United States General Accounting Office Report to the Chairmen, Subcommittee on Surface Transportation and Subcommittee on Aviation, Committee on Public Works and Transportation, House of Representatives AVIATION SAFETY Management Improvement Needed in FAA’s Airworthiness Directive Program , s ‘ United States GAO General Accounting Office Washington, D.C. 20648 Resources, Community, and Economic Development Division B-238278 February 16,199O The Honorable Norman Y. Mineta Chairman, Subcommittee on Surface Transportation Committee on Public Works and Transportation House of Representatives The Honorable James L. Oberstar Chairman, Subcommittee on Aviation Committee on Public Works and Transportation House of Representatives . This report responds to an October 11, 1988, request that we explore, through several assignments, the Federal Aviation Administration’s (FAA) ability to provide meaningful oversight of aviation safety. Pursu- ant to your request we previously reported on FAA’S removal procedures for pilot examiners,’ the agency’s inspection management system,2 and changes under consideration regarding aviation medical standards3 As agreed with both subcommittee staffs, this report focuses on FAA’S over- sight of the airworthiness directive (AD) program because ADS are key elements of FAA’S safety responsibility. ADS are rules that FAA issues requiring airlines to correct conditions in their aircraft, such as cracking and corrosion, that can jeopardize safety. Because of the AD program’s critical nature, we agreed to examine whether (1) FAA’S oversight is ade- quate to determine airline compliance with AD requirements and (2) safety-related information maintained by FAA could be used to improve the program’s effectiveness. Our review focused on FAA’S oversight of the AD process, not on the merits or validity of individual ADS. I FAA’S oversight of the AD program is inadequate to determine whether Rebults in Brief the airlines are complying with ADS.The National Transportation Safety Board’s (NTSB) investigation of the widely publicized April 28, 1988, Aloha accident, in which the aircraft lost the upper portion of its cabin ‘Aviation Safety: FAA Has ImprovedIts RemovalProceduresfor Pilot Examinen (GAO/ ‘Aviation Safety:FAA’s Safety InspectionManagementSystemLacksAdequateOversight(GAO/ RCD-90-36, 3Aviation Safety:FAA Is ConsideringChangesto Aviation MedicalStandards(GAO/RCED-90-68FS, Jan. 9,lQQO). Page 1 GAO/RCED-99-94 Management Improvement Needed in FAA’s AD Program while in flight, and other FAAspecial inspections have documented sev- eral important examples of AD noncompliance. FAAdoes not know the extent of AD compliance throughout the airline industry because FAA inspectors do not, and are not required to, verify AD compliance during each inspection. In addition, when AD compliance is checked, current guidance requires inspectors to report only noncompliance. As a result, FAAdoes not have information on the extent to which airlines are in compliance with ADS.FAAneeds to measure the extent of noncompliance and decide what actions it should take to ensure AD compliance through- out the airline industry. FAAinspection guidelines are very broad, in that they allow inspectors to decide whether to check for AD compliance, which ADSto include in their check, and which aircraft to inspect. FAAinspectors do not routinely check for AD compliance during airline inspections. For example, during our review we accompanied FAAinspectors on 10 routine surveillance inspections and noted that AD compliance was checked in only 2 instances because the inspector decided to emphasize non-AD-related items. In addition to obtaining more complete and comprehensive information from its safety inspectors, FAAshould better use its existing safety data bases. Use of such data could help focus airline inspections on problem areas, thereby improving the effectiveness and efficiency of FAA’Slim- ited inspector work force. the safe travel of U.S. air passengers-about 492 million in 1988-as a joint responsibility of FAAand the airlines. FAApromotes aviation safety by issuing regulations that stipulate certain requirements that airlines must meet to operate commercial aircraft. FAA’Saviation inspectors then monitor to ensure that airlines comply with these safety requirements. When airline accidents do occur, mechanical failures-such as in land- ing gear and aircraft structural components-are factors in 40 percent of the accidents. FAA’sAD program addresses such unsafe conditions and is a key element of FAA’S safety responsibility. Although FAAcertifies new aircraft models as safe before they are used in commercial service, it also issues ADS to address unsafe mechanical conditions that surface after the aircraft has been in use. ADSare requirements that FAA issues for the airlines to identify and correct unsafe aircraft conditions that have occurred, or are likely to occur, in Page 2 GAO/RCED-90-94 Management Improvement Needed in FM’s AD Program B238278 other aircraft of the same design. ADSalso prescribe corrective actions that airlines must take to correct identified problems in their aircraft. For example, a 1988 AD identified several incidents involving Boeing 737s in which an engine bolt failed. In one case, the engine separated from the aircraft. The AD required airlines to add a secondary engine mount support, install a failure indicator for the bolt, and periodically inspect the indicator. FAA becomes aware of most unsafe aircraft conditions through communi- cations with aircraft manufacturers or reports of significant incidents or accidents. Using engineering judgment, FAA decides if these conditions warrant an AD. If needed, FAA engineers work with the aircraft manufac- turer to develop the AD. The manufacturer usually prescribes the proce- dures needed to monitor and correct the unsafe condition, and FAA reviews and approves these procedures. FAA also establishes the time allowed for the airlines to comply with the AD. In determining the com- pliance time frames, which can range from immediate action to several years, FAA engineers consider factors such as the severity of the unsafe condition, the availability of parts needed to correct the condition, and the potential economic impact the AD will have on the aviation industry. In addition, some ADSrequire a one-time repair of an unsafe condition, while others require a combination of repetitive inspections, monitoring, and eventual repair. FAA issues about 200 ADSeach year for large commercial aircraft. At any one time, a particular aircraft type may have many ADS requiring recur- rent inspections or repairs. After issuing an AD, FAA monitors compliance through its airline inspec- tion program. When FAA inspectors find AD noncompliance, the airlines must comply with the AD before operating the aircraft again. Inspectors may also recommend that FAA impose a civil penalty or take other administrative action against the airline. Recent accident investigations and special inspections by the NTSB and FAA’s Management FAA found instances in which airlines have not complied with ADS. and Oversight of AD Because FAA'S guidelines allow aircraft inspectors discretion in deciding Compliance Needs when, what ADS, and how many aircraft to inspect for AD compliance, the inspectors do not always verify compliance during their routine Improveme’nt inspections. In addition, FAA does not know the industry-wide extent of AD noncompliance because it does not receive information about the Page 3 GAO/RCJD-90-94 Management Improvement Needed in FM’s AD Program B-228278 number of ADSinspected for or the number of airlines found in compliance. Incjdents of AD Recent NTSB accident investigations and special inspections by FAA found that several airlines were not complying with ADS. For example, FAA Nopcompliance issued an AD in October 1987 that addressed deterioration in the Boeing 737 fuselage, the aircraft type that was later involved in the Aloha acci- dent. The AD required airlines to conduct visual inspections for cracking; and if cracks were found, the AD required airlines to perform additional technical inspections. According to NTSB'S accident report, Aloha’s records neither documented these required additional inspections nor accurately reflected the condition of the aircraft. Further, FAA records showed no evidence that its inspector had verified AD compliance. NTSB concluded that Aloha Airline’s maintenance program and FAA's over- sight of the airline’s flight operations were inadequate. NTSB therefore recommended that FAA'S inspection program place greater emphasis on evaluating the actual condition of each aircraft and the airlines’ compli- ance with specific ADS. FAA'S own experience with direct inspection of airline operations has also found instances of serious AD noncompliance. In 1986, as a result of an Eastern Airline aircraft accident, FAA conducted an in-depth review of the company’s adherence to FAA regulations. This effort combined records reviews and direct aircraft inspections. FAA found, among other problems, that Eastern had flown two aircraft for almost 6 years with- out complying with an AD pertaining to the landing gear and without FAA inspectors’ detecting the noncompliance. One aircraft made over 10,000 flights while not in AD compliance. The aircraft was damaged when the landing gear-which was the subject of the An-failed. Another aircraft flew more than 8,900 flights while not in compliance with the same AD. Furthermore, during one 6-day period, Eastern operated 37 aircraft on over 1,100 flights without properly complying with an AD that required recurrent inspections for cracks. On the basis of this review, FAA imposed a $9.6 million fine for a variety of violations, including AD non- compliance. Also, FAA'S 1987 “white glove” inspection of 8 airlines found 26 instances of AD violations. Monitoring ADdCompliance When issuing an AD, FAA relies on the airlines to comply with its require- ments but does not require them to report compliance. Instead, FAA relies on its aircraft inspectors to monitor airlines’ compliance during routine surveillance inspections. However, FAA'S guidelines are too discretionary Page 4 GAO/RCELW@94 Management Improvement Needed in FAA’s AD Program ‘? + ’ , B-228278 to ensure that critical safety areas, such as AD compliance, are ade- quately covered during inspections. We also found that FAA does not have information on the number of ADSchecked by inspectors or the extent to which airlines comply with ADS. J&q&ion GuidanceDoesNot ADSare critical elements in maintaining aviation safety because they Requ@e Verification of AD require airlines to correct potentially unsafe aircraft conditions. How- Corn liance ever, FAA cites AD compliance as 1 of more than 20 area&such as pilot training and emergency equipment, that inspectors may consider includ- f ing in an inspection. The guidelines do not specify how much emphasis inspectors should place on reviewing each inspection area, nor do they I require inspectors to verify AD compliance during each inspection. In , addition, the guidelines allow inspectors discretion to determine how often to conduct AD inspections, which ADSto check for compliance, and how many of the airline’s aircraft to inspect. Inspectors do not always verify AD compliance during routine inspec- tions because some discretion in the inspection process is necessary to allocate FAA's limited work force among an increasing number of aircraft and, hundreds of ADS. In addition, some types of inspections, such as enroute and ramp inspections,4 do not lend themselves to inspecting for AD compliance. Other types of inspections, however, such as mainte- nance records checks and spot checks of aircraft undergoing periodic service, are appropriate for verifying AD compliance. Because Aos pre- scribe corrective actions that airlines must take to correct known unsafe conditions in their aircraft and because several recent incidents suggest serious weaknesses regarding airline compliance with ADS,FAA needs to require its inspectors to test for AD compliance as part of each inspection. We accompanied FAA inspectors on 10 routine inspections. In only two instances did the inspector check for compliance with an AD during the inspection-once by reviewing maintenance records and once by com- paring maintenance records to work done on the aircraft. In the other eight instances, due to FAA'S discretionary inspection policy, the inspec- tor decided to emphasize non-m-related inspection duties, such as fol- lowing up on reported mechanical problems during a flight, reviewing maintenance records for current work being performed on an aircraft, 4An “enroute”inspectionis a checkof an aircraft during a flight and includesobservingthe flight crew’susageof aircraft equipmentand the performanceof the equipment.A “ramp”inspectionis a checkof an in-serviceaircraft and includesobservingthe refuelingof the aircraft, passengerhan- dling, and the conditionof the aircraft. Page 5 GAO/RC~BO-94 Management Improvement Needed in FAA’S AD tim B-238278 and reviewing training records for airline personnel, instead of checking for AD compliance. F* NeedsMoreCompleteAD FAA does not know the extent of AD compliance throughout the airline Compliance Information industry because the agency does not have information on the extent to which inspectors check for, or find airlines in compliance with, ADS. Inspectors, in accordance with FAA'S procedures, report instances in which they find that airlines are not in compliance; they are not, how- ever, required to report when they observe AD compliance. Requiring inspectors to both verify AD compliance during inspections and report complete inspection results-compliance as well as noncompliance- would provide FAA with (1) a data base to measure the extent of AD non- compliance and (2) a management tool to help administer the program more effectively. Without such data, FAA cannot determine whether it has established appropriate emphasis and policies regarding AD compli- ance inspections. -FA1A.Can More safety-related information to help focus its inspections and thereby Acitively Use Available improve the effectiveness of its limited inspector resources. Because FAA Safety Data relies on inspectors to verify An compliance, a more effective inspection effort would also help FAA identify where changes should be made in the AD program to further improve aircraft safety. The NTSB and the Airwor- thiness Assurance Task Force, a technical panel of industry and govern- ment aviation experts, both recently concluded that FAA'S inspection efforts are largely reviews of paper records with limited hands-on air- craft observation. FAA acknowledges it would prefer more hands-on inspections, but it cites manpower limitations as precluding it from inspecting all aircraft and verifying compliance with all ADS. Therefore, FAA must judiciously allocate its limited resources to achieve the maxi- mum effectiveness from its inspection program. Analyzing existing safety-related data, such as in FAA'S Service Diffi- culty Report system, could provide this focus by raising warning signals regarding specific aircraft, aircraft types, or airlines that warrant a closer hands-on inspection. FAA uses these data to help focus its special inspections but does not as a matter of course use the data for routine airline surveillance. FAA has maintained extensive safety information in various data bases. For example, FAA requires airlines to report mechanical problems that occur in aircraft, such as problems with landing gear, identification of Page 6 GAO/RCED-90-94 Management Improvement Needed in FM’s AD Program B-238278 corrosion and cracking, and engine shutdowns, to its Service Difficulty Report system, which has existed for over 2 decades. In 1988, commer- cial airlines reported approximately 19,000 mechanical difficulties to the system. We analyzed Service Difficulty Report data by aircraft type, such as the Boeing 727 and McDonnell Douglas DC-g, to determine whether this information could be used to augment FAA’S inspection process. We reviewed a limited sample of reports that airlines had submitted to FAA between January 1983 and June 1989 and noted a wide range in the frequency of problems reported across aircraft type and airlines. For example, in reviewing Service Difficulty Report data for 727s, we noted that one airline reported 37 instances of structural problems involving cracking and corrosion found during inspections of one of its aircraft. All of these problems were found during eight aircraft inspections that began in September 1984 and concluded in January 1988. During the last inspection, the airline found and repaired 28 instances of cracks and corrosion in areas such as floor beams, brackets, and fuselage skin. However, the airline did not submit any additional Service Difficulty Reports to FAA from January 1988, when repairs were made, to June 1989, the ending date of our analysis. If FAA analyzed these safety data, the agency could better focus its inspection efforts by selecting for rein- spection this or other aircraft with similar findings to determine whether any additional cracking and corrosion have occurred since the last reported inspection. FAA could also use Service Difficulty Report information to monitor whether airlines are finding and correcting generic problems, such as cracking and corrosion, that can occur in aircraft. For example, for one type aircraft, we examined six Service Difficulty Reports that airlines had submitted to FAA. Five of the reports stated that the airlines had found and corrected cracks and corrosion in their aircraft. Airlines oper- ating other aircraft makes and models also reported finding these and other types of deterioration. While this type information cannot be used exclusively to determine problems with specific aircraft, aircraft type, or airlines, FAA can use the data as a warning signal of potential safety problems. FAA inspectors could use these signals, along with their knowledge of other factors- such as aircraft age, utilization, airline maintenance practices, and vari- ous types of normally expected deterioration-to determine which spe- cific aircraft, aircraft type, or airline should receive increased inspection emphasis. FAA notes that its reporting system may have some problems Page 7 GAO/RCEDW-94 Management Improvement Needed in FAA’s AD Program B-288278 regarding the quality and timeliness of the data. In a separate review, we are currently assessingFAA’S Service Difficulty Report system to determine how to make these data more useful. In 1987, we issued two reports that identified improvements needed in FAA’S airline inspection program. In one report, we recommended that FAA use safety data to target its inspector resources toward high-risk conditions.6 In the other, we recommended that FAA revise its inspection guidance to take into account the need to target airlines displaying char- acteristics that may indicate safety deficiencies.6 Consistent with these recommendations, FAA could use the kind of data maintained in the Ser- vice Difficulty Report system, as well as other systems and data bases maintained by the agency, to develop indicators of potentially unsafe conditions in aircraft as tools to help target its limited inspection resources. Because by their nature ADSaddress critical safety conditions, airline Conclusions compliance with AD requirements and FAA’s ability to effectively oversee and manage the AD program are vital to aviation safety. Poor implemen- tation of AD requirements can potentially result in aircraft accidents. This is illustrated by the Eastern Airlines incident involving the failed landing gear. Furthermore, the NTSB investigation of the Aloha accident and FAA’S special inspection results demonstrate that (1) some airlines are not complying with ADSand (2) AD noncompliance can remain unde- tected by FAA for long periods of time. FAA does not compile information on the extent of AD noncompliance because (1) the number of ADS inspected for is left to the discretion of each aviation safety inspector and (2) the inspectors report only when airlines are not in compliance. Consequently, FAA is not aware of the number of ADSinspected for or the number verified as being in compli- ance. This information is needed, however, to measure the extent of AD compliance throughout the airline industry. The instances of noncompli- ance discovered by recent accident investigations and special FAA inspec- tions, coupled with the wide discretion given inspectors in verifying ADS and incomplete inspection reporting, indicate a need for FAA to improve its management and oversight of this critical safety program. FAA needs %zpartmentof Trans rtation: EnhancingPolicy and ProgramEffectivenessThroughImproved Management(GAO/&87-3, Apr. 13,1987). 6AviationSafety:NeededImprovementsin FAA’sAirline InspectionProgramAre Underway(GAO/ m-62, May 19,1987). Page 8 GAO/RCED-W94 Management Improvement Needed in FAA’s AD Program to know if its inspectors are checking for AD compliance and the extent to which airlines are complying with ADS.With more complete informa- tion, FAA would have the management tools needed to determine whether (1) airlines are complying with ADSand (2) additional regula- tory action is needed to ensure AD compliance. Also, FAA can improve the management of its AD process by analyzing existing safety data to help determine where to focus its limited inspec- tion resources, FAA can use this information to identify which aircraft, aircraft type, or airlines warrant additional hands-on inspection emphasis. To improve FAA’s management and oversight of the AD program, we rec- Recommendations ommend that the Secretary of Transportation direct the Administrator, FAA,to require a systematic testing for AD compliance as part of each routine airline inspection, require inspectors to report which ADSare tested and the extent of air- line compliance found during each inspection, and maintain and analyze compliance information to determine the extent of AD noncompliance and any additional actions necessary to ensure that airlines comply with ADS. We also recommend that the Secretary direct the Administrator, FAA, to analyze and use available aircraft safety data as a management tool to focus FAA'S limited inspection work force. We discussed the facts presented in this report with cognizant FAA offi- cials. They agreed that the report is accurate. However, as requested by your office, we did not obtain official agency comments. Details on our objectives, scope, and methodology are contained in appendix I. Unless you publicly announce its contents earlier, we plan no further distribution of this report until 16 days from the date of this letter. At that time, we will send copies to the Secretary of Transportation and the Administrator, FAA, and make copies available to others upon request. Page 9 GAO/RCED9@94 Management hprovement Needed in FM’s AD Program . n: 7 We performed our work in accordance with generally accepted gov- ernment auditing standards under the direction of Kenneth M. Mead, Director for Transportation Issues, who can be reached on (202) 276 1000. Other major contributors to this report are listed in appendix II. m&J J. Dexter Peach Assistant Comptroller General Page 10 GAO/RCED-9044 Management Improvement Needed in FM’s AD Program Y Page 11 GAO/RCELMJO-94Management Improvement Needed in FAA’s AD Program Appendix I objectives,Scope,and Methodology On October 11, 1988, the then-Chairman of the Subcommittee on Avia- tion, House Committee on Public Works and Transportation, Norman Mineta, requested that we explore FAA's ability to provide meaningful oversight of aviation safety. This report is one of several assignments responding to the chairman’s request. As agreed with the subcommittee staff, we concentrated this current review on FAA'S AD program because ADS are key elements of FAA'S safety responsibility and critical to ensuring aviation safety. We reviewed how FAA identifies and addresses unsafe conditions in commercial aircraft and whether (1) FAA'S oversight is effective to determine AD compliance and (2) safety-related information maintained by FAA could be used to improve the program’s effectiveness. We limited our review to the AD process for commercial aircraft because these aircraft transport 96 per- cent of all aviation passengers. We did not review the process for smaller, general aviation aircraft. We focused on FAA'S management of the AD process and did not assess the merits or validity of individual ADS. To determine how FAA identifies and addresses unsafe conditions, we (1) interviewed FAA engineers who write ADS in Seattle, Washington; Long Beach, California; and Boston, Massachusetts; (2) reviewed AII documen- tation and files; and (3) accompanied FAA inspectors on aircraft inspec- tions, In addition, we interviewed engineers from two aircraft manufacturers, the Boeing Aircraft Company and McDonnell Douglas. Also, we interviewed officials from NTSB and the Air Transport Associa- tion regarding their roles in the AD development process. We discussed methods that FAA uses to identify unsafe aircraft conditions, the manu- facturer’s input into the AD process, how an AD is developed, and the engineering judgment used in the process. We reviewed files from 14 major ADSand identified the extent of documentation supporting the judgments and decisions made as the engineers developed the ADS. We also tested to determine whether the files contained evidence of proper internal controls, such as documented levels of review and approvals. To assess the adequacy of FAA's oversight of AD compliance, we accom- panied FAA inspectors on aircraft inspections in Pt. Lauderdale and Miami, Florida, and Atlanta, Georgia, and observed their routine surveil- lance and AD compliance activities. We supplemented these observations by reviewing inspector guidelines and discussing inspection practices with the field inspectors and with the directors of FAA'S Flight Standards in F’t. Lauderdale, Miami, Atlanta, and FAA headquarters. We also reviewed FAA requirements for reporting inspection results. Finally, we Page 12 GAO/RCED-90-94 Management Improvement Needed in FAA’s AD Program Appendix I Objectivee, Scope, and Methodology reviewed the NTSB'S report on its investigation of the Aloha Airlines acci- dent, FAA'S actions taken against Eastern Airlines for AD noncompliance, and the results of FAA's 1987 special inspection program. To determine whether safety data could be used to improve the AD pro- gram, we interviewed FAA officials and analyzed safety-related data reported by airlines to the Service Difficulty Report system in Washing- ton, D.C., and the agency’s computer center in Oklahoma City, Oklahoma. We discussed the feasibility of using existing safety-related data as a means to focus inspections and identify unsafe conditions in aircraft. We conducted our review between February and October 1989 using generally accepted government auditing standards. Page 13 GAO/RCED99-94 Management Improvement Needed in FAA’s AD Program Major Contributors to This Report eResources, Robert W. Shideler, Assistant Director Cbnmunity, and William D. McDowell, Assignment Manager Charles R. Chambers, Evaluator E$onomic Development Division, CarolynKirkpatrick, Evaluator Washington, D. C. -Atlanta Regional Allan C. Richardson, Evaluator-in-Charge Office Cherie M. Starck, Site Senior Janice S. Villar, Evaluator (341203) Page 14 GAO/RCED-99-94 Management Improvement Needed in FAA’s AD Program / I J.S. General Accouut.ing Office Ptrsl Offiw Hex 6015 Gaithersburg, Marylauci 20877 The first, five copies of each rt~port, are free. Additional copies are !$z.oo ca<th. ‘rhthrt: is a 25% discount, on orders for 100 or mm? copies mailtd to a single adciwss. Ordt~rs must, he prepaid by cash or by check or money order made out, to t.ht? Sup~?rint.~ndt?nt. of DocurneutS. r
Aviation Safety: Management Improvement Needed in FAA's Airworthiness Directive Program
Published by the Government Accountability Office on 1990-02-16.
Below is a raw (and likely hideous) rendition of the original report. (PDF)