Aviation Safety: New Airlines Illustrate Long-Standing Problems in FAA's
Inspection Program (Letter Report, 10/17/96, GAO/RCED-97-2).
Pursuant to a congressional request, GAO reviewed the safety performance
of new airlines having 5 or fewer years of operating experience,
focusing on: (1) the frequency with which the Federal Aviation
Administration (FAA) inspects new airlines compared with its inspections
of established airlines; and (2) FAA efforts to correct long-standing
problems that limit the effectiveness of its safety inspection program.
GAO found that: (1) although FAA has taken actions to better target its
inspection resources to the areas with the greatest safety risks, these
efforts have been inadequate; (2) FAA may initiate enforcement actions
in response to apparent or alleged violations of federal aviation
regulations by suspending or revoking airlines' operating certificates,
imposing civil penalties, issuing warning notices and letters of
correction, and issuing emergency revocation orders; (3) FAA initiated
twice as many enforcement actions against new airlines as a group than
it did against established airlines during 1990 through 1994; (4) new
airlines are not required to be inspected any differently from
established airlines according to FAA national inspection guidelines;
(5) FAA has targeted specific airlines and areas of commercial airline
operations for increased surveillance on the basis of a variety of
factors, such as noncompliance with federal aviation regulations,
increased frequency of aircraft incidents, and deteriorating financial
conditions; (6) there is no clear pattern between inspection rates and
the rate of airline incidents or FAA-initiated enforcement actions; and
(7) FAA has hired additional aviation safety inspectors, developed a
comprehensive strategy for upgrading FAA computer tracking and data
systems, and reviewed inspector training and work assignments to
strengthen its inspection operations.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: RCED-97-2
TITLE: Aviation Safety: New Airlines Illustrate Long-Standing
Problems in FAA's Inspection Program
DATE: 10/17/96
SUBJECT: Commercial aviation
Air transportation operations
Airline industry
Transportation safety
Inspection
Human resources training
Aircraft accidents
Aircraft maintenance
Law enforcement
Management information systems
IDENTIFIER: FAA Safety Performance Analysis System
FAA National Aviation Safety Inspection Program
FAA Accident/Incident Data System
FAA Enforcement Information System
FAA ASAS Program Tracking and Reporting Subsystem
FAA ASAS Vital Information Subsystem
FAA Service Difficulty Reporting System
FAA Flight Operations Quality Assurance Initiative
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to Congressional Requesters
October 1996
AVIATION SAFETY - NEW AIRLINES
ILLUSTRATE LONG-STANDING PROBLEMS
IN FAA'S INSPECTION PROGRAM
GAO/RCED-97-2
FAA's Inspection of New Airlines
(341468)
Abbreviations
=============================================================== ABBREV
BTS - Bureau of Transportation Statistics
DOT - Department of Transportation
EIS - Enforcement Information System
FAA - Federal Aviation Administration
FMFIA - Federal Managers' Financial Integrity Act
FOQA - Flight Operations Quality Assurance
GAO - General Accounting Office
NTSB - National Transportation Safety Board
OMB - Office of Management and Budget
OST - Office of the Secretary of Transportation
PTRS - Program Tracking and Reporting Subsystem
SPAS - Safety Performance Analysis Subsystem
Letter
=============================================================== LETTER
B-265984
October 17, 1996
The Honorable James L. Oberstar
Ranking Democratic Member
Committee on Transportation and Infrastructure
House of Representatives
The Honorable John J. Duncan, Jr.
Chairman
The Honorable William O. Lipinski
Ranking Democratic Member
Subcommittee on Aviation
Committee on Transportation and Infrastructure
House of Representatives
The deregulation of the commercial airline industry in 1978 has
stimulated the formation of a significant number of new airlines.
For example, a total of 79 airlines with fewer than 5 years of
operating experience provided scheduled service to the public from
January 1990 through December 1994. In his former capacity as
Chairman of the House Subcommittee on Aviation, Committee on
Transportation and Infrastructure, the current Ranking Democratic
Member of the Committee asked us to review the process of certifying
new airlines and the safety performance of new airlines. As agreed
with the Subcommittee, this report is also addressed to the current
Chairman and Ranking Democratic Member of the House Subcommittee on
Aviation. We conducted our work in two segments to address these
issues. In January 1996, we reported on the certification process,
through which the Department of Transportation (DOT) and the Federal
Aviation Administration (FAA) authorize airlines to begin
operations.\1
This report addresses (1) the safety performance of new airlines
(airlines having 5 or fewer years of operating experience)\2 compared
with that of established airlines (airlines with more than 5 years of
experience) in terms of accidents, incidents, and FAA-initiated
enforcement actions and (2) the frequency with which FAA inspects new
airlines compared with its inspections of established airlines. In
addition to our analysis of new airlines, we also assessed the status
of FAA's efforts to correct long-standing problems that limit the
effectiveness of its overall safety inspection program. Finally,
this report discusses publishing airline-specific safety data for use
by the traveling public.
--------------------
\1 See Certification of New Airlines: Department of Transportation
Has Taken Action to Improve Its Certification Process (GAO/RCED-96-8,
Jan. 11, 1996).
\2 We selected 5 years as the analysis period because it provides
insight into the early years of an airline's development. The period
analyzed was January 1990 through December 1994 (this period provided
the most recent data available at the time of our analysis). We
discussed our definition of new airlines with officials of FAA, DOT,
and the National Transportation Safety Board, none of whom raised any
objection or concern. For additional information on the data
included in this report, see the Objectives, Scope, and Methodology
section in app. I.
BACKGROUND
------------------------------------------------------------ Letter :1
Before commencing operations, new airlines must obtain two separate
authorizations from DOT--"economic" authority from the Office of the
Secretary of Transportation (OST) and "safety" authority from FAA.
Within OST, the Air Carrier Fitness Division is responsible for
assessing whether applicants have the managerial competence,
disposition to comply with regulations, and financial resources
necessary to operate a new airline. FAA's Flight Standards Service
uses a multiphase process to determine whether an applicant's
manuals, aircraft, facilities, and personnel meet federal safety
standards.
Once airlines begin actual operations, FAA is responsible for
monitoring the operations, primarily by conducting safety
inspections. FAA conducts two types of inspections: routine and
special. Routine inspections are generally spot checks performed by
individual inspectors on a periodic basis. FAA's special inspections
complement routine inspections by providing more comprehensive
evaluations of airlines' operations.
To analyze the safety performance of new airlines, we used three sets
of data--data on accidents from the National Transportation Safety
Board (NTSB), FAA's data on incidents, and FAA's data on enforcement
actions initiated against airlines. We discussed the selection of
these data sets with officials from FAA, DOT, and NTSB, who agreed
that they were appropriate for our analysis. However, it should be
noted that all three have limitations. Specifically, some of NTSB's
files on accidents did not definitively specify the airline that was
operating the aircraft; FAA's data on incidents may be subject to
some underreporting; and the data on the number of enforcement
actions initiated, while complete, may reflect differences among FAA
field offices in the emphasis they placed on initiating enforcement
actions. We reviewed and made refinements to these data, where
appropriate, to address these concerns.
NTSB, the official source of information on airline accidents,
defines accidents as events in which individuals are killed or suffer
serious injury, or the aircraft is substantially damaged. By NTSB's
definition, accidents can range from fatal crashes in which the
aircraft is destroyed and all crew and passengers aboard are killed,
to events in which only one person suffers a broken bone and the
aircraft is not damaged, to still others in which there is
substantial aircraft damage, but no fatalities or serious injuries.
FAA generally defines incidents as occurrences other than accidents
associated with the operation of an aircraft that affect or could
affect the safety of operations. Among the commonly recorded types
of incidents are engine malfunctions, system failures, landing gear
collapses, and losses of directional control. Other types of
incidents include collisions with various structures, such as runway
lights, fences, wires, or poles; fires; and in-flight turbulence
resulting in damage to the aircraft or less serious personal injury.
FAA may initiate enforcement actions in response to apparent or
alleged violations of the Federal Aviation Act or federal aviation
regulations. The actions that can be taken under FAA's compliance
and enforcement program include administrative actions, such as
warning notices and letters of correction, and legal enforcement
remedies, such as revoking, suspending, or amending an airline's
operating authority. Examples of violations that can lead to
enforcement actions range from an airline's failure to perform proper
aircraft maintenance to a pilot's failure to maintain the altitude
directed by air traffic control. Another example is a pilot who
possesses a valid pilot certificate but inadvertently pilots an
aircraft without the certificate in his or her possession.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :2
Our analysis showed that during their first 5 years of operations,
new airlines, on average, had higher accident, incident, and
enforcement action rates than established airlines. This does not
mean, however, that new airlines do not provide safe transportation
to the traveling public. Our analysis demonstrates the need for
better targeting FAA's limited inspector resources. For years, we
and others have reported on numerous problems with the inspection
program in such areas as inspector training and the oversight of
aging aircraft. To its credit, FAA has made some progress to correct
its problems, and recent initiatives by DOT and FAA, if implemented,
should go a long way toward strengthening the program.
Although the available data show that both new and established
airlines experience accidents infrequently, we found that, on
average, new airlines had higher accident rates than established
airlines during their early years of operations. From 1990 through
1994, new airlines had an accident rate of 0.60 per 100,000
departures, compared with the established airlines' rate of 0.36 per
100,000 departures. NTSB's definition of accident includes events
ranging from major aviation catastrophes to much less serious
occurrences. As a result, the use and interpretation of accident
data require caution. Of the 201 accidents that occurred from 1990
through 1994, 45 involved fatalities, and 5 of those 45 involved new
airlines.
There were 2,879 incidents and 3,982 FAA-initiated enforcement
actions during the same period, thus providing much more information
for analyzing safety trends. It should be noted that new airlines
begin operations with fewer departures compared to established
carriers (complete data are shown in app. II). As a result, as with
accident data, caution must be exercised in the interpretation of
incident and enforcement data. Rates based on relatively few
departures are susceptible to large fluctuations and may not
accurately predict longer-term performance. Nevertheless, new
airlines experienced an average of 8.1 incidents per 100,000
departures, which was 52 percent higher than the established
airlines' average of 5.4 incidents per 100,000 departures. Included
in these data are the airlines that had no incidents--about half of
the new airlines and one-fifth of the established airlines. During
their second and third years of operations, new airlines, as a group,
experienced elevated incident rates that were twice the average rates
of established airlines. For example, the average incident rate for
new commuter airlines in their third year of operations was 11.6
incidents per 100,000 departures, twice the average incident rate for
established commuter airlines. For new large airlines, the average
incident rate more than tripled between their first and second years
to 12.5 incidents per 100,000 departures, more than twice the average
rate for established new airlines.\3
New airlines, as a group, also had higher rates of FAA-initiated
enforcement actions. FAA initiated 14.9 enforcement actions per
100,000 departures for new airlines, more than twice the rate of
enforcement actions initiated against established airlines (7.3 per
100,000 departures). New airlines experienced their highest number
of FAA-initiated enforcement actions during their first 3 years.
However, most of the enforcement actions initiated during the period
were concentrated among a relatively small group of new airlines, and
over 40 percent of the new airlines had no enforcement actions
initiated against them.
FAA officials told us that they did not know why new airlines as a
group experienced higher incident and enforcement rates, especially
during their early years of operations. However, they theorized that
new airlines may encounter more incidents because their fleets
expanded faster than their organizational ability to absorb growth,
train their staffs, and maintain their fleets. Other factors can
also be a cause for concern and may warrant closer scrutiny. These
include precarious financial conditions (which some new airlines
encountered) or the level at which major functions, such as
maintenance, are contracted out, which can lead to a loss of control
or oversight--a concern that FAA recently acknowledged in its review
of ValuJet Airlines.
FAA's national inspection guidelines in effect during the period of
our review did not call for new airlines to be targeted for increased
surveillance. In actual practice, from January 1990 through December
1994, FAA field offices inspected new large airlines, as a group,
about 3 times as often on average as large established airlines, and
it inspected new commuter airlines at about the same frequency on
average as established commuters. However, no clear pattern in the
inspection rates distinguished those airlines that had relatively
high rates of incidents and enforcement actions from those that had
few or no such problems. Some airlines with high incident and
enforcement rates were inspected less frequently than the average,
while other airlines with no accidents, incidents, or enforcement
actions were inspected more frequently than the average.
For nearly a decade, we have reported on numerous shortcomings in
FAA's aviation safety inspection program, some of which still exist.
These include insufficient training of FAA safety inspectors, the
inadequacy of aviation safety databases, and the need to improve the
oversight of aging aircraft.\4 For example, as early as 1987, we
identified the need for FAA to develop criteria for targeting safety
inspections to those airlines which have characteristics that may
indicate safety problems. We also noted that targeting is important
because of the magnitude of FAA's inspection responsibilities.\5
Although FAA has taken steps to better target its inspection
resources to the areas with the greatest safety risks, these efforts
are still not complete. In 1991, FAA began developing the Safety
Performance Analysis System (SPAS), which draws on information from a
number of safety-related databases to better establish priorities for
FAA's inspections. However, the system is not expected to be fully
operational until 1999. Furthermore, some databases that may provide
source data for SPAS contain incomplete, inconsistent, and inaccurate
data. FAA has developed, but not yet implemented, a data quality
improvement strategy to ensure that these source databases used for
SPAS provide reliable information. Until the reliability of these
databases is improved, the new targeting system will not realize its
full potential to target FAA's resources to high-risk aviation
activities. Moreover, once the reliability of the data is improved,
they could be used to publish airline-specific safety information to
help the traveling public in making transportation decisions.
Even with a resource-targeting system, FAA's inspectors must be
properly trained to effectively carry out their responsibilities. In
1989, we reported that FAA's aviation safety inspectors were not
receiving needed training.\6 On April 30, 1996, we testified that
this problem continued, as some inspectors told us that they had not
been trained on the specific types of aircraft that they were
responsible for inspecting. We also testified that FAA's funding of
technical training had been reduced by 42 percent from fiscal year
1993 through fiscal year 1996.\7 Moreover, the training problem could
become worse in the future as FAA attempts to hire additional
inspectors during a period of constrained budgets.
In the aftermath of the ValuJet accident, DOT and FAA undertook a
number of initiatives related to FAA's inspection program.
Specifically, on May 14, 1996, the Secretary of Transportation, in a
memorandum for the President, outlined efforts to (1) accelerate the
hiring of safety inspectors, (2) strengthen FAA's data collection and
tracking systems, and (3) review FAA's inspection operations,
including inspector training and assignments. On June 18, 1996, the
FAA Administrator initiated a safety review that addressed, among
other things, the certification of new airlines, resource targeting
to address safety risks, newly certificated airlines' operations and
growth, and inspector training and resources. This effort culminated
in a September 16, 1996, report entitled FAA 90 Day Safety Review,
which addressed a number of long-standing problems and made over 30
recommendations. These initiatives, if properly implemented, have
the potential to significantly improve the efficiency and
effectiveness of FAA's safety inspection program.
--------------------
\3 For the purposes of this report, we separated the airlines into
two groups--ones that use large aircraft ("large airlines"), defined
as having a seating capacity of more than 30 persons or a maximum
payload capacity of more than 7,500 pounds. Most of these are jet
aircraft. We refer to the others as "commuter airlines," generally
those that operate smaller aircraft. For additional information on
how we categorized the large and commuter airlines, see app. I.
\4 See Related GAO Products at the end of this report for a list of
prior GAO reports and testimonies on the problems in FAA's inspection
program.
\5 FAA employs about 2,500 aviation inspectors to oversee about 7,300
scheduled commercial aircraft, more than 11,100 charter aircraft,
about 184,400 active general aviation aircraft, about 4,900 repair
stations, slightly more than 600 schools for training pilots, almost
200 maintenance schools, and over 665,000 active pilots.
\6 Aviation Training: FAA Aviation Safety Inspectors Are Not
Receiving Needed Training (GAO/RCED-89-168, Sept. 2, 1989).
\7 Aviation Safety: Targeting and Training of FAA's Safety Inspector
Workforce (GAO/T-RCED-96-26, Apr. 30, 1996).
ACCIDENT, INCIDENT, AND
ENFORCEMENT ACTION RATES
------------------------------------------------------------ Letter :3
The available data show that both new and established airlines
experience accidents infrequently. Nevertheless, from 1990 through
1994, new airlines had an average accident rate of 0.60 per 100,000
departures compared with the established airlines' average rate of
0.36 per 100,000 departures.\8 However, NTSB's definition of accident
can range from fatal crashes in which the aircraft is destroyed and
all crew and passengers aboard are killed, to events where there is
substantial damage to the aircraft but no fatalities or serious
injuries, and to still others where only one person may suffer a
broken bone, but the aircraft suffers no substantial damage. As a
result, the use and interpretation of accident data require caution.
Of the 201 accidents that occurred in 1990 through 1994, 45 involved
fatalities, of which 5 involved new airlines.
Both new and established airlines had a higher number of incidents
and enforcement actions from 1990 through 1994, thus providing much
more information for analyzing safety trends. During 1990 through
1994, there were a total of 2,879 incidents and 3,982 enforcement
actions. Both new large and commuter airlines experienced higher
average rates of incidents and enforcement actions, as a group, than
established large and commuter airlines. In particular, for new
airlines, the rates of incidents and enforcement actions peaked
during their early years of operations. However, there was some
clustering of these events among the new airlines. More than half of
the new airlines had no incidents during the period of our analysis,
and 42 percent of the new airlines had no enforcement actions
initiated against them. Thus, while these rates provide useful
information for analysis, it would not be appropriate to conclude
that new airlines provide unsafe service. (Detailed information on
new airlines' and established airlines' departures, accidents,
incidents, FAA-initiated enforcement actions, and their respective
rates is contained in app. II.)
--------------------
\8 We analyzed data for all new and established airlines that
provided scheduled domestic service during 1990 through 1994 and that
reported data to DOT. We excluded air taxis and other nonscheduled
airlines. Our universe of 262 airlines comprised 29 new large
airlines, 60 large established airlines, 50 new commuters, and 123
established commuters. During the review period, 20 new airlines
reached their sixth year of operations and were then analyzed as
established airlines.
ACCIDENTS
---------------------------------------------------------- Letter :3.1
In 1990 through 1994, NTSB reported 201 accidents by commercial
airlines that provide scheduled service. Most airlines--both new and
established--had no accidents during 1990-94. For example, among the
29 new large airlines in our review, 3 had accidents; the other 26
had no accidents during the 5-year period. Similarly, of the 50 new
commuters, 7 had accidents. Of the 203 established airlines, 69 had
accidents. The remaining 134 had no accidents.
Of the 201 accidents, 45 involved fatalities. These accidents ranged
from 1 accident in which 132 people on board the aircraft were killed
to 12 separate accidents in which 1 person was killed; in 8 of those
12 accidents, the person killed was not on board the aircraft. In
one case, for example, an airline employee was killed after walking
into a rotating propeller blade. The remaining 156 accidents
involved serious injury and/or substantial aircraft damage.
New airlines experienced 13 of the 201 total accidents and 5 of the
45 fatal accidents. The new airlines' accidents resulted in a rate
of 0.60 per 100,000 departures, while the established airlines'
accidents resulted in a rate of 0.36 per 100,000 departures.\9 More
specifically, new large airlines had an accident rate of 1.35 per
100,000 departures, while large established airlines had a rate of
0.30 per 100,000 departures. In contrast, new commuters had an
accident rate of 0.48 per 100,000 departures, while established
commuters had a rate of 0.46 per 100,000 departures.
Aware that the current definition of accident does not distinguish
among the varying degrees of accidents' severity, NTSB and FAA have
undertaken an effort to develop new subclassifications of aviation
accidents. One option that has been explored is to define accidents
according to the significance of damage, recording and grouping data
accordingly. However, according to officials in FAA's Office of
Accident Investigations and NTSB's Office of Research and
Engineering, the results of the joint effort have not yet been
completed, and no completion date has been set.
--------------------
\9 The number of accidents attributed to new and established airlines
do not total to 201 for two reasons: (1) NTSB's accident information
may have contained ambiguous data on the operator of the aircraft
involved or (2) the airline may not have reported departure data for
the period in which the accident occurred. Without departure data
for the airline, we were unable to compute airline-specific accident
rates, and consequently excluded the accident from our totals.
INCIDENTS
---------------------------------------------------------- Letter :3.2
During 1990 through 1994, new large and commuter airlines had
incident rates that were, on average, 52 percent higher than those of
established airlines (overall, a rate of 8.1 incidents per 100,000
departures compared with a rate of 5.4 incidents per 100,000
departures for established airlines). For new large airlines, the
incident rate was over twice that of large established airlines (a
rate of 11.5 incidents per 100,000 departures compared with a rate of
5.1 incidents per 100,000 departures for large established airlines).
The average incident rate for new commuters during 1990 through 1994
was also higher than that of established commuters, although the
difference was not as great. (See table 1.)
Table 1
Summary of Incidents and Incident Rates
for New and Established Airlines,
January 1990 Through December 1994
Establishe Establishe
Category of airline New d New d
---------------------- ---------- ---------- ---------- ----------
Large 34 1,721 11.50 5.13
Commuter 142 982 7.61 5.80
======================================================================
Total 176 2,703 8.14 5.35
----------------------------------------------------------------------
Source: GAO's analysis of FAA's and DOT's data.
As with our analysis of accidents, these rates represent the combined
experiences of the airlines in each of the different categories over
the entire 5-year period. Of the new airlines, 38 (48.1 percent)
experienced at least one incident sometime during 1990-94, while the
other 41 experienced no incidents. Of the new airlines that
experienced incidents, the incident rates ranged from 2.8 to 666.7
incidents per 100,000 departures.\10 Of the 203 established airlines,
162 (79.8 percent) had one or more incidents during the same period,
while the other 41 experienced no incidents.
At certain times during their first 5 years of operations, new
airlines that experienced incidents had rates that greatly exceeded
the average rates for established airlines. For new large airlines,
these times were during their second, fourth, and fifth years of
operations. For example, the rate for new large airlines more than
tripled between their first and second years of operations. Of the
18 new large airlines that had their second year of operations
sometime during 1990 through 1994, 7 (38.9 percent) had incidents.
The other 11 second-year new airlines had no incidents.
In commenting on a draft of this report, DOT noted that one adverse
event for a new airline with a limited number of departures can
significantly affect accident, incident, or enforcement rates. We
agree that because new airlines have fewer departures, the rates at
which they experience problems must be viewed with caution.
Nevertheless, our review included the entire data sets of departures,
accidents, incidents, and enforcement actions for new and established
airlines for a 5-year period, and thus these data are important
pieces of information in FAA's efforts to oversee the airline
industry. The purpose of our analysis of these data was to assess
analytically whether there were differences between new and
established airlines overall that might warrant FAA's increased
oversight of new airlines. Figure 1 shows the change in the incident
rates for new large airlines over their first 5 years of operations
and compares them with the average rate for large established
airlines.
Figure 1: Average Incident
Rates for New Large Airlines
and Large Established Airlines,
by Years of Experience
(See figure in printed
edition.)
Source: GAO's analysis of DOT's and FAA's data.
For new commuters, the average incident rate during their first year
of operations was about the same as for established commuters. But
by their third year of operations, new commuters had an incident rate
that was twice as great as the rate for established commuters (11.6
versus 5.8 per 100,000 departures) and more than twice the rate they
experienced in their first year of operations. (Of the 23 new
commuters that operated for at least 3 years during 1990 through
1994, 10 experienced incidents in their third year.) During the new
commuters' fourth and fifth years of operations, the incident rate
declined. Figure 2 shows the change in the incident rates for new
commuters over their first 5 years of operations and compares them
with the average rate for established commuters.
Figure 2: Average Incident
Rates for New and Established
Commuters, by Years of
Experience
(See figure in printed
edition.)
Source: GAO's analysis of DOT's and FAA's data.
Our analysis did not specifically identify the reasons why new
airlines experienced higher levels of incidents during certain
periods of their first 5 years of operations. We discussed the
results of our analysis with FAA officials. They said that they were
unaware of these trends--they had not done an analysis similar to
ours for new airlines--nor were they aware of any other studies
addressing this issue. Nevertheless, they theorized that new
airlines may encounter more incidents because their fleets expanded
faster than their organizational ability to absorb the growth, train
their staff, and maintain their fleets. Other factors can also be a
cause for concern and may warrant closer scrutiny. These include
precarious financial conditions (which some new airlines encountered)
or the level at which major functions, such as maintenance, are
contracted out, which can lead to a loss of control or oversight--a
concern that FAA recently acknowledged in its review of ValuJet
Airlines.
--------------------
\10 The rate at the high end of the range (666.7 incidents per
100,000 departures) was for an airline that had very few departures.
Specifically, that airline had one incident but only 150 departures.
Other airlines that had relatively high incident rates also had a
relatively low number of departures. Consequently, we aggregated the
data for new large and commuter airlines into groups to deal with the
statistical effects of this phenomenon.
ENFORCEMENT ACTIONS
---------------------------------------------------------- Letter :3.3
FAA's compliance and enforcement program is designed to promote
compliance with both statutory and regulatory requirements. Under
this program, the agency may initiate enforcement actions in response
to apparent or alleged violations of the laws governing federal
aviation or of federal aviation regulations.
Enforcement actions may be initiated on the basis of FAA's inspection
results or on information provided by other sources such as air
traffic controllers or employees in the airline industry.
Enforcement actions include administrative actions, such as warning
notices and letters of correction; legal enforcement remedies, such
as amending, suspending, or revoking airlines' operating
certificates; and punitive actions, such as imposing civil
(financial) penalties and temporarily suspending certificates. For
example, FAA may pursue civil penalties against an airline that
operates aircraft that are not airworthy, repairs equipment using
unacceptable methods, or violates regulations on the transportation
of hazardous materials. When an immediate safety need exists, FAA
inspectors can also issue an emergency revocation order--the most
severe action that can be taken against a domestic airline--to
prevent an airline from conducting flight operations.
In 1990 through 1994, FAA initiated twice the rate of enforcement
actions against new airlines as a group than it initiated against
established airlines.\11
FAA initiated 14.8 enforcement actions per 100,000 departures against
new airlines and 7.3 per 100,000 departures against established
airlines. In addition, just as both new large and commuter airlines
experienced elevated rates of incidents during their early years of
operations, they also experienced higher rates of enforcement actions
during their early years of operations.
FAA initiated considerably higher rates of enforcement actions
against new large airlines, as a group, than it did against large
established airlines. In 1990 through 1994, new large airlines had 8
times more enforcement actions than their established
counterparts--an average of 64.3 actions initiated against them per
100,000 departures compared with 7.8 actions per 100,000 departures
for large established airlines. Figure 3 shows the change in the
rate of enforcement actions initiated against new large airlines
during their first 5 years of operations.
Figure 3: Average Annual
Enforcement Actions That FAA
Initiated Against New Large
Airlines During Their First 5
Years of Operations
(See figure in printed
edition.)
Source: GAO's analysis of DOT's and FAA's data.
Most of the enforcement actions that FAA initiated against new large
airlines were concentrated among relatively few airlines. Of the 190
total enforcement actions initiated against new large airlines during
the period, FAA initiated 141 (74.2 percent) against 10 airlines and
49 against 11 other airlines. FAA initiated no enforcement actions
against eight airlines that were new airlines during the period.
FAA initiated relatively fewer enforcement actions against both new
and established commuters, and the difference in the average number
of enforcement actions initiated was smaller. In 1990 through 1994,
FAA initiated an average of 7.0 enforcement actions against new
commuters per 100,000 departures compared with 6.2 against
established commuters per 100,000 departures.
As with incident rates, new commuters tended to experience rising
rates of enforcement actions until after their third year of
operations. Figure 4 shows the incidence of FAA-initiated
enforcement actions during the new commuters' first 5 years of
operations. FAA initiated an average of 10.7 enforcement actions
against new commuters during their third year of operations--more
than 70 percent higher than the average rate for established
commuters. During the new commuters' fourth and fifth years of
operations, the rate of enforcement actions initiated declined
markedly.
Figure 4: Average Annual
Enforcement Actions That FAA
Initiated Against New Commuters
During Their First 5 Years of
Operations
(See figure in printed
edition.)
Source: GAO's analysis of DOT's and FAA's data.
Similar to the pattern observed for new large airlines, most of the
enforcement actions were initiated against relatively few new
commuters. Of the 130 total actions initiated against new commuters
in 1990 through 1994, FAA initiated 106 (81.5 percent) against 10
airlines; the other 24 enforcement actions were divided among another
15 airlines. FAA initiated no enforcement actions against the
remaining 25 new commuters.
FAA's data reveal that most enforcement cases initiated against
scheduled airlines resulted in administrative actions, rather than
other actions. Of the total 2,286 enforcement cases that had been
initiated in 1993 for which data on final action are available, 1,538
(67.3 percent) concluded with an administrative action, 84 (3.7
percent) concluded with a civil penalty, 79 (3.5 percent) concluded
with a certificate suspension, and 18 (0.8 percent) concluded with a
revocation. In another 567 cases (24.8 percent), FAA took no action.
--------------------
\11 According to FAA, a lag as long as 2 to 3 years may occur between
the time that an enforcement action is opened and the case is closed.
Accordingly, we used actions initiated rather than cases closed as
our measure because cases initiated in 1994 may not be closed.
FAA'S POLICIES DID NOT TARGET
NEW AIRLINES FOR INCREASED
SURVEILLANCE
------------------------------------------------------------ Letter :4
FAA is responsible for promoting safety in air transportation, and
the airlines are responsible for operating their aircraft safely in
compliance with the requirements in title 14 of the Code of Federal
Regulations that cover the aircraft and its systems, maintenance, and
personnel and training. FAA oversees the airlines' programs by
monitoring the safety of all operating airlines and conducting
periodic inspections.
FAA's national inspection guidelines in effect during the period of
our review, which set priorities and established a minimum standard
for the number and type of inspections, did not call for new airlines
to be inspected any differently from established airlines. However,
the guidelines grant latitude to FAA's regional and district offices
to identify the areas that they determine to be important in the
interest of safety. This discretionary surveillance allows
inspectors and their supervisors at FAA's field offices to develop
work programs that can be tailored to their particular environments
and be balanced against such competing priorities as accident
investigations.
Over the years, FAA has targeted specific airlines and areas of
commercial airline operations for increased surveillance on the basis
of a variety of factors. For example, FAA has used an increased
frequency of noncompliance with federal aviation regulations, an
increased frequency of incidents by individual airlines, the
deteriorating financial conditions of individual airlines, and
non-airline-specific attributes (such as aging aircraft) to target
its surveillance activities. However, FAA has not compared the
performance characteristics of new airlines, as a group, with those
of established airlines to determine whether new airlines should be
targeted for increased surveillance. In general, we found that in
1990 through 1994, FAA's field offices inspected new large airlines,
as a group, more frequently than large established airlines. On
average, for large airlines, FAA conducted one inspection for every
20.3 new airline departures and one for every 65.5 established
airline departures. For new commuters FAA conducted, on average, one
inspection for every 113.1 departures and for established commuters,
one inspection for every 107.8 departures. However, there was
considerable variation in the relative frequency with which FAA
inspected individual airlines. At the extremes, the data showed that
a few airlines received more than one inspection for every departure,
while a few others made hundreds of flights between inspections.
FAA's inspection effort also varied widely among the new airlines
that had the greatest average annual number of departures. Of the 10
new large airlines with the highest average number of departures,
inspection rates ranged from once every 8 departures to once every 92
departures. Similarly, of the 10 new commuters with the highest
average number of departures, the data indicate that FAA's inspection
rates ranged from once every 38 departures to once every 340
departures.
We also found no clear pattern between inspection rates and the
airlines' rate of incidents or FAA-initiated enforcement actions.
For example, among the 17 new large airlines responsible for 85
percent of the incidents and enforcement actions in 1990 through
1994, the frequency of inspections varied from one inspection for
every two departures to one inspection for every 66 departures.
Similarly, among the 13 new commuters that accounted for
approximately 80 percent of the incidents and enforcement actions
initiated against that group, the frequency of inspections varied
from one inspection for every 21 departures to one inspection for
every 188 departures. On the other hand, some airlines that had had
no accidents, incidents, or enforcement actions initiated against
them were inspected by FAA once every several hundred departures.
One other, however, was inspected every two departures. More
specifically, of the seven new large airlines that were inspected
less frequently than the average for all new large airlines,
one--ValuJet--had an incident rate that was 40 percent higher than
average, but it was inspected only about one-third as frequently as
all new large airlines through calendar year 1994. For new
commuters, 8 of the 17 that were inspected less frequently than
average had incident or enforcement action rates that were higher
than average. FAA officials told us that the low inspection rates
for new airlines with relatively high problem rates may be due to the
fact that some new airlines, particularly new commuters, may serve
airports that are not closely located to the field office where their
inspectors are assigned.
The recent disclosures about safety problems at ValuJet Airlines and
FAA's oversight of ValuJet illustrate the need for FAA to closely
monitor new airlines. ValuJet began operations in October 1993 with
2 aircraft and expanded its operations to 47 aircraft about 2 years
later. In October 1994, FAA conducted a detailed inspection of
ValuJet and found 35 violations of FAA's air safety regulations. The
two most serious violations--flying an aircraft with broken forward
and aft cargo door locks and flying an aircraft over 140 flights with
a leaking hydraulic line--resulted in a fine of $8,500. In September
1995, FAA conducted another detailed inspection of ValuJet and found
58 violations, including the absence of a continued analysis and
surveillance program, conflicts between the airline's general
maintenance manual and the federal aviation regulations, and the
conduct of maintenance with unapproved procedures.
In February 1996, FAA initiated a "special emphasis program" for
ValuJet. The May 6, 1996, preliminary report on this effort
identified 130 findings on several aspects of ValuJet's operations,
including flight operations training, crew qualifications, manuals
and procedures, and maintenance. After the May 11, 1996, crash,
which killed all 110 passengers and crew, FAA intensified its special
emphasis review through an intensive 30-day review of ValuJet and its
fleet. That review led to a June 1996 consent order, under which
ValuJet agreed to suspend its operations. FAA's announcement of
ValuJet's agreement cited multiple quality assurance shortcomings,
systemwide maintenance deficiencies, the inability to establish the
airworthiness of aircraft, and a lack of engineering capability. On
August 29, 1996, FAA returned ValuJet's operating certificate,
permitting it to resume operations if the airline was found
managerially and financially fit by DOT. On the same day, DOT issued
an order tentatively finding ValuJet fit, willing, and able to
provide domestic scheduled air service. Under agreement with FAA,
upon returning to service, ValuJet would operate a substantially
smaller fleet, starting with up to nine aircraft and adding up to six
more within the following days. ValuJet resumed limited flight
operations on September 30, 1996.
FAA's 90 Day Safety Review recognized that FAA's surveillance system
does not differentiate between established airlines and newly
certificated airlines and stated that additional surveillance during
the first several years of operations is warranted. The safety
review recommended a heightened level of surveillance of newly
certificated airlines for at least the first 5 years of the
companies' operations.
LONG-STANDING PROBLEMS LIMIT
EFFECTIVENESS OF FAA'S
INSPECTION PROGRAM
------------------------------------------------------------ Letter :5
To do its job effectively, and because its resources are limited, FAA
must target its inspectors to the areas of greatest risk. To do so,
FAA needs to have performance-based criteria to gauge various aspects
of aviation safety, and the criteria or measures of safety must be
underpinned by reliable data. Even if FAA inspectors are targeted to
the areas of greatest risk, they must be adequately trained to
effectively carry out their responsibilities. For nearly a decade,
we have reported on long-standing shortcomings in these two areas.
Although FAA has agreed with most of our recommendations and taken
actions to implement them, until all of the these problems are
effectively resolved, the effectiveness of FAA's inspection program
will be limited.
TARGETING SYSTEM REQUIRES
RELIABLE DATA
---------------------------------------------------------- Letter :5.1
In 1987, we reported on the need for FAA to develop criteria for
targeting safety inspections to airlines with characteristics that
may indicate safety problems.\12 In 1991, FAA began designing a
resource-targeting system called the Safety Performance Analysis
System, but it is not yet fully operational. As of August 1996, SPAS
was in place and undergoing operational tests at 47 field offices.
FAA expects the next version of SPAS to be available to inspectors in
late 1997 and the system to be fully operational in 1999.
When fully operational, SPAS could rely on over 25 databases within
FAA, other government agencies, and the aviation industry, including,
potentially, the Improved Accident/Incident Data Subsystem and the
Enforcement Information Subsystem. The current SPAS version uses
four: the Program Tracking and Reporting Subsystem (in which
inspection results are entered), the Vital Information Subsystem
(which contains key data on such items as airlines, pilot and
mechanic schools, and repair stations), the Service Difficulty
Reporting Subsystem (which contains data on instances of abnormal and
potentially unsafe mechanical conditions aboard aircraft), and a
non-FAA database of information and analyses on financial credit
risks.
Building on inspection results and other data, SPAS is intended to
assist FAA in applying its limited inspection resources to those
entities and areas that pose the greatest risk to aviation safety.
The system is also expected to highlight particular types of aircraft
or particular airlines for increased surveillance (inspection) or
oversight if they are experiencing problems at rates that exceed the
averages for that group. Specifically, if problems in a particular
inspection category are found at rates exceeding 50 percent of the
average experience for that group, the SPAS will trigger "advisory"
notifications to the inspector that he or she should look into the
situation. If problems are found at rates exceeding 100 percent of
the average, the system will trigger a notice of "concern" (alert) to
the principal inspectors, who are to respond with a written plan of
action.
In a 1995 report, however, we concluded that SPAS will not be
effective if the quality of its source data is not improved.\13
Specifically, we reported that SPAS may rely on data from numerous
databases that contain incomplete, inconsistent, and inaccurate data.
To address these concerns, we recommended that FAA develop and
implement a comprehensive strategy to improve the quality of those
data. FAA agreed to this recommendation and stated that such a
strategy would be developed by the end of 1995.
In August 1996, FAA reported that this strategy would not be
completed until October 1996. The strategy is to provide clear and
measurable data quality objectives, accurate assessments of the
quality of the current data in each database (including an analysis
and possible redirection of FAA's existing data quality improvement
initiatives), milestones for attaining the stated quality objectives,
and estimates of the resources required. An FAA official said that
implementation would begin immediately afterward. Until FAA
implements its data quality improvement strategy, problems with data
quality may limit SPAS' usefulness and prevent it from realizing its
full potential to target resources to higher-risk activities.
--------------------
\12 See Department of Transportation: Enhancing Policy and Program
Effectiveness Through Improved Management (GAO/RCED-87-3, Apr. 13,
1987).
\13 Aviation Safety: Data Problems Threaten FAA Strides on Safety
Analysis (GAO/AIMD-95-27, Feb. 8, 1995).
INSPECTORS MUST BE
ADEQUATELY TRAINED
---------------------------------------------------------- Letter :5.2
Although FAA management officials told us that inspectors generally
have the experience and basic training necessary to accomplish their
mission, we and others have reported for several years that FAA's
aviation safety inspectors are not receiving needed training. For
example, in 1989 we reported that (1) pilot flight checks were being
made by operations inspectors who had not received recurrent flight
training and whose qualifications to make pilot flight checks had
expired and (2) airworthiness inspectors received only about 50
percent of the training that was planned for them. Recognizing that
some of its employees had received expensive training they did not
need to do their jobs while others did not receive essential
training, in 1992 FAA developed a centralized process to determine,
set priorities for, and fund its technical training needs. This
centralized process is intended to ensure that funds are first
allocated for the training that is essential to fulfilling FAA's
mission. In accordance with this process, each FAA entity has
developed a needs assessment manual tailored to the entity's
activities and training needs. In addition, FAA is also providing
training through such alternative methods as computer-based
instruction, interactive classes televised via satellite, and
computer-based training materials obtained from manufacturers.
Although these initiatives can help improve the efficiency of FAA's
training programs, we testified in 1996 that the adequacy of
inspector training continues to be a concern.\14
During the course of our work on new airlines, we interviewed 37 FAA
inspectors who were involved with the initial certification or
continuing surveillance of new airlines. Although the results of
these interviews are not projectable to the universe of inspectors,
they do indicate a continuing concern among FAA safety inspectors
about the adequacy of the training they receive. Sixteen of the
inspectors said they had gaps in training that affected their
effectiveness in doing their jobs. For example, one inspector
requested training on Airbus aircraft when the airline he inspected
began using that aircraft, but he did not receive the training until
2 years after the airline went out of business. In another case, a
maintenance inspector told us he was responsible for inspecting
several commuter airlines but had never attended maintenance training
school for the types of aircraft he inspects. Instead, FAA sent the
maintenance inspector to training on Boeing 727s and composite
materials, which were not related to the aircraft he was responsible
for. Finally, several inspectors told us that despite their
responsibility to approve global positioning system receivers, a
navigational system increasingly being used in aircraft, they have
had no formal training on this equipment.
We also reported that in fiscal years 1993 through 1996, decreases in
FAA's overall budget have significantly reduced the funding available
for technical training. FAA's overall training budget decreased from
$147 million to $85 million (42 percent) during this period. FAA's
reduced funding for technical training has occurred at a time when it
had received congressional direction to hire over 230 additional
safety inspectors in fiscal year 1996. Because new staff must be
provided with initial training to prepare them to perform their
duties effectively, the cost of this training, combined with overall
training budget reductions, may further constrain FAA's ability to
provide training to its existing inspectors in the future.
--------------------
\14 Aviation Safety: Targeting and Training of FAA's Safety
Inspector Workforce (GAO/T-RCED-96-26, Apr. 30, 1996).
INSPECTION PROGRAM
IDENTIFIED FOR INCREASED
MANAGEMENT OVERSIGHT
---------------------------------------------------------- Letter :5.3
The Federal Managers' Financial Integrity Act of 1982 (FMFIA)
requires that executive agencies prepare an annual statement on the
adequacy of internal controls based on assessments conducted in
accordance with Office of Management and Budget (OMB) Circular A-123.
FMFIA and the circular require that the agency's annual statement on
internal controls include a description of any material weaknesses
(and related plans for corrective actions) identified as part of the
internal control assessment process. Under OMB Circular A-123,
agency managers are requested to use Inspector General reviews and
GAO reports to help them identify and correct deficiencies in
management controls. In addition, the circular states that the
agency should pay particular attention to the views of the agency's
Inspector General in identifying and assessing the relative
importance of deficiencies in management controls. According to
OMB's guidelines, management control weaknesses are material when the
weaknesses meet one or more of the following criteria, among others:
Weaknesses are significant enough to be reported to the President or
the Congress; resources are not being used consistently with the
agency's mission; reliable and timely information is not being
obtained, maintained, reported, and used for decision-making; and a
failure to report a known deficiency may reflect adversely on the
agency.
In December 1993, the DOT Inspector General stated that FAA's
oversight and inspection program represented both a material weakness
and a high-risk area reportable to the President and the Congress.
The Inspector General cited several GAO and Inspector General reports
as the basis for this conclusion and identified the need for FAA to
(1) target inspection resources to areas posing the greatest risks,
(2) accomplish planned/targeted inspections, (3) perform quality
inspections, (4) record deficiencies and ensure that they are
corrected, (5) resolve inspection staff imbalances and retrain or
refocus inspectors where necessary, and (6) enforce certification
requirements relating to aviation parts. The Secretary of
Transportation's 1993 FMFIA report to the President stated that the
DOT Inspector General and GAO had identified deficiencies in some
program areas administered by the FAA (e.g., Aviation Inspection and
Airport Security) and that, taken as a whole, the deficiencies that
were identified may constitute "material weaknesses" in a "high-risk"
area. The report, however, did not identify FAA's oversight and
inspection program as a "high-risk" area. The Secretary stated that
FAA was actively reviewing all of the issues within the context of
FMFIA reporting requirements and that these issues would be reflected
in future FMFIA reports, as appropriate.
In December 1994, the Inspector General again identified FAA's
aviation oversight and inspection activities as a "high-risk" area
and recommended that the Secretary of Transportation include FAA's
safety oversight and inspection activities as a "high-risk" area in
DOT's 1994 FMFIA Report to the President and the Congress. The FAA
Administrator, however, disagreed with the Inspector General's
position, stating that there was an insufficient basis to conclude
that the FAA's safety and inspection program was a "material
weakness" as defined by FMFIA. The Secretary of Transportation's
1994 FMFIA report to the President stated that he continued to be
concerned about ensuring that the aviation oversight and inspection
program meets the highest standards, but did not designate this
program as "high risk," concluding that no new areas of "material
weakness" were reported that year.
For 1995, the DOT Inspector General did not specifically cite FAA's
aviation oversight and inspection activities in her December 1995
letter to the Secretary on FMFIA issues. However, she stated that
past and ongoing work indicated that significant management
weaknesses existed in many of the Department's safety programs and
recommended that safety oversight be reflected in the Secretary's
FMFIA report as a "problem area." An official of the DOT Inspector
General's office told us that a "problem area" is not as serious a
designation as a "high risk" or "material weakness." The Secretary's
1995 FMFIA report, however, did not discuss safety oversight.
Beginning August 1, 1996, OMB no longer requires agencies to
designate "high-risk" areas in their FMFIA reports. Agencies will
still be required, however, to report any "material weaknesses" in
their internal controls. However, as discussed in the following
section, DOT and FAA have recently undertaken a number of initiatives
that, taken together, have the potential to address these concerns.
RECENT INITIATIVES ADDRESS
LONG-STANDING PROBLEMS
------------------------------------------------------------ Letter :6
In a May 14, 1996, memorandum for the President, the Secretary of
Transportation outlined several initiatives to strengthen FAA's
inspection operations. These initiatives included accelerating the
hiring of additional aviation safety inspectors; examining FAA's
computer systems and developing a comprehensive strategy for
upgrading FAA's computer tracking and data systems; and conducting a
comprehensive review of FAA's inspection operations, including
reviewing inspector training and work assignments.
Between May 28 and June 7, 1996, FAA's Flight Standards Service
conducted a self-assessment that looked at various issues, including
the effectiveness of inspector training. A number of recommendations
to improve training resulted from the process, including defining
requirements for the currency and recurrent training needs of safety
inspectors. FAA plans to implement all of these recommendations
within 2 years.
On June 18, 1996, the FAA Administrator initiated a safety review on
"lessons learned" from FAA's oversight experience with ValuJet--the
FAA 90 Day Safety Review. On September 16, 1996, FAA's Deputy
Administrator issued a report that addressed the certification of new
airlines, resource targeting to address safety risks, newly
certificated airlines' operations and growth, contracting out,
inspector training and guidance material, and inspector resources.
The report made over 30 recommendations and included proposed
implementation strategies.
For example, the report noted that FAA could improve its resource
targeting to address safety risks and that the only way to
significantly improve aviation safety is through changing FAA's
methods of assessing risk and using new analysis techniques on more
complete data. The report said that using systems such as SPAS will
allow FAA to more effectively use inspection, surveillance, and
enforcement resources where they are most likely to improve safety.
While recognizing that the inspector workforce is central to FAA's
ability to ensure compliance and maintain a high level of safety, the
report also acknowledged that inspector levels have historically been
understaffed. It also recognized that FAA's training programs do not
always provide the frequency of training or meet the specific needs
identified by employees, managers, and industry. It included
recommendations to ensure that FAA's resources and training are
adequate to meet safety requirements.
As noted in the 90 Day Safety Review, an effective inspection program
requires a stable source of financing. The recently signed Federal
Aviation Reauthorization Act of 1996 creates a National Civil
Aviation Review Commission that will analyze financial needs and
safety trends and make specific recommendations for change. Recent
experience with the lack of authority to collect aviation excise
taxes underscores the need to develop a long-term financing solution
for FAA that will ensure adequate funding of aviation inspectors and
required training.
Similarly, the report concluded that no guidance directs FAA to
maintain heightened surveillance during a new airline's formative
years, when it may be the most unstable. The report recommended
heightened levels of surveillance of newly certificated airlines
during the first 5 years of the companies' operations and periodic
reviews of new airlines that assess management, financial, and
operational capabilities.
The Administrator endorsed the recommendations and called for the
development of a strategy and timetable to implement the recommended
actions. Once implemented, he wrote, these actions will enhance
FAA's ability to target resources more strategically and to respond
more rapidly to changes in the aviation industry.
Following the crash of TWA Flight 800 on July 25, 1996, the President
established a commission headed by the Vice President (commonly known
as the Gore Commission) to review aviation security and safety. The
Commission is scheduled to issue its final report early next year.
In our opinion, these initiatives, taken together, have the potential
to address several of FAA's long-standing problems.
PUBLISHING AIRLINE- SPECIFIC
SAFETY DATA
------------------------------------------------------------ Letter :7
DOT regularly publishes certain consumer-related information on
individual airlines--such as information on on-time performance and
lost luggage. Consumer advocates, academics, and some Members of
Congress have expressed an interest in having FAA publish
airline-specific safety data. The aviation system safety indicators
that FAA already publishes, such as accident rates, incident rates,
near mid-air collisions, and pilot deviations, are aggregated rather
than published on an individual airline basis.
The FAA Administrator and other FAA officials have raised concerns
about the potential negative effect of publishing airline-specific
safety data. For example, under the Flight Operations Quality
Assurance (FOQA) initiative, FAA is encouraging the airlines to
monitor and analyze flight data recorder information to determine
aviation system weaknesses before they become incidents or accidents.
Because the airlines might react negatively to how such data would be
used, FAA officials have said that airlines might be hesitant to
share such information, which would impair FAA's efforts to improve
the system's overall safety. We recognize FAA's desire to obtain
such information from the airlines on a voluntary basis. However,
FAA's mission to promote air safety argues that it should have access
to whatever data that can help it to better improve air safety. If
the airlines do not choose to share such data voluntarily, FAA could
pursue the appropriate regulatory or legislative remedies to gain
such access.
Before publishing airline-specific safety data, FAA would need to
address a number of issues. First, FAA would need to develop a
consensus among the affected and interested parties (airlines,
passengers, aviation safety system analysts, etc.) on the most
appropriate criteria for measuring airline safety performance.
Second, FAA would need to gather and analyze the data and develop a
monitoring system to verify the completeness and accuracy of the
data. Third, FAA would need to take appropriate measures, including
enforcement actions, where necessary to ensure that airlines comply
with data requirements.
While such an endeavor is a formidable task, the benefits could be
substantial. It would not only allow FAA to publicly disclose
airline-specific safety data to help the public in making
transportation decisions but, just as importantly, better equip FAA
to identify and preemptively act on emerging aviation safety trends.
FAA's current effort to develop a strategy to improve the quality of
SPAS databases is an important step that can help solidify the
foundation on which an airline-specific safety analysis and a public
reporting system would potentially be based.
CONCLUSIONS
------------------------------------------------------------ Letter :8
New airlines face a formidable challenge in beginning and sustaining
operations, managing growth, and developing their management and
maintenance infrastructures. The recent disclosures about ValuJet
and FAA's oversight of this airline reinforce this point. Our
analysis of new airlines over a 5-year period shows that, on average,
they experienced higher rates of incidents and FAA-initiated
enforcement actions than established airlines, particularly during
their early years of operations. While such information can be
useful for better targeting FAA's inspection resources, it does not
mean that new airlines are unsafe.
FAA's policies that were in effect during the period of our review
did not call for new airlines to be monitored any differently from
established airlines, and actual inspection rates varied widely among
new airlines--some airlines with high incident and enforcement action
rates were being inspected less frequently than airlines with few or
no such problems. We believe that the basic challenges of starting a
new airline, and the overall results of our analysis, argue for
closely monitoring the performance of new airlines during their first
several years of operations and conducting increased or comprehensive
inspections of those airlines with elevated rates of safety-related
concerns. The recent disclosures about ValuJet reinforce this
argument. FAA's 90 Day Safety Review recommended heightening the
level of surveillance of newly certificated airlines for at least the
first 5 years of the airlines' operations. This recommendation is
consistent with our observations and, if properly implemented, would
largely address our concerns in this area.
On a broader scale, serious problems that hamper the effectiveness of
FAA's aviation safety inspection program have remained unresolved for
nearly a decade. While FAA has taken steps to better target its
inspection resources and has evaluated safety inspector training and
work assignments, concerns in those areas have persisted for years
and a number of unresolved issues remain.
DOT and FAA have recently undertaken a number of initiatives to
address these and other problems, with the FAA 90 Day Safety Review
making over 30 recommendations for improvement. We believe that
these initiatives have the potential to significantly improve FAA's
inspection program, but only if they are effectively implemented. We
believe that, to be effective, DOT's and FAA's implementation
strategy must be underpinned by (1) clear goals and objectives with
measurable performance elements, (2) a monitoring and evaluation
element to measure progress, and (3) a reporting mechanism to keep
the Secretary of Transportation and the Congress informed about
progress and problems.
Resource constraints resulting from budgetary reductions in such
areas as safety inspector training provides a continuing challenge
for FAA. Evaluating the use of and managing existing resources as
efficiently as possible is important given the current tight budget
situation. Such evaluations could also provide the basis for
reprogramming funds to meet critical safety-related needs, or to
justify the need for additional resources should they be found
necessary.
Public concern about the safety of the nation's aviation system has
escalated over the last several months as a result of the ValuJet and
TWA crashes, and several groups have expressed interest in having FAA
publish airline-specific safety data. While FAA would have to
address a number of issues--including gaining consensus on safety
parameters, obtaining and verifying data, and ensuring that airlines
comply with requirements--before publishing such data, we believe
that the time has come for FAA to begin the process that can lead to
publishing such data. One step in this process would involve NTSB's
and FAA's ongoing effort to refine the definition of accident, but
the completion date for this effort has not been established.
RECOMMENDATIONS
------------------------------------------------------------ Letter :9
We recommend that the Secretary of Transportation instruct the
Administrator of FAA to (1) closely monitor the performance of new
airlines, particularly during the early years of operations, and
conduct increased and/or comprehensive inspections of those new
airlines that experience elevated rates of safety-related problems;
(2) evaluate the impact of recent budget reductions on FAA's critical
safety-related functions, including--but not limited to--inspector
training, and report the results to the Congress through the
appropriations process; and (3) study the feasibility of developing
measurable criteria for what constitutes aviation safety, including
those airline-specific safety-related performance measures that could
be published for use by the traveling public.
Furthermore, to ensure the timely and effective implementation of the
recommendations included in FAA's 90 Day Safety Review, we recommend
that the Secretary of Transportation require the Administrator of FAA
to establish (1) clear goals and objectives addressing the safety
review's identified problem areas; (2) measurable performance
criteria to assess how the goals and objectives are being met; and
(3) a monitoring, evaluation, and reporting system so that FAA's
implementation of the recommendations contained in FAA's 90 Day
Safety Review can be reported to the Secretary and the Congress on a
regular basis.
We also recommend that the Chairman of NTSB and the Administrator of
FAA jointly establish a date for completing the ongoing reevaluation
of the definition of accident.
AGENCY COMMENTS AND OUR
EVALUATION
----------------------------------------------------------- Letter :10
DOT and FAA generally agreed with our findings, conclusions, and
recommendations. However, they raised concerns about the statistical
foundation of the report. Specifically, they noted that the number
of accidents, incidents, and departures for new airlines is small in
comparison to the number for established airlines and produces
substantial negative bias in comparing accident and incident rates
for new and established airlines. We agree that accident and
incident rates based on relatively few departures are susceptible to
large fluctuations and may not accurately predict longer-term
performance, and we have noted that prominently in the report.
However, our calculations included 100 percent of these events and
not just a sample and therefore show the actual rates as of the
period of our analysis. The analysis that is of concern to DOT and
FAA provides additional evidence on how FAA might want to target
inspection resources and, therefore, does not impact any of our
conclusions or recommendations.
We have made a number of changes to the report on the basis of the
events that have occurred since the draft was provided to DOT for
comment on September 6, 1996, as well as DOT's written comments.
Most notable among these events was FAA's publication of its 90 Day
Safety Review on September 16, 1996. That review confirmed the
validity of the major issues discussed in our report--the need to
closely monitor the performance of new airlines during their early
years of operations, as well as the need to better target FAA's
resources, improve data quality, and ensure that FAA's resources and
training programs are adequate to meet safety requirements.
Our September 6, 1996, draft of this report contained a proposed
recommendation calling for FAA's aviation safety inspection program
to be designated an area of material weakness in DOT's Federal
Managers' Financial Integrity Act report. In light of the fact that
FAA's 90 Day Safety Review recognized the long-standing concerns that
gave rise to our proposed recommendation and made over 30
recommendations that, if properly implemented, have the potential to
correct these problems, we have deleted that recommendation from our
final report. However, we believe there is a need for continued
vigilance on the part of DOT, FAA, and the Congress to ensure that
the recommendations in the 90 Day Safety Review are effectively
implemented in a timely manner. Consequently, we have added a
recommendation that calls for FAA to report periodically to the
Secretary of Transportation and the Congress on its progress in
implementing the recommendations from the 90 Day Safety Review.
A copy of DOT's comments is included as appendix III.
--------------------------------------------------------- Letter :10.1
We conducted our review from August 1995 through September 1996 in
accordance with generally accepted government audit standards. A
detailed discussion of our objectives, scope, and methodology appears
in appendix I.
We will send copies of this report to the Secretary of
Transportation; the Administrator, FAA; the Chairman, NTSB; the
Director, Office of Management and Budget; and other interested
parties. We will also make copies available on request. This report
was prepared under the direction of John H. Anderson, Jr., Director,
Transportation Issues, who can be reached at (202) 512-2834 if you
have any questions. Other major contributors to this report are
listed in appendix IV.
Keith O. Fultz
Assistant Comptroller General
OBJECTIVES, SCOPE, AND METHODOLOGY
=========================================================== Appendix I
The former Chairman, Subcommittee on Aviation, House Committee on
Public Works and Transportation, asked us to examine, as the second
segment of work addressing issues concerning the Federal Aviation
Administration's (FAA) oversight of new airlines, the agency's
efforts to ensure that new airlines meet safety standards. As agreed
with the Subcommittee's staff, we also addressed this report to the
current Chairman and Ranking Democratic Member of the Subcommittee on
Aviation.
To address this issue, we focused on three questions: Did new
airlines perform differently from established airlines during the
5-year period between January 1, 1990, and December 31, 1994, with
regard to accidents, incidents, and enforcement actions?\1 At what
frequency does FAA inspect new airlines compared with established
airlines? And what impediments hinder the effectiveness of FAA's
overall safety inspection program?
Before we were able to answer the first question, we had to determine
which airlines were "new airlines." We defined a new airline as one
that provided scheduled domestic air service for 5 or fewer years at
any time from the beginning of 1990 to the end of 1994. For example,
an airline that began service in 1994 would be considered a new
airline, since its first year of operations was within the study
period. Similarly, an airline that began operating in 1986 would
also be considered a new airline in our analysis of 1990 data,
because that airline's fifth year of operations occurred in 1990.
However, beginning with the analysis of 1991 data, that same
airline's operations would then be included in the comparison group
of established airlines--those that had provided scheduled domestic
service for more than 5 years during the 1990-94 period. Thus, we
considered any airline that began scheduled operations between
January 1986 and December 1994 to be a new airline during relevant
portions of the 1990-94 period.
This definition of new airline differs from that normally applied in
other aviation safety research. Those studies have tended to define
new airlines as being airlines that began interstate operations
following the Airline Deregulation Act of 1978. However, airlines
such as Southwest Airlines that began interstate operations
immediately after that act have now been operating for nearly two
decades. We believe that a review that focuses more on airlines with
considerably fewer years of experience would provide more insight
into the safety performance of new airlines. We discussed our
definition of new airlines with FAA, the Department of Transportation
(DOT), and the National Transportation Safety Board (NTSB), none of
whom raised any objection or concern.
To determine which specific airlines should be designated as new
airlines and which should be designated as established airlines, we
reviewed records from DOT's Airline Fitness Division within the
Office of the Secretary of Transportation (OST), the Bureau of
Transportation Statistics (BTS), and FAA to develop a list of
airlines subdivided into large and commuter new airlines and large
and commuter established airlines. First, we obtained historical
information from OST's files on airlines that it had found "fit" and
to which DOT had issued operating authority. We initially included
as new airlines those that OST had recertificated following a
substantial change in their operations. Second, we used industry
financial and operating records from BTS to help determine the year
in which airlines began scheduled operations, and divided the airline
list into "new" and "established" by the year indicated in the
records.\2 Because none of the automated databases we analyzed
recorded any specific distinction between scheduled commuter airlines
and on-demand air taxi services (i.e., chartered airlines), we relied
on FAA officials to provide this distinction. As a result, we
eliminated on-demand airlines from our list. However, some commuters
that operated as both commuters and on-demand airlines at different
points during our 5-year period are included among our group of
established commuters. BTS and FAA verified our airline lists.
At FAA's suggestion, we made two additional adjustments to our list
of new airlines. First, we reclassified as established airlines
those airlines that DOT had newly authorized to provide scheduled
service at some point between 1986 and 1994 but which had earlier
operated as on-demand air taxis. Second, we reclassified as
established airlines those that DOT had recertificated following a
substantial change of operations. FAA suggested that those airlines
should be considered established because they had essentially
maintained an unbroken chain of operations from a previous status.
To determine which airlines to categorize as "large" or as
"commuters," we analyzed information from OST, BTS, and FAA. OST and
BTS use definitions of large and commuter aircraft that differ from
FAA's. According to DOT's regulations, a large certificated airline
is one that holds a certificate issued under section 401 of the
Federal Aviation Act of 1958 and that operates aircraft designed to
have a maximum passenger seating capacity of more than 60 seats or a
maximum payload capacity of more than 18,000 pounds, or that conducts
international operations. Small certificated airlines and commuter
airlines ("commuters") generally operate only aircraft with 60 seats
or fewer or a payload capacity of 18,000 pounds or less. FAA's
definitions follow the distinction made by parts 121 and 135 of the
Federal Aviation Regulations, which basically define an aircraft as
"large" or as a "commuter" depending upon whether or not it seats
more than 30 passengers.
While we relied on FAA to indicate exactly which airlines it
considered to be commuters, our distinction between large and
commuter airlines was also consistent with DOT's definitions. This
occurred because FAA's list of commuter airlines included not just
those that operated "part 135" aircraft exclusively, but also
airlines that operated "part 121" aircraft ("split certificate"
airlines). According to information from FAA, those airlines' part
121 aircraft were turboprop aircraft, such as the De Havilland
Dash-8, that may seat between 36 and 56 passengers. FAA's list of
large airlines included only airlines that exclusively operated large
aircraft. Most of those large airlines operated jet aircraft in
1994.
As a result, we analyzed data for all new airlines and established
airlines that provided scheduled domestic service during the 1990
through 1994 period and that reported data to DOT. We excluded air
taxis and other airlines providing nonscheduled service. Our
universe of 265 airlines comprised 29 new large airlines, 60 large
established airlines, 50 new commuters, and 123 established
commuters. During the review period, 20 new airlines reached their
sixth year of operations and were then analyzed as established
airlines.
To answer the first question regarding the airlines' experiences with
accidents, incidents, and enforcement actions, we analyzed three
different sets of data. First, to analyze data on all airline
accidents that occurred from 1990 through 1994, we reviewed
information from NTSB, the official source of information on airline
accidents. Some of NTSB's accident data included ambiguous
information about the airline operator's identity. To resolve the
uncertainty, we reviewed more extensive information on each accident
in question. Still, of the 201 accidents that occurred from 1990
through 1994, for 8 accidents we were unable to determine with
certainty which company operated the aircraft involved. For example,
NTSB's files include information on a commuter airline accident in
January 1991 involving US Air Express. However, more than one
airline company conducts business as US Air Express, and because NTSB
did not record the airline's designator code, which FAA assigns to
individual operators, we were unable to assign the accident to any
specific company.
Second, we analyzed FAA's data on aviation incidents that occurred
during the period. FAA records data on various airline incidents,
which the agency defines as an occurrence other than an accident
associated with the operation of an aircraft, that affects or could
affect the safety of operations. To improve the data's reliability
and the relevance of the analysis, we excluded certain categories of
incidents clearly outside the control of the airline, such as birds'
being ingested into jet engines and lightning strikes. We made these
changes at the suggestion, and with the assistance, of FAA.
Third, we analyzed data on enforcement investigations initiated from
FAA's Enforcement Information System (EIS). EIS includes information
on all enforcement actions taken by FAA, whether administrative or
legal. FAA's Assistant Chief Counsel processes reports requiring
legal enforcement action or referral for possible criminal
investigation and prosecution. Because such actions may take years
to conclude (for example, FAA closed its last enforcement actions
against Eastern Air Lines in August 1995, although Eastern ceased
operations in January 1992), we used the actions initiated to measure
enforcement activity.
We did not assess the reliability of the incident or enforcement
data. However, we discussed the issue with FAA officials, who told
us that while there may be omissions in these data, they were the
best available for the purposes of our review. For example, the
officials told us that although FAA's incident data may be subject to
some underreporting, those data were preferable to NTSB's airline
safety incident data, because NTSB exercises great discretion in
deciding which events to investigate. Similarly, the data on the
number of enforcement actions initiated, while complete, may be
underreported because of differences in how FAA field offices
implement the agency's enforcement program. That is, confronted with
similar sets of factual circumstances, some field offices may
recommend that FAA initiate an enforcement action while others would
not.
To provide the basis for comparing the number of accidents,
incidents, and enforcement actions across airlines, we divided all
such data points by a base of 100,000 (domestic) departures, a common
comparative measure of aviation safety. According to FAA and NTSB,
since most accidents occur during arrival or departure, the number of
departures is considered to be the best normalizing variable. We
obtained the departure data from BTS, which received those data
directly from individual airlines. However, we did not independently
verify the data sent by the airlines or review BTS' procedures for
handling those data. Also, in our calculations of the various rates
for each group of airlines, we included data on accidents, incidents,
and enforcement actions only if an airline also reported departure
data for that year. For example, Eastern Air Lines stopped reporting
departure data to BTS in 1991; however, FAA's data indicate that it
initiated an enforcement action against Eastern in 1992. Our
calculations of the enforcement rate for large established airlines
did not include that 1992 action against Eastern.
We analyzed accidents, incidents, and enforcement actions of new
airlines by years of operating experience. Such an analysis compares
the records of airlines with the same number of years of operations,
regardless of the calendar year in which the observation occurred.
For example, we compared airlines within their second year of
operations, whether that year was 1990 or 1993, against those with
fewer and more years of experience. This method focuses on examining
the airline's records over time, as the airlines gain operating
experience.
To answer the second question on the relative level of surveillance
applied to new airlines and established airlines during the 1990-94
period, we compared the number of inspections of new airlines to the
number of inspections of established airlines, normalized for
departures in each year. We obtained those data from FAA's Program
Tracking and Reporting Subsystem (PTRS). We have long reported on
problems with the data in FAA's safety inspection management
system.\3 Because of continuing concerns about the reliability of the
data on inspection results, we used the PTRS data only to determine
the number of inspections done, and not their outcomes. We also
reviewed the national program guidelines for airline surveillance and
spoke to responsible FAA officials to determine whether FAA
distinguished between new and established airlines in its
surveillance and inspection efforts.
To answer the third question, we reviewed GAO products, both reports
and testimonies published over the last decade, reporting on many
aspects of FAA's aviation safety inspection program. To assess FAA's
progress in addressing the problems that were discussed in those
reports and testimonies, we reviewed documentation that monitors the
extent of FAA's implementation of GAO's recommendations.
After completing our analysis, we discussed our preliminary findings
with officials of FAA and NTSB. We also provided a draft of our
report to DOT for its review and comment. The agency's letter in
response is reproduced in appendix III. We performed our work
primarily at FAA headquarters in Washington, D.C., from August 1995
through September 1996 in accordance with generally accepted
government auditing standards.
--------------------
\1 We selected the period 1990 through 1994 for the analysis because
those were the latest years for which we could obtain complete data
for all accidents, incidents, and enforcement measures across all of
the databases used.
\2 Some airlines, though certified for operations by DOT, do not
begin operating for several months afterwards. Others never fly;
after a year of not having begun operations, those airlines may have
their certifications revoked by DOT for dormancy.
\3 See, for example, Aviation Safety: FAA's Safety Inspection
Management System Lacks Adequate Oversight (GAO/RCED-90-36, Nov. 13,
1989), Aviation Safety: Problems Persist in FAA's Inspection Program
(GAO/RCED-92-14, Nov. 20, 1991), and Aviation Safety: Data Problems
Threaten FAA Strides on Safety Analysis System (GAO/AIMD-95-27, Feb.
8, 1995).
DEPARTURES, ACCIDENTS, INCIDENTS,
AND FAA-INITIATED ENFORCEMENT
ACTIONS FOR LARGE AND COMMUTER NEW
AIRLINES AND ESTABLISHED AIRLINES
========================================================== Appendix II
Years of
operatin
g Enforcem Enforcem
Category of experien Departur Accide Incide ent Accide Incide ent
airline ce es nts nts actions nts nts actions
----------------- -------- -------- ------ ------ -------- ------ ------ --------
New large 1 27,030 0 1 23 0.00 3.70 85.09
2 112,435 1 14 90 0.89 12.45 80.05
3 115,325 0 9 42 0.00 7.80 36.42
4 14,826 2 5 24 13.49 33.72 161.88
5 26,021 1 5 11 3.84 19.22 42.27
=========================================================================================
Subtotal 295,637 4 34 190 1.35 11.50 64.27
=========================================================================================
Established large Subtotal 33,539,7 102 1,721 2,610 0.30 5.13 7.78
48
New commuter 1 196,631 1 11 13 0.51 5.59 6.61
2 331,735 1 28 26 0.30 8.44 7.84
3 421,158 4 49 45 0.95 11.63 10.68
4 377,405 2 28 23 0.53 7.42 6.09
5 539,073 1 26 23 0.19 4.82 4.27
=========================================================================================
Subtotal 1,866,00 9 142 130 0.48 7.61 6.97
2
=========================================================================================
Established Subtotal 16,943,5 78 982 1,052 0.46 5.80 6.21
commuter 88
All new airlines 1 223,661 1 12 36 0.45 5.37 16.10
2 444,170 2 42 116 0.45 9.46 26.12
3 536,483 4 58 87 0.75 10.81 16.22
4 392,231 4 33 47 1.02 8.41 11.98
5 565,094 2 31 34 0.35 5.49 6.02
=========================================================================================
Total 2,161,63 13 176 320 0.60 8.14 14.80
9
=========================================================================================
All established Total 50,483,3 180 2,703 3,662 0.36 5.35 7.25
36
-----------------------------------------------------------------------------------------
Note: Includes departure, accident, incident, and enforcement action
data for deregulated all-cargo airlines and commercial operators of
aircraft when those operations or events occurred during operations
under either 14 C.F.R. 121 or 14 C.F.R. 135.
Source: GAO's analysis of data from DOT, FAA, and NTSB.
(See figure in printed edition.)Appendix III
COMMENTS FROM THE DEPARTMENT OF
TRANSPORTATION
========================================================== Appendix II
(See figure in printed edition.)
(See figure in printed edition.)
(See figure in printed edition.)
MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV
Charles Barchok, Jr.
Gerald L. Dillingham
Peter J. Espada
Curtis L. Groves
Julian L. King
Steven C. Martin
RELATED GAO PRODUCTS
============================================================ Chapter 0
Aviation Safety: Targeting and Training of FAA's Safety Inspector
Workforce (GAO/T-RCED-96-26, Apr. 30, 1996).
FAA Budget: Issues Related to the Fiscal Year 1996 Request
(GAO/T-RCED/AIMD-95-131, Mar. 13, 1995).
Aviation Safety: Data Problems Threaten FAA Strides on Safety
Analysis System (GAO/AIMD-95-27, Feb. 8, 1995).
Aviation Safety: FAA Can Be More Proactive in Promoting Aviation
Safety (GAO/T-RCED-95-81, Jan. 12, 1995).
Aviation Safety: FAA's Efforts to Improve Oversight of Foreign
Carriers (GAO/T-RCED-95-33, Oct. 4, 1994).
FAA Technical Training (GAO/RCED-94-296R, Sept. 26, 1994).
Aviation Safety: Unresolved Issues Involving U.S.-Registered
Aircraft (GAO/RCED-93-135, June 18, 1993).
Aircraft Maintenance: FAA Needs to Follow Through on Plans to Ensure
the Safety of Aging Aircraft (GAO/RCED-93-91, Feb. 26, 1993).
Aviation Safety: Increased Oversight of Foreign Carriers Needed
(GAO/RCED-93-42, Nov. 20, 1992).
Aviation Safety: Additional Actions Needed for Three Safety Programs
(GAO/T-RCED-92-90, Aug. 4, 1992).
Aviation Safety: Commuter Airline Safety Would Be Enhanced With
Better FAA Oversight (GAO/T-RCED-92-40, Mar. 17, 1992).
Aviation Safety: Better Oversight Would Reduce the Risk of Air Taxi
Accidents (GAO/T-RCED-92-27, Feb. 25, 1992).
Aviation Safety: FAA Needs to More Aggressively Manage Its
Inspection Program (GAO/T-RCED-92-25, Feb. 6, 1992).
Aviation Safety: Air Taxis--The Most Accident-Prone Airlines--Need
Better Oversight (GAO/RCED-92-60, Jan. 21, 1992).
Aviation Safety: Problems Persist in FAA's Inspection Program
(GAO/RCED-92-14, Nov. 20, 1991).
Aviation Safety: Emergency Revocation Orders of Air Carrier
Certificates (GAO/RCED-92-10, Oct. 17, 1991).
Aging Aircraft Maintenance: Additional FAA Oversight Needed
(GAO/T-RCED-91-84, Sept. 17, 1991).
Aircraft Maintenance: Additional FAA Oversight Needed of Aging
Aircraft Repairs (GAO/RCED-91-91A and B, May 24, 1991).
Aviation Safety: Limited Success Rebuilding Staff and Finalizing
Aging Aircraft Plan (GAO/RCED-91-119, Apr. 15, 1991).
Serious Shortcomings in FAA's Training Program Must Be Remedied
(GAO/T-RCED-90-86, June 6, 1990).
Staffing, Training, and Funding Issues for FAA's Major Work Forces
(GAO/T-RCED, 90-42, Mar. 14, 1990).
Aging Aircraft: FAA Needs Comprehensive Plan to Coordinate
Government and Industry Actions (GAO/RCED-90-75, Dec. 22, 1989).
Aviation Safety: FAA's Safety Inspection Management System Lacks
Adequate Oversight (GAO/RCED-90-36, Nov. 13, 1989).
Meeting the Aging Aircraft Challenge: Status and Opportunities
(GAO/T-RCED-90-2, Oct. 10, 1989) and (GAO/T-RCED-89-67, Sept. 27,
1989).
Aviation Training: FAA Aviation Safety Inspectors Are Not Receiving
Needed Training (GAO/RCED-89-168, Sept. 14, 1989).
Aviation Safety: FAA Has Improved Its Removal Procedures for Pilot
Examiners (GAO/RCED-89-199, Sept. 8, 1989).
FAA Staffing: Recruitment, Hiring, and Initial Training of
Safety-Related Personnel (GAO/RCED-88-189, Sept. 2, 1988).
Aviation Safety: Measuring How Safely Individual Airlines Operate
(GAO/RCED-88-61, Mar. 18, 1988).
Aviation Safety: Needed Improvements in FAA's Airline Inspection
Program Are Under Way (GAO/RCED-87-62, May 19, 1987).
Department of Transportation: Enhancing Policy and Program
Effectiveness Through Improved Management (GAO/RCED-87-3, Apr. 13,
1987).
Compilation and Analysis of the Federal Aviation Administration's
Inspection of a Sample of Commercial Air Carriers (GAO/RCED-85-157,
Aug. 2, 1985).
*** End of document. ***