources on mission needs. These needs are to ‘In 1984,FAAs&&d MartinMariettaasits Systems Engineering andIntegrationContractorto act asa technicaladviserfor integratingall theNASPlansystems. Page2
Air Traffic Control: The Interim Support Plan Does Not Meet FAA's Needs
Published by the Government Accountability Office on 1990-09-11.
Below is a raw (and likely hideous) rendition of the original report. (PDF)