Military Aircraft Safety: Significant Improvements Since 1975 (Briefing
Report, 02/01/96, GAO/NSIAD-96-69BR).

Pursuant to a congressional request, GAO reviewed military aircraft
accidents involving fatalities or severe aircraft damage, also known as
Class A flight mishaps, focusing on: (1) historical trends in Class A
mishaps; (2) the causes of such mishaps and how the causes are
investigated; and (3) actions taken to avoid Class A mishaps.

GAO found that: (1) military aircraft were involved in 3,828 Class A
mishaps between fiscal years (FY) 1975 and 1995, which resulted in 3,810
fatalities and 3,483 destroyed aircraft; (2) between FY 1975 and 1995,
the annual number of mishaps decreased from 309 to 76, fatalities
decreased from 285 to 85, and the mishap rate decreased from about 4.3
accidents per 100,000 flying hours to 1.5; (3) the value of Class A
losses has remained between $1.2 billion and $1.6 billion over the past
6 years; (4) the Department of Defense requires the military services to
report and investigate Class A mishaps, but each service has its own
investigation reporting requirements; (5) the independence of accident
investigators may be compromised, since some investigators are tied to
the mishap command; (6) human error contributed to 73 percent of Class A
mishaps in FY 1994 and 1995; (7) it is difficult to draw correlations
between causes of aviation accidents and flight mishap rates due to the
low incidence of safety mishaps; and (8) actions the military services
have taken to reduce mishaps include tracking mishap investigation
recommendations, distributing safety information, and conducting risk
management studies in order to provide guidance.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  NSIAD-96-69BR
     TITLE:  Military Aircraft Safety: Significant Improvements Since 
             1975
      DATE:  02/01/96
   SUBJECT:  Military aircraft
             Aircraft accidents
             Investigations by federal agencies
             Accident prevention
             Reporting requirements
             Property losses
             Transportation statistics
             Transportation safety
             Air transportation operations
IDENTIFIER:  OH-58 Helicopter
             
******************************************************************
** 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


Briefing Report to the Rfanking Minority Member, Subcommittee on
Military Procurement

February 1996

MILITARY AIRCRAFT SAFETY -
SIGNIFICANT IMPROVEMENTS SINCE
1975

GAO/NSIAD-96-69BR

Military Aircraft Safety

(703108)


Abbreviations
=============================================================== ABBREV

  DOD - Department of Defense

Letter
=============================================================== LETTER


B-270647

Letter Date Goes Here

The Honorable Ike Skelton Ranking Minority Member Subcommittee on
Military Procurement Committee on National Security House of
Representatives

Dear Mr.  Skelton: 

In May 1995, you noted that a number of military aircraft accidents
had occurred over a period of a few weeks, resulting in the death of
more than a dozen crew members and passengers.  You requested that we
conduct a review of military aircraft accidents.  Our objectives were
to identify (1) historical trends in aircraft accidents involving
deaths or extensive aircraft damage (Class A flight mishaps), (2)
investigations performed to determine the causes, and (3) examples of
actions taken to reduce the number of aviation accidents.  We also
analyzed investigation summaries to identify the primary factors
contributing to mishaps and reviewed studies which addressed the
relationship of operating tempo to aviation safety.  As agreed, we
did not address your concern about alleged mishandling of mishap
investigations in the Air Force because the Department of Defense
(DOD) Inspector General was already examining those allegations. 
This report reflects the information in our briefing to you on
January 22, 1996. 


   BACKGROUND
------------------------------------------------------------ Letter :1

A measure of aviation safety within DOD is the mishap rate--number of
Class A flight mishaps per 100,000 flying hours.  DOD defines a Class
A flight mishap as one involving a DOD aircraft with an intent to
fly, that resulted in damages totaling $1 million or more, a
destroyed aircraft, a fatality, or a permanent total disability.  DOD
requires that Class A mishaps be investigated so that causes can be
identified and corrective actions taken to prevent future
occurrences. 

Service safety centers\1 play a key role in maintaining aviation
mishap statistics, establishing safety policies, disseminating safety
information, reviewing mishap investigation reports, tracking
recommendations, and performing safety studies.  In addition, the
safety centers analyze trends to identify potential safety hazards. 


--------------------
\1 These centers are the Naval Safety Center, Norfolk, Virginia; Air
Force Safety Agency, Kirtland Air Force Base, New Mexico; and the
Army Safety Center, Ft.  Rucker, Alabama.  In addition to its
involvement in Navy safety, the Naval Safety Center also monitors
investigations of Marine Corps aviation mishaps. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

DOD aviation safety has improved significantly over the last 21
years.  Between fiscal years 1975 and 1995 for example, the annual
number of Class A mishaps decreased from 309 to 76, while the number
of fatalities decreased from 285 to 85.  During this period, Class A
mishaps per 100,000 flying hours, referred to as the mishap rate,
also decreased from about 4.3 to 1.5.  The value of Class A losses
remained fairly constant over the last 6 years, ranging from a high
of about $1.6 billion in fiscal year 1993 to a low of $1.2 billion in
fiscal year 1994. 

Although DOD requires that the services report and investigate Class
A aviation mishaps, service requirements differ as to who convenes
the investigation board and who participates as voting members.  For
example, until recently only the Army required safety center
investigators as voting board members.  Past GAO and Air Force
studies have questioned the independence of the investigators because
of their organizational ties to the mishap command.  The Air Force
has recently directed changes to enhance the independence of its
investigations. 

Each of the services have taken steps to reduce aviation mishaps,
such as tracking mishap investigation recommendations and
disseminating safety information in manuals, newsletters, videos, and
messages.  Recent safety initiatives include risk management and
human factor studies.  The Army, for example, is developing a series
of profiles for predicting whether an aviation training mission is
low, medium, or high risk.  A subsequent system will provide guidance
for assessing operational risks and reducing them to acceptable
levels. 

In fiscal years 1994 and 1995, human error was reported as a
contributing cause in 73 percent of the Class A flight mishaps. 
However, in a 1994 report,\2 examining its historical flight mishap
data, the Air Force found no direct correlation between operating
tempo and safety mishaps.  In 1995, the Air Force Blue Ribbon panel
reported some evidence connecting pace of operations to aviation
safety.  Service statisticians, however, told us that the relatively
low incidence of Class A flight mishaps makes it difficult to draw
inferences and identify statistical correlations of variables with
mishap rates. 


--------------------
\2 Safety Challenge:  Identifying and Addressing Recent Trends in
USAF Flight Mishaps, Office of the Chief of Safety, Headquarters,
United States Air Force, September 1994. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :3

In commenting orally on a draft of this report, officials from the
Office of the Secretary of Defense, Air Force, Army, and Navy
generally concurred with the findings.  The Air Force provided
documentation showing that the recommendations of the Blue Ribbon
Panel were being implemented.  Other recommended technical changes
were incorporated throughout the report as deemed appropriate. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :4

We obtained and analyzed annual statistics, beginning in 1975, on the
number of Class A flight mishaps, fatalities, destroyed aircraft, and
dollar losses.  We compared the services' data, validated mishap
rates, and documented trends.  We also analyzed mishap investigation
summaries to document causes contributing to flight mishaps. 

We reviewed agency instructions and procedures regarding flight
mishap investigations, interviewed safety officials, and reviewed
final reports.  We observed an Air Force safety investigation,
discussed the investigation with board members, and attended the
outbrief at the major command.  We did not observe the board's
deliberations.  We interviewed National Transportation Safety Board
officials and reviewed their investigation procedures to provide some
comparison with DOD's safety investigations.  We reviewed studies
dealing with the independence of investigations and the possible
relationship of operating factors to aircraft accident rates.  We
also identified service initiatives to reduce aviation accidents. 

We conducted our work from June 1995 to January 1996 in accordance
with generally accepted government auditing standards. 

As arranged with your staff, unless you publicly announce its
contents earlier, we plan no further distribution of this report
until 30 days from its issue date.  At that time, we will send copies
to other interested congressional committees and Members of Congress,
and the Secretaries of Defense, the Army, the Navy, and the Air
Force.  We will also make copies available to other interested
parties on request. 

The major contributors to this report are identified in appendix I. 
If you or your staff have any questions concerning this report,
please contact me on (202) 512-5140. 

Sincerely yours,

Mark E.  Gebicke Director, Military Operations and Capabilities


Briefing Section I BACKGROUND
============================================================== Letter 


   CRITERIA FOR CLASSES OF FLIGHT
   MISHAPS
------------------------------------------------------------ Letter :5



   (See figure in printed
   edition.)

Flight mishaps involve any reportable damage to an aircraft that is
preparing to fly, in flight, or completing a landing.  Flight mishaps
are classified by DOD according to the severity of resulting injury
or property damage.  Class A mishaps involve damage of $1 million or
more, a destroyed aircraft, or a fatality or permanent total
disability.  The remaining classes of mishaps are distinguished
primarily by their loss value and severity of injury:  Class B
accidents involve damage of $200,000 to less than $1 million,
permanent partial disability, or inpatient hospitalization of five or
more people; Class C accidents involve damage of $10,000 to less than
$200,000, or a lost-time injury; and Class D accidents involve damage
of less than $10,000. 

Our review focused on Class A flight mishaps only. 


Briefing Section II AIRCRAFT
MISHAP TRENDS
============================================================== Letter 


   DECREASE IN NUMBER OF CLASS A
   FLIGHT MISHAPS (FISCAL YEARS
   1975-95)
------------------------------------------------------------ Letter :6



   (See figure in printed
   edition.)

Between fiscal years 1975 and 1995, military aircraft were involved
in 3,828 Class A mishaps, which resulted in 3,810 fatalities and
3,483 destroyed aircraft. 

The annual number of DOD Class A flight mishaps decreased from 309 in
fiscal year 1975 to 76 in fiscal year 1995.  The Navy/Marine Corps
reduced the number from 158 to 34, the Air Force from 99 to 32, and
the Army from 52 to 10. 


   DECREASE IN DOD CLASS A MISHAP
   RATE (FISCAL YEARS 1975-95)
------------------------------------------------------------ Letter :7



   (See figure in printed
   edition.)

DOD's Class A mishap rate, calculated as the number of accidents per
100,000 flying hours, declined from about 4.3 in fiscal year 1975 to
1.5 in fiscal year 1995. 


   DECREASE IN SERVICE CLASS A
   FLIGHT MISHAP RATES (FISCAL
   YEARS 1975-95)
------------------------------------------------------------ Letter :8



   (See figure in printed
   edition.)

As with the number of mishaps, each of the services has also
experienced an overall downward trend in its mishap rate since fiscal
year 1975.  In particular, the Navy/Marine Corps mishap rate dropped
significantly from 7.3 mishaps per 100,000 flying hours in fiscal
year 1975 to 2.2 in fiscal year 1995.  Air Force rates were reduced
from about 2.8 to 1.5 during that period.  Army aviation experienced
its best year in fiscal year 1995 with a Class A mishap rate of
0.8--a reduction from 3.5 in fiscal year 1975. 


   DECREASE IN CLASS A FLIGHT
   MISHAP FATALITIES (FISCAL YEARS
   1975-95)
------------------------------------------------------------ Letter :9



   (See figure in printed
   edition.)

The annual number of aviation fatalities has dropped significantly
since fiscal year 1975, when DOD reported 285, including 141 in the
Air Force, 103 in the Navy/Marine Corps, and 41 in the Army.  In
fiscal year 1995, fatalities had fallen to 85, including 53 in the
Air Force, 17 in the Navy/Marine Corps, and 15 in the Army. 


   DECREASE IN DOD RATE OF CLASS A
   FATALITIES (FISCAL YEARS
   1975-95)
----------------------------------------------------------- Letter :10



   (See figure in printed
   edition.)

The number of DOD aviation fatalities per 100,000 flying hours fell
from about 4 in fiscal year 1975 to 1.7 in fiscal year 1995. 


   DECREASE IN NUMBER OF DESTROYED
   AIRCRAFT (FISCAL YEARS 1975-95)
----------------------------------------------------------- Letter :11



   (See figure in printed
   edition.)

The number of destroyed aircraft resulting from flight mishaps fell
significantly between fiscal years 1975 and 1995.  In fiscal year
1975, 221 were destroyed, including 117 Navy/Marine Corps, 52 Air
Force, and 52 Army.  The number of destroyed aircraft fell to 67 in
fiscal year 1995, when the Navy/Marine Corps reported 31, the Air
Force 29, and the Army 7. 


   DECREASE IN RATE OF DESTROYED
   AIRCRAFT (FISCAL YEARS 1975-95)
----------------------------------------------------------- Letter :12



   (See figure in printed
   edition.)

Since fiscal year 1975, the annual rate of destroyed aircraft per
100,000 flying hours was reduced from 3.1 to 1.3. 


   VALUE OF CLASS A FLIGHT MISHAP
   LOSSES (FISCAL YEARS 1990-95)
----------------------------------------------------------- Letter :13



   (See figure in printed
   edition.)

Since fiscal year 1975, the services report the cost of Class A
flight mishaps at about $21 billion.  The value of Class A losses has
been fairly constant over the last 6 years, ranging from a high of
approximately $1.6 billion in fiscal year 1993 to a low of $1.2
billion in fiscal year 1994.  Even given that fiscal years 1994 and
1995 had generally low mishap rates, the value of Class A losses
still exceeded $2.5 billion during that time, totaling about $1.2
billion in fiscal year 1994 and $1.3 billion in fiscal year 1995. 


Briefing Section III MISHAP
INVESTIGATIONS
============================================================== Letter 


   EACH SERVICE HAS OWN MISHAP
   INVESTIGATION REQUIREMENTS)
----------------------------------------------------------- Letter :14



   (See figure in printed
   edition.)

DOD requires investigations of all Class A flight mishaps in order to
determine the causes and reduce future risks of property damage,
injuries, or deaths.  However, each service has established its own
requirements regarding how the investigations are performed and
reported. 

The Army assigns at least two safety investigators from its safety
center to each Class A mishap investigation--the board president and
recorder.  Other members of the investigation team are assigned from
the command having the mishap, other Army commands or DOD agencies,
or private industry.  The investigating team is responsible for
finding the cause(s) and preparing the final report.  Command review
and approval occur after preparation of the draft investigation
report. 

The Navy requires that aircraft reporting custodians appoint and
maintain standing aircraft mishap boards to investigate Class A
flight mishaps.  According to Naval Safety Center data, in about 75
percent of the Class A mishap investigations, a professional
investigator from the center is sent to assist the board.  The Safety
Center representative is not a voting member of the board.  However,
the Commander of the Naval Safety Center does not see this as a
problem since he is the final endorser of the report.  In addition,
although not required by regulation, the Commander also receives an
independent briefing from his representative on the board's tentative
findings and conclusions.  Each endorser up the chain of command must
comment on the report in turn, presenting evidence, if appropriate,
for suggested changes.  Endorsements become part of the report and
are available for review as the report goes up the chain.  Following
its final endorsement, the Safety Center tracks recommended
corrective action to completion. 

Prior to October 1995, the responsible Air Force major command
routinely delegated the responsibility for convening the
investigative board to the numbered Air Force that was the custodian
of the mishap aircraft.  The board president, as well as other
members, were normally selected from units within the numbered Air
Force.  The Air Force Safety Agency generally did not provide a
voting member to participate in the investigation.  Following the
investigation, the board briefed its findings to the major command
prior to finalization of the report.  In October 1995, the Air Force
implemented several changes to its investigation procedures as a
result of recommendations made by a Blue Ribbon Panel appointed by
the Air Force Chief of Staff. 


   CONCERNS ABOUT INDEPENDENCE OF
   INVESTIGATORS
----------------------------------------------------------- Letter :15



   (See figure in printed
   edition.)

A recurring issue of concern has been the independence of accident
investigators.  Our 1994 report\3 noted the Navy investigators' lack
of independence because of the investigators' organizational ties to
the mishap command.  The Navy has not taken action on that finding. 
A Blue Ribbon Panel convened by the Air Force to address aviation
safety stated in its 1995 report that mishap investigation boards
should be more independent of the convening authority.\4 As a result
of the panel's recommendations, the Secretary of the Air Force (1)
directed that the major command not delegate the convening authority
for Class A investigations, (2) mandated safety courses for key board
members, (3) directed that an Air Force safety center representative
be a voting member on all Class A flight mishap investigations, and
(4) clarified that only the voting members of the board can change
the final report. 

Briefing Section IV

--------------------
\3 Military Training Deaths:  Need to Ensure That Safety Lessons Are
Learned and Implemented (GAO/NSIAD-94-82, May 5, 1994). 

\4 The Blue Ribbon Panel on Aviation Safety, Department of the Air
Force (Sept.  5, 1995). 


ACTIONS TAKEN BY SERVICES TO
REDUCE MISHAPS
============================================================== Letter 


   ACTIONS TAKEN BY SERVICES TO
   REDUCE MISHAPS
----------------------------------------------------------- Letter :16



   (See figure in printed
   edition.)

Service safety centers track the implementation of recommendations
stemming from mishap investigations.  Within the Air Force, the
Safety Agency maintains a centralized database of open
recommendations and solicits status reports semi-annually.  Major
operational commanders track and report their disposition of open
recommendations for inclusion in the safety center database.  In the
Army, the responsibility for tracking the status of open
recommendations rests with the Safety Center in coordination with the
major commands.  The Army Safety Center conducts quarterly reviews of
open recommendations and documents the actions taken on each
recommendation.  The Navy places responsibility for ensuring that
recommendations are implemented with safety center analysts, who
monitor the recommendations associated with one or more aircraft
models. 

In addition to tracking recommendations, the services disseminate a
variety of information aimed at reducing flight mishaps.  This
information includes changes to manuals and procedures, safety
newsletters, videos of specific mishaps that identify the causes, and
safety messages. 

The services also undertake special initiatives aimed at reducing
mishap rates.  Earlier this year, for example, the Air Force convened
a panel to review aviation safety.  This panel was appointed on June
23, 1995 following a spate of aircraft mishaps during the early part
of fiscal year 1995, the highly publicized allegations that Air Force
mishap investigations lacked quality and objectivity, and your
request that GAO undertake this review.  While the panel found that
the long-term trends of reduced mishap rates reflected favorably on
the general soundness of the Air Force's safety program, the panel
made a number of recommendations aimed at improving investigation
objectivity and reducing human errors as a causal factor. 

Additionally, the Army undertook a study to determine the causes for
an increase in OH-58 helicopter mishaps.  The Army determined that a
disproportionate number of accidents had taken place at night and
involved human error.  The Army then developed a series of profiles
for predicting whether a mission was low, medium, or high risk.  From
this, the Army began using a system to assess the risk prior to each
OH-58 night mission, as well as guidance for reducing the risk to
acceptable levels.  The Army attributes the subsequent downward trend
in OH-58 accidents to this initiative.  Army officials informed us
that they plan to expand the use of this risk management system to
include other aircraft.  The Air Force and Navy are also developing
risk assessment programs. 

Operational commanders have also ordered flight operations to cease
for a specific time period so that personnel can focus on safety
issues.  Additionally, the services have undertaken a number of
studies looking at cause-and-effect relationships and their impact on
mishap rates.  An example of an ongoing study concerns two types of
Air Force mishaps, those involving air traffic control operations and
those involving rescue helicopters.  Both kinds of mishaps have
experienced notable increases during the past 2 years. 


Briefing Section V MAJOR MISHAP
CAUSES
============================================================== Letter 


   HUMAN ERROR WAS CONTRIBUTING
   FACTOR IN HIGH PERCENTAGE OF
   FLIGHT MISHAPS
----------------------------------------------------------- Letter :17



   (See figure in printed
   edition.)

In their efforts to reduce aviation mishaps, the services focus on
the contributing causes that can be grouped by human error, material,
and environmental.  On the basis of data reported by the services and
our own analyses, human error contributed to 73 percent of the Class
A flight mishaps in fiscal years 1994 and 1995.  Human error, as a
contributing factor, ranged from a high of 76 percent in Army mishaps
to approximately 71 percent in Air Force mishaps.  The Naval Safety
Center provided data showing that human error was a causal factor in
80 percent of Navy and Marine Corps Class A flight mishaps for fiscal
years 1990 through 1994. 


   CORRELATION BETWEEN OPERATING
   TEMPO AND MISHAPS IS
   INCONCLUSIVE
----------------------------------------------------------- Letter :18



   (See figure in printed
   edition.)

While service studies have addressed the relationship of operating
tempo to aviation mishaps, direct correlation has been inconclusive. 
For example, in 1994 the Secretary of Defense requested the Air Force
examine the relationship between operating tempo and other readiness
indicators and flight mishap rates.  The Air Force's final report
issued in September 1994, provided no direct correlation between
operating tempo and safety mishaps. 

In a September 1995 report, the Blue Ribbon Panel noted that "the
current 'Operations Tempo' in the Air Force presents an obvious
operational safety risk." The report pointed to many factors
contributing to this assessment such as organizational change,
reduced maintenance and leadership manning within squadrons, and
extended duty days of both flying and maintenance personnel.  The
panel believed that the Air Force should take actions to reduce
stress on aircrew and maintenance personnel and recommended that the
Air Force review the safety impact of reduced manning and aircraft
numbers without corresponding changes in operational requirements. 

Service statisticians, however, told us that the relatively low
incidence of Class A flight mishaps make drawing inferences and
statistical correlations of variables with mishap rates difficult. 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix I

NATIONAL SECURITY AND
INTERNATIONAL AFFAIRS DIVISION,
WASHINGTON, D.C. 

Sharon A.  Cekala, Associate Director William E.  Beusse, Assistant
Director

NORFOLK REGIONAL OFFICE

Hugh Brady, Evaluator-in-Charge Sharon Reid, Evaluator Harry E. 
Taylor, Jr., Evaluator


*** End of document. ***