Rejected Takeoffs: Causes, Problems and Consequences · controller as a safety factor in RTO events...

FLIGHT SAFETY FOUNDATION • FLIGHT SAFETY DIGEST • JANUARY 1993 1

... As we started our taxi, I asked the captain if hepreferred using normal or alternate takeoff power... My thoughts were occupied with the upcomingtakeoff procedure since it was my leg. After receiv-ing takeoff clearance, I advanced the throttles ... tothe normal takeoff power setting. I first heard theaural takeoff configuration warning horn ... Look-ing down at the flap handle I was absolutely ...surprised to see that it was in the up position. Weaborted ... before reaching 40 knots. [NationalAeronautics and Space Administration (NASA)Aviation Safety Reporting System (ASRS) recordnumber 78074]

Rejected Takoffs InvolveMultiple Risks

Accidents and incidents involving air trans-port aircraft have renewed interest in the pro-cess and problems associated with a rejectedtakeoff (RTO). The La Guardia Boeing 737-400runway overrun in 1989 involved flight crewperformance deficiencies before, during andafter the takeoff rejection. [The accident oc-curred September 20, 1989, at La Guardia Air-

port, Flushing, New York. There were 63 per-sons involved, including two fatalities.] Con-tributing to these human performance errorswere external conditions that were not per-ceived by the flight crew as being relevant totheir operating decisions.

The 1987 Detroit DC-9-82 and 1988 Dallas Boeing727-232 no-flap takeoffs also underscored thepossibility that flight crews could fail to properlyconfigure their aircraft for takeoff and not de-tect their acts of omission. [The DC-9 accidentoccurred August 16, 1987, at Detroit Metro-politan/Wayne County Airport, Romulus, Michi-gan. There were 162 persons involved, includ-ing 156 fatalities. The Boeing 727 accidentoccurred August 31, 1988, at Dallas Interna-tional Airport, Texas. There were 108 personsinvolved, including 14 fatalities.]

RTOs introduce multiple risks — those associ-ated with the takeoff abort and those associ-ated with the events that may follow the abort.RTOs are also symptomatic of a breakdown inhuman performance that can lead to improperaircraft conditions or configurations. A

Rejected Takeoffs:Causes, Problems and Consequences

Rejected takeoffs involve multiple risks and require a high level of pilotperception, judgment and procedural skill. This study looks at human

factors associated with rejected takeoffs.

byCapt. Roy W. Chamberlin

National Aeronautics and Space AdministrationAviation Safety Reporting System

FLIGHT SAFETY FOUNDATION • FLIGHT SAFETY DIGEST • JANUARY 19932

successfully managed RTO involves a skillfulblending of pilot perception and appropriateaction to conclude the abort procedure safelyand avoid dangerous follow-on events.

The following data present a small part of alarger ongoing effort that seeks to categorizethe causes, problems and effects of rejectedtakeoff events as reported through ASRS.

Study Focuses onHuman Error RTOs

Human errors associated with rejected take-offs reported to ASRS were analyzed. Incidentreports were studied to understand the flightcrew human factors that led to RTOs; to iden-tify decision-making and procedural issues as-sociated with RTO initiation and execution;and to analyze problems that occurred in thewake of rejected takeoffs.

Initially, 507 incidents occurring between January1, 1983, and November 30, 1990, were retrievedfrom the ASRS database. Only reports submit-ted by flight crew members of transport cat-egory aircraft (in excess of 60,000 lbs./27,000kg. gross weight) were considered. Of these,168 were found to be relevant to flight crewdecision-making and procedures. The findingsof this study are based on this 168-report sub-set (Table 1) and focus on the flight crew per-formance problems that are factors before, duringand after an RTO.

Causal Factors Sought

The reports were read and analyzed for causalfactors underlying the rejected takeoff event.Primary causal factors were labeled as flightcrew procedural errors or conditions that pre-dispose such errors. Secondary contributingfactors were also considered. In each incident,the abort maneuver was examined for poten-tial problems with its initiation and execution.Finally, flight crew decisions made in the wakeof the rejected takeoff were also evaluated.

The RTO study subset was limited to reportssubmitted by flight crew members, since onlythey could shed light on the cockpit proce-dures employed, crew members’ roles and crewperceptions of aircraft operating conditions.

Data collected by the ASRS are subject to bothknown and unknown biases. Since reports arevoluntarily submitted, they constitute a non-random sample of the actual population ofaviation safety incidents. In addition, report-ers’ incident descriptions are colored by theirindividual motivations for reporting. They usu-ally give only one perspective of the event,and this is not balanced by any additionalinvestigation or verification.)

Flight Crew Errors LeadingTo RTOs Identified

Ninety-four RTOs were caused by crew er-rors. Five categories of crew-induced rejectedtakeoff scenarios were identified. These were:

Unauthorized Takeoffs. An aircraft departedprematurely or used the wrong runway fortakeoff. In wrong-runway takeoffs, the air-craft was authorized onto a runway butthen deviated from air traffic control (ATC)directives (22 incidents).

Taxiway Takeoffs. An aircraft departed froma taxiway rather than a runway (seven in-cidents).

Off-runway Takeoffs. An aircraft errone-ously aligned with the runway edge lights

Table 1Distribution of Study Reports

Category Number of Incidents*

Flight Crew Errors Leading to RTOs 94

RTO Initiation and Execution Problems 13

Post-RTO Problems 84

*Some reports apply to more than one category

Source: Aviation Safety Reporting System


instead of the centerline lights while posi-tioning for takeoff (three incidents).

Aircraft Configuration Anomalies. An air-craft was improperly configured before,during or following the RTO. The aircraftconfiguration anomalies included four con-ventional abnormals often practiced in re-current training: improperly set flaps;unstowed spoilers or spoiler handle; sta-bilizer trim not in agreement with presetparameters; and failure to observe that acockpit window was unlatched (34 inci-dents).

Loss of Aircraft Control. An initial mis-management of thrust leverscreated a loss of aircraft head-ing control. Loss of control wasoften worsened by misuse ofprimary ground steering devices(10 incidents).

The remaining 18 incidents involvedaircraft discrepancies unrelated toconfiguration that were attributableto flight crew errors.

In addition to identifying RTO eventcategories, flight crew proceduralerrors contributing to RTOs werealso determined. Procedural errorswere classified as improper infor-mation transfer, deficiencies in taskmanagement and crew coordina-tion, and aircraft configuration anomalies.

Improper Information Transfer. Some re-ports revealed that the interaction betweenthe flight crews and tower controllers wasnot effectively monitored by the captain.The lack of an ATC response to a flightcrew communication was often interpretedas approval by the flight crew. Other com-munication failures were associated withunauthorized and premature takeoff inci-dents. In these, flight crews failed to chal-lenge partial or doubtful clearances. Thesefindings suggest that intra-crew commu-nications can be compromised by insuffi-cient cross-checking.

Some reportsrevealed that

the interactionbetween the

flight crews andtower

controllers wasnot effectivelymonitored bythe captain.

Deficiencies in Task Management and CrewCoordination. Flight crews did not alwayschoose the right time to perform a requiredfunction. Often, the error was in allowingthe other pilot to change radio frequencyto make a company radio call at an inap-propriate time. Wrong-runway takeoffs usu-ally were associated with a rushed cockpitenvironment and poor crew coordinationin the areas of cross-checking, mutual sup-port and use of proper charts. Off-runwaytakeoff and taxiway takeoff events also werecharacterized by rushing and lack of flightcrew coordination. In the latter events (whichusually occurred at night), it was commonfor one pilot to be head-down in the cock-

pit while performing the checklistduring the runway entry.

Failure to monitor control inputsand to detect inappropriate controlm o v e m e n t s l e d t o f i v eloss-of-control incidents. These typi-cally were caused by improperthrottle application and the result-ing uneven spool-up of large by-pass engines. Asymmetrical thrustwas further exaggerated by snow,ice, moisture or rubber deposits onthe runway surface. Sometimes, amisapplication of corrective controland throttle movements placed theaircraft off the runway, because theloss of control was so unexpected.

Aircraft Configuration Anomalies. Improp-erly performed checklists were the leadingcause of aircraft configuration anomalies.Most reporters stated that they missed achecklist item — most frequently the flaps— during a period of busy cockpit activityor during interruptions of routine moni-toring and cross-checking functions. Therewere eight incidents where leading edgeflaps created a warning, usually becauseof a circuit breaker being out of normalposition. Another six reports cited errorsin trim setting.

Several conditions were identified as poten-tially predisposing flight crew procedural er-rors. These consisted of radio frequency


congestion, schedule pressure, environmentalfactors and transfer of control to the first offi-cer.

Frequency Congestion. This was often as-sociated with clipped transmissions, missedclearances and readbacks and flight crewsresponding to wrong call signs. Pilots weremore prone to act on their expectations,during periods of excessive radio trafficrather than on ATC’s actual instructions.Frequency congestion was the most fre-quently cited predisposing condition.

Schedule Pressure. Other reports reflectedschedule-related pressures that compelledflight crews to hurry. Driven by company“on-time” considerations or ATCtraffic flow priorities, flight crewsimprovised callouts and alteredcockpit procedures to meet sched-ule demands. These proceduralshortcuts led crews into incom-plete readbacks, nonstandardphraseology and inadequate in-tra-cockpit communication. Hur-rying also led to missed items inthe checklist. The assumption oftoo many tasks in too brief a timeoverloads a flight crew and re-sults in “forgetfulness” that theU.S. Federal Aviation Adminis-tration (FAA) has identified as a causal factorin runway incursions. This work load-in-duced forgetfulness was also associated withmany RTO events in the study data.

Environmental Factors. Weather adverselyaffected visibility and lighting conditionsand contributed to flight crew performanceerrors in 10 reports. In some cases, runwaylighting was also instrumental in creatingdisorientation and resulted in either wrong-runway takeoffs or taxiway takeoffs at night.A few flight crews requested that towercontrollers dim or turn off the lights oninactive runways.

Transfer of Control to the First Officer.As represented by the opening report ex-cerpt, a disproportionate number of RTOsoccurred when the first officer was con-

The role ofthe tower

controller as asafety factor in

RTO eventsreported toASRS was

significant.

ducting the takeoff and had control of thethrottles. These events included off-run-way and unauthorized takeoffs, improperaircraft configurations and loss of aircraftcontrol. From the character of these reports,it appeared that problems sometimes re-sulted from the first officer ’s failure to ex-ecute the initial phase of the takeoff in themanner expected by the captain. The captain’sexpectations were often shaped by the firstofficer ’s past performance. However, whenthe first officer and the captain were unfa-miliar with each other, a captain was proneto assess a first officer ’s capabilities onlyby his length of experience. In either cir-cumstance, captains exhibited complacencyregarding their responsibility to monitor

first officers’ actions.

Tower Controllers Cited AsSignificant Safety Factor

The role of the tower controller as asafety factor in RTO events reportedto ASRS was significant. RTOs weremost often initiated by the tower con-troller during unauthorized, wrong-runway takeoffs and taxiway take-offs. These events were usually caughtin the controller ’s scan, and a low-speed abort resulted. In contrast, run-

way excursions and off-runway takeoffs wereoften detected by the flight crew. The flightcrew disorientation inherent in these eventsusually resulted in relatively high-speed re-jections; however, off-runway incidents some-times continued into takeoffs where potentialaircraft damage could go undetected by theflight crew.

Aircraft configuration problems that resultedfrom flight crew procedural errors were usu-ally announced by the takeoff warning systemand typically resulted in a low-speed abort.The low-speed RTO was generally a reactiveclosing of the throttles and coasting to thenext turn-off.

Abort decisions related to aircraft system fail-ures — including engine failures — were moreapt to be derived from multiple warnings and


to result in high-speed aborts. There were 13reports where the abort speeds were at V1,and in some cases the speed was as high as Vrand into liftoff. Most crews seemed to basetheir go/no-go decisions not onlyon speed but also on the number ofwarnings received, runway remainingand their perception as to whetherthe aircraft could safely fly. The sen-sory advisories stimulating crew de-cision-making were audibles, suchas compressor stalls, tower alertsand warning systems; visual indi-cations of engine problems; and tactilesensing of vibrations. The most com-mon audible was the compressor stall.

It appeared that if pilots receivedtwo related engine indications suchas a compressor stall and a fire warn-ing, they were more likely to abortat a speed above that which train-ing dictates. Another decision fac-tor was that aircraft vibration by it-self appeared to create doubt in theflight crew as to the ability of the aircraft tocontinue safely. The visual aspects of runwayremaining also entered into pilots’ perceptionsand decision-making. Lower-speed aborts atV1 or less were related to the pilots’ percep-tions of runway remaining and braking re-quired. Tire considerations were the main de-cision factor in these cases.

Crew Perceptions DeterminedDecisions Following an RTO

Decisions made by flight crews in the wake ofa rejected takeoff were based largely on theirperceptions of aircraft integrity. These percep-tions were shaped by warning system alerts,tactile sensing, engine instrument indications,aircraft-generated noises and observations ra-dioed by tower controllers or other externalobservers.

Request for Emergency Equipment. Theperceived requirement for emergency equip-ment was based again on whether therewere two or more warnings associated withthe condition of the aircraft. A compressor

stall by itself did not produce a call for thefire truck, but if it was accompanied by atower warning or a system warning of anengine fire, the crew would usually call for

the emergency equipment. Moreoften than not, a compressor stallwould result in engine damage orengine fire. Severe vibrations alsoled to a call for assistance. Controltower operators were of great as-sistance to RTO aircraft and in manycases actually initiated the call foremergency equipment when fire wasindicated. Other RTOs were trig-gered by door lights and other sys-tem light warnings. These were falsewarnings in many cases, but theyusually appeared when the aircrafthad reached a high speed, therebyrequiring not only great finesse inexecution of the RTO but also gen-erating heat within the braking sys-tems.

In the aftermath of an RTO, flightcrew decisions regarding emergency equipmentwere based largely on their perceptions of air-craft integrity and passenger safety. Flight crewsdid not always call for emergency vehiclesafter a successful RTO if they believed thateverything was under control. The same mind-set that prevents a crew from declaring anemergency during takeoff or landing seemedto drive the crew’s decision not to ask for as-sistance from the emergency vehicles. Dataindicate that this is not always a correct as-sumption or position to maintain.

Use of Brake Energy Charts. To determinetire condition, the brake energy chart isone of the decision tools available to flightcrews after an RTO. Crews did not alwaysuse the charts, and sometimes they misin-terpreted them. A few reporters indicatedthat the charts did not take into accountlong taxi distances; thus, subsequent take-offs resulted in deflated tires. Another re-sult of nonuse or misuse of these chartswas a return to the gate area by an aircraftwith exceedingly hot brakes. In one inci-dent, maintenance called the fire trucks tothe gate area because the aircraft’s brakes

Some flightcrews were

more likely toinitiate a

second takeoffif their RTO

was inresponse to a

false orcorrectedcockpit

warning.


were glowing red and the crew was notaware of the danger to the aircraft or toground crew personnel. Data indicated onlyoccasional use of hot-brake areas by flightcrews following an RTO.

Takeoffs Following RTOs. Some flight crewswere more likely to initiate a second take-off if their RTO was in response to a falseor corrected cockpit warning. The decisionfactors involved were schedule pressureand the flight crew’s assessment of brakeand tire conditions. Schedule pressure attimes allowed unqualified ramp personnelt o v e r i f y c a r g o d o o r i n t e g r i t y t opreclude a return to the gate. There weremany problems with landings on deflatedtires as a result of second takeoffs.

Crew Communication with EmergencyGround Vehicles. Some reports indicatedthat flight crews wanted to communicatedirectly with ground vehicles and that towertransmissions sometimes interfered withthat capability. During one aircraft evacu-ation, a ground vehicle blocked an exit dooras the vehicle crew was assisting the flight.Another reported that a fire crew identi-fied the wrong engine that was on fire. Ineach of these examples, there was no com-munication between the aircraft crew andground vehicle/fire crew.

Aircraft Evacuation. A flight crew deci-sion to evacuate an aircraft was most oftendriven by a concern of fire in an engine orlanding gear. Smoke in the cabin was an-other reason for evacuation. Although datadid not consistently reveal the level andquality of flight crew interactions, therewere some indications that, during aircraftevacuations, flight crews advised passen-gers of the situation and cabin crews func-tioned as trained. Calling for emergencyequipment provided the flight crew withassurance regarding the condition of theaircraft, in some cases, and led to a deci-sion to evacuate when fire was observedfrom outside the aircraft by emergencypersonnel.

The data indicate the following conclusionsand recommendations:

In these data, the most significant causes ofcrew-induced RTOs appeared to be impropercommunications procedures influenced by ex-ternal conditions of frequency congestion andschedule pressure. These external factors in-duced a hurry-up attitude and lowered theexchange of information required to manage awell-coordinated cockpit. They also predisposeda lack of coordination and vigilance whereone pilot could monitor the other, particularlyduring runway entries at night.

Rushing caused by schedule pressure and ATCtraffic flow priorities also interfered with theaccurate completion of cockpit checklists. Dataindicated that high work loads, operations dis-tractions and complacency were involved inmost cases of checklist errors and omissions.The most serious of these resulted in no-flapstakeoffs, leaving the takeoff warning systemsas the only “safety straps.”

Some flight crews deviated from their train-ing guidelines when a high-speed abort deci-sion was involved and did not adhere to V1 asa go/no-go boundary. This may indicate thatcurrent training scenarios lack realism whenmodeling RTO conditions. To reduce these er-rors, the problem of risk assessment may haveto be addressed in either simulator training orground school classes.

Some flight crews were overly optimistic intheir assessment of aircraft condition after re-jecting a takeoff. Flight crews should be en-couraged to ask for assistance after just oneindication of a problem, as external examina-tions and communication are vital for a trueunderstanding of aircraft condition.

Some flight crews failed to use brake energycharts following an RTO and often did notconsider other pertinent factors, especially taxidistances. In some circumstances, brake en-ergy-chart estimates may be unrealistically low.This can lead to inadequate cooling times andmay result in tire failure either on the groundor in the air.


Simulator and crew resource management pro-grams may improve aspects of pilot perfor-mance associated with aborted takeoffs by:

• Incorporating real-life scenarios intosimulator sessions, including the pres-ence of vibrations and multiple warn-ings;

• Introducing subtle equipment failure intosimulator training such as engine in-strument fluctuation, door warning lightactivation, stick shaker activation andabnormal control column force;

• Demonstrating proper radio proceduresusing tower-tape examples of misun-derstood clearances to promote the useof readbacks by flight crews and to en-courage intracockpit communicationprocedures to verify or question vagueclearances; and,

• Promoting rigorous methodologies forthe execution of checklists, especiallyflap settings.

It may be helpful to consider formal criteriafor allowing takeoffs by the first officer. Suchcriteria might address:

• First officer performance;

• First officer experience level; and,

• Weather factors.

The captain’s responsibilities during the firstofficer ’s takeoffs should also be formalized.

Real-life scenarios, such as those described inASRS reports, could serve as a beneficial ele-ment of a complete training curriculum. In-corporating these reports into ground schoolvideo presentations and simulator line-orientedflight training (LOFT) programs would pro-vide, at the very least, the realism that is neededto expose flight crews to the subtleties of thehuman factor problems of rejected takeoffs. ♦

About the Author

Capt. Roy W. Chamberlin is an aviation safetyanalyst with more than 30 years of flight experience.He has logged more than 22,000 flight hours inmany aircraft, including Boeing 747s withTransWorld Airlines. At ASRS, Chamberlin analyzesreport submissions from air carrier and generalaviation pilots.


Training for Advanced CockpitTechnology Aircraft

A group of 100 pilots was asked to assess safety issues related to lineoperations of new generation, high technology aircraft. In an earlierstudy, another group of 48 pilots identified a variety of training and

procedural concerns.

byHarry W. Orlady and William A. Wheeler

National Aeronautics and Space AdministrationAviation Safety Reporting System

Shortly after advanced cockpit technology(ADVTECH) aircraft were introduced, the Avia-tion Safety Reporting System (ASRS) was askedby the National Aeronautics and Space Ad-ministration (NASA) and others in the avia-tion community to determine pilot opinionsof the overall safety of these new-generationaircraft in line operations.

A group of 48 pilots who flew ADVTECH air-craft in regular service and who had reportedincidents was selected for comprehensive tele-phone interviews using a stratified, randomsampling procedure. The pilots identified trainingand, on some airlines, operating proceduresas problem areas. Their views supported a gen-eral industry consensus that training practiceshad not kept pace with cockpit technology.

Although flight crew training and operatingprocedures are obviously interrelated (theyprovide the interface between aircraft and thepilots who fly them), a follow-up study wasrestricted to training issues for several rea-

sons. First, training and training concepts arerelatively independent of variations in oper-ating procedures. Second, if specific problemareas in training can be verified, improvementscan be made with relative ease. Third, thissubject was particularly amenable to explora-tion with the data-gathering tools available tothe ASRS. Finally, because the adequacy oftraining directly affects cockpit work load (par-ticularly during high work load periods), atraining study could be expected to provideadditional data on the controversy of cockpitwork load in ADVTECH aircraft.

The follow-up study’s objectives were identi-fied as follows:

• Determine line pilots’ views of the ini-tial and recurrent training that they re-ceived to fly ADVTECH aircraft;

• Determine the strengths and weaknessesof current training and the sensitivityof this training to widely varying needs;


and,

• Identify the most effective methods forinstructing flight crews of ADVTECHcockpit aircraft with the hope of identi-fying model training curricula.

Preliminary findings focused on training forcrew coordination and communication withADVTECH aircraft and maintenance of basicflying skills.

One of the great strengths of theASRS is its ability to contact thepilots who report to it during thevery short period that ASRS holdsreporter identification slips. Be-tween October 1988 and Febru-ary 1989, approximately 100 pi-lots, who were flying ADVTECHaircraft and reported incidents tothe ASRS, were called and askedto participate in the survey. Noone refused, although they wereunder no obligation to cooper-ate. Participants were selectedfrom the much larger base ofADVTECH pilot reports to ob-tain a reasonable (albeit not per-fect) distribution between cap-tains and first officers amongcurrent ADVTECH transports,from established trunk and in-ternational carriers to newly established com-muter airlines.

The surveyed pilots represented 12 airlinesand included the following aircraft:

A300-600; Boeing 737-300/400/500; Boeing 757/767; MD-80; and MD-88. Selected pilot popu-lation variables are shown in Table 1.

Pilots were sent a list of the general subjectsto be discussed before the telephone inter-views, which required about one hour. Thepilots stated this had been helpful becausethey were able to present well-thought opin-i o n s , n o t s n a pjudgments.

In addition to basic demographic data, the

questions were based on nearly 30 trainingissues identified by ADVTECH pilots in theearlier study or developed from Working Pa-per on Training for Advanced Technology Air-craft.

Interviews were conducted by experienced air-line pilots because they were able to establisha rapport with the respondents quickly andthey understood the technical aspects of thequestions. These interviewers were given train-

ing on interviewing techniques, with empha-sis on the importance of controlling interviewerbias.

Even without the complication of advancedcockpit technology aircraft, airline pilot train-ing is a complex subject. Its complexity is in-creased by many factors. These include a broadrange of aircraft, differences in airline opera-tions and operating philosophies, and a vari-ety of airline training resources. In addition,there are different training needs for pilotswith a wide range of skills and experience andfor pilots who operate in an air traffic control(ATC) system that is not always sensitive toaircraft performance characteristics.

Two training issues examined closely are:

Table 1Pilot Survey Demographics

Captains First Officers

Flight HoursAverage 13,000 6,777Range 4,500-25,000 2,500-12,500

Years with Present AirlineAverage 17.7 6.9Range 1.5-35.0 0.5-17.5

Hours in TypeAverage 1,025 720Range 50-3,500 50-2,500

First Time in PresentCockpit Position 27% 19%

TransitionedFrom 3-person Crew 52% 63%

Source: Aviation Safety Reporting System


• Intracockpit communication and crewcoordination in ADVTECH aircraft; and,

• Maintenance of basic flying skills.

Communication and CrewCoordination Viewed as Primary

As suggested in the previous study, many ofthe surveyed pilots believe that good crewcoordination and good cockpit communica-tion are more important in ADVTECH aircraftthan in non-ADVTECH airplanes.

Their importance is also recognized in indus-try practices. For example, approximately 70percent of the surveyed pilots received theirtransition training utilizing a “full-crew con-cept” during simulator training. Nearly one-half of the remaining 30 percent, who weretrained instead by cockpit position (i.e., cap-tains with captains and copilots with copi-lots), believed this was not a good practice.

One of the major innovations in airline flightcrew training during the past decade has beenthe development of formalized crew resourcemanagement (CRM) training, which stressescrew coordination and intracockpit communi-cation. Companies for 85 percent of the pilotshad formalized CRM training programs. In analmost unanimous response, 97 percent of thesurveyed pilots believed that “there is a realneed for such programs.”

The comments listed in Table 2 illustrate thevariety of reasons for the support of the linepilots for CRM training, despite some criti-cism of their airlines’ current programs, andother qualifications.

Pilots Expressed Concern aboutMaintenance of Basic Flight Skills

Although there is nothing new about the problemof maintaining basic flying skills, there is con-siderable evidence that this difficulty has beenexacerbated with ADVTECH operations. Priorto the introduction of ADVTECH aircraft, theproblem was largely confined to long-rangeflight operations marked by fewer takeoffs andlandings. However, when highly automatedADVTECH aircraft were introduced, the poli-cies and procedures of many companies stressedthe maximum use of their automatic systems.This policy, which has been sometimes calledthe “we bought it, you use it” philosophy, cre-ated a maintenance-of-skills problem for thepilots flying these aircraft. (Several airlines

Table 2Responses to Question:

“As a professional pilot, do you think thereis a real need for a crew resource

management (CRM) program?”

Captains’ Comments

“YES”

“However, one year the programs are good, thenext year no good.”

“Pilots are mechanical — not people-oriented(i.e., people managing). Our program still doesn’trecognize this.”

“It’s needed for some individuals.”

“Timid crew members are sometimes afraid tovoice their opinions.”

“Have less personality conflicts than in the past.Captain regards the rest of the crew as humanbeings and accepts inputs from the rest of thecrew.”

“NO”

“Has diluted Captain’s authority. Not neces-sary!”

First Officers’ Comments

“YES”

“Everyone’s opinions are considered.”

“Puts things into perspective and recognizesF/O intelligent input.”

“Don’t feel free to volunteer input with certaintypes of captains (personality).”

“Not enough feedback to company on lineoperations being used in training.”

“Have noticed changes in crews recently, butI’m not flying with the old timers.”

“NO”


the quality of ADVTECH pilot training sincethese airplanes were introduced. This is notsurprising because there are usually shake-down periods in new training programs. Thequality of individual programs still varies, andindividual needs are not always recognized.

However, pilot attitudes toward ADVTECHtraining and operating policies have changed.The pilots who were interviewed for this studybelieve that current operating policies and train-ing show greater sensitivity to line operatingneeds, unlike their peers who were surveyedin the earlier study.

Still, there is room for improvement. Some

Table 3Responses to Question:

“Do you think that attainment and reten-tion of manual flying skills is a particular

problem for low-time pilots?”

have modified their policies since the earlierADVTECH study.)

One captain explained that maintenance ofmanual skills had indeed been a problem untilhis airline had moderated its policy regardingmaximum use of the automatic systems. Hesaid that it is still a problem for low-time pi-lots, but in this case the problem is in theinitial development of skills — not in the main-tenance of skills that have already been devel-oped. He said, “It’s not their fault, but manyof the new copilots have never had a chance tolearn these skills and they don’t have enoughopportunity to practice.” Table 3 presents typicalcomments made by captains and first officerswhen asked if the maintenance of skills was aproblem for the low-time pilot.

Even without the additional complication ofadvanced cockpit technology, airline pilot train-ing is a very complex business. There are manyvariables, and some of them are critical. Withfew exceptions, industry-wide generalizationscan be made only at considerable peril.

The data indicate the following conclusions:

• Pilots like these airplanes;

• The addition of sophisticated automatedsystems has not reduced the level ofbasic airmanship skills required of anairline pilot;

• Automation has not reduced trainingneeds;

• Computer-designed or computer-assistedtraining is not yet an unqualified suc-cess, despite glowing testimonials in itssupport; and,

• Major advances in information display,as exemplified in glass cockpits, havecreated some problems that may be re-lated to training. Moving map displaysare an exception; they are universallyliked.

There have been significant improvements in

Captains’ Comments

“NO”

“Most fly manual too much when they shoulduse auto.”

“Has to do with personal discipline — not lowtime.”

“YES”

“They get so enthralled with all the magic theyforget how to flip switches back to raw data.”

“You can develop additional skills without fly-ing. When you need to hand fly, you need yourhighest skills. The only way to attain and main-tain them is to fly manually.”

“Yes, but also for everybody else.”

First Officers’ Comments

“NO”

“Because most of us fly a lot manually andcompany encourages this.”

“YES”

“Have to force yourself to fly manually. Somecaptains are more comfortable if you are onauto. Some captains are not good on PNF [pilot-not-flying] duties.”


resources management from Pepperdine University.

Capt. Harry W. Orlady is an independent researchconsultant with more than 53 years of aviationexperience, including 38 years with United Airlines.

In 34 years as a United captain, Orlady loggedmore than 31,000 hours of flight time and flew allof the airline’s routes in 12 different aircraft. Heflew as captain on the DC-3, DC-4, DC-6, DC-7,DC-8, Boeing 720 and the Boeing 747.

Orlady has worked with NASA’s Aviation SafetyReporting System (ASRS) since 1982 as an aviationsafety research consultant specializing in air carriermulti-crew operations, with a focus on aeromedicaland human factors. He is president of OrladyAssociates, an aviation consulting firm affiliatedwith Battelle’s ASRS program.

References

Norman, S.D. and Orlady, H.W. Flight DeckAutomation: Promises and Realities: Final Re-port of a National Aeronautics and Space Ad-ministration/Federal Aviation Administration/Industry Workshop, Carmel Valley, Califor-nia, August 1-4, 1988.

Orlady, H.W. Working Paper on Training for Ad-vanced Cockpit Technology Aircraft. Battelle Avia-tion Safety Reporting System Office, Moun-tain View, California, July 1988.

training methods seem more effective than others.Some carriers appear more adept at trainingtheir crews for service in ADVTECH aircraft.Future studies should identify those operat-ing policies and training procedures that havedemonstrated their value and should be imple-mented universally. ♦

About the Authors

William A. Wheeler, a research scientist and formernaval flight officer, joined Battelle (which operatesNASA’s Aviation Safety Reporting System ASRS)in 1982. He is assigned to Battelle’s Human AffairsResearch Center in Seattle, Washington. In additionto his flight experience, Wheeler has more than 20years of experience working in naval aircraftmaintenance, both as a technician and as a manager.

Wheeler has been involved in a wide range of researchprojects, many of which focused on humanperformance on complex systems while under stress.He holds advanced degrees in psychology and human


Aviation Statistics

The recently released annual General AviationActivity and Avionics (GAAA) Survey conductedby the U.S. Federal Aviation Administration(FAA) provides information about the activi-ties and avionics equipment of the general avia-tion aircraft fleet for calendar year 1991. Theinformation obtained from the survey enablesthe FAA to monitor the general aviation fleetso that the FAA can, among other activities,anticipate and meet demand for airways fa-cilities and services, assess the impact of regu-latory changes on the general aviation fleetand implement measures to ensure the safeoperation of all aircraft in U.S. airspace.

“General aviation” is not always defined thesame in the aviation community. The generalaviation aircraft represented in this report rangein complexity from simple gliders and bal-loons to sophisticated four-engine turbojets.These aircraft are used for a variety of pur-poses such as air taxi, cargo, agricultural, ex-ecutive/business, personal, research, instruc-tional, and recreational. The survey excludesaircraft operated by scheduled airlines.

Following are some of the survey’s significantfindings.

• The estimated 198,475 active general avia-tion aircraft in the fleet flew more than30 million hours, with an average an-nual flight time per aircraft of 149 hours.These active aircraft represent approxi-

mately 75 percent of the registered generalaviation fleet, which is five percent lowerthan was estimated in 1990.

• Active general aviation aircraft under-took nearly 96 million operations (takeoffsand landings). About 69 percent werein local flight vs. 31 percent in cross-country flight.

• General aviation aircraft flew more than3.9 million nautical miles.

• Approximately 87 percent of generalaviation flying took place during theday.

• Forty-five percent of the hours flownby the general aviation fleet were flownwith no flight plan, and an additionalseven percent of hours flown were un-der some other/unknown flight plan.Only 25 percent of the aircraft hourswere flown under visual flight rules/daylight visual flight rules (VFR/DVFR)flight plan, and 23 percent were flownunder instrument flight rules (IFR).

• An estimated 930 million gallons of fuelwere consumed by the active generalaviation fleet. Approximately 38 per-cent of the total fuel consumed was avia-tion gasoline, and 62 percent was jetfuel.

Survey Tracks U.S. General AviationFleet Activities

byEditorial Staff


• Almost 38 percent of the active generalaviation fleet flew under instrument flightrules (IFR).

• The three regions with the greatest num-ber of active aircraft were: the WesternPacific Region with 18.4 percent, theGreat Lakes Region with 17.5 percent,and the Southern Region with 16.3 per-cent. The region with the smallest numberof active aircraft was the Alaskan Re-gion, which constituted 3.3 percent ofthe active general aviation fleet.

• States represented by the largest num-ber of active general aviation aircraftinclude California with 14.7 percent,Texas with 8.2 percent and Florida with6.2 percent.

Personal Use Ranks Highest

• Rotorcraft, turboprop and turbojet air-craft types averaged 452, 308 and 290flight hours per aircraft, respectively.In contrast, active fixed-wing piston air-craft, which make up 78 percent of theactive fleet and represent 68 percent ofthe total flight time, averaged only 137flight hours per aircraft.

• Turbine rotorcraft had the most aver-age hours flown per aircraft (592). Theaircraft types with the least number ofaverage hours flown were the “other”piston, averaging 41 hours, and aircrafttypes in the “other” category (e.g., glidersand balloons), which averaged 61 hoursflown per aircraft.

• The most popular primary-use categoryof the active general aviation aircraft ispersonal use, with 58 percent of theactive fleet falling into this category.The next closest primary-use categorywas business with 16 percent, followedby instructional use with nine percent.

Several Activities MeasureGrowth and Activity

Several aviation activity measures indicategrowth trends and activity levels in the gen-eral aviation fleet, including measures of thesize of the general aviation population, num-ber of active aircraft, total flight hours, aver-age flight hours per aircraft and number oflandings. (Figures 1& 2, page 15; Figure 3,page 16)

The data indicate that:

• A great deal of variation in the numberof active aircraft, total hours and aver-age aviation hours exists among all typesof general aviation aircraft.

• More than 30 million hours were flownby the estimated 198,475 active generalaviation aircraft.

• The average flight time per active air-craft was 149 hours. Active aircraft con-stituted about 75 percent of the regis-tered general aviation fleet, which isfive percent lower than was estimatedin 1990.

• Single-engine piston aircraft, with a popu-lation of 206,371 or 78 percent of theregistered general aviation fleet, domi-nated the general aviation fleet, althoughthe average hours flown (134) were lowerthan most aircraft types. This aircrafttype accounted for 78 percent of theactive aircraft but only 68 percent ofthe total flight time.

• Turbine rotorcraft averaged the mosthours per aircraft of any aircraft typeat 592 average hours. Fixed-wing tur-boprops with 13 or more seats were aclose second with 589 average hours.This aircraft type’s high average hoursare most likely attributable to heavycommercial use as commuter a ircarriers.

• The two manufacturer/model groupswith the largest representation in the


general aviation fleet were theCessna 172, with 23,918 regis-tered aircraft (nine percent of theregistered general aviation fleet),of which 88 percent were active,and the Piper PA28, with 21,423registered aircraft (eight percentof the registered general avia-tion fleet), of which 86 percentwere active. The Cessna 172 ac-counted for 13 percent of the to-tal hours flown, and the PiperPA28 accounted for eight per-cent of the total hours flown.

• The percentages of registered air-craft active in each region arerelatively similar, ranging froma low of 71 percent in the Alas-kan Region to a high of 80 per-cent in the New England Region.

• The three regions with the greatestnumber of active aircraft werethe Western Pacific Region with36,545 active aircraft; the GreatLakes Region with 34,792; andthe Southern Region with 32, 428.

• The Western Pacific Region accountedfor the most flight time of any region,5.5 million hours, with the Southern,Southwestern and Great Lakes Regionsclose behind.

• The state with the largest estimated num-ber of active aircraft was California with29,261, followed by Texas with 16,206and Florida with 12,336.

• The state with the highest estimatedaverage flight hours was Hawaii, with534.3 hours. Montana averaged the lowestflight hours at 86.1.

• The general aviation fleet made almost48 million landings. About 69 percentof the landings were in local flight com-pared with 31 percent in cross-countryflight.

1991 General Aviation Active AircraftBy Aircraft Type

1991 General Aviation Total Flight HoursBy Aircraft Type

Source: U.S. Federal Aviation Administration

Other

2 Engines, 7+ Seats

2 Engines, 1-6 Seats

1 Engine, 4+ Seats

1 Engine, 1-3 Seats 55.7

98.5

13.6

7.6

0.1

Fixed-wing Piston

Other

2 Engines, 13+ Seats

2 Engines, 1-12 Seats 3.8

0.6

0.5

Fixed-wing Turboprop

Other

2 Engines 4.1

0.3

Fixed-wing Turbojet

Turbine

Piston

0 20 40 60 80 100

2.5

3.8

Rotorcraft

7.6Other Aircraft

Number of Active Aircraft (Thousands)

Air

craf

t T

ype

Figure 2

Figure 1


Single-engine Piston

20.5 Million Hours68%

Multi-engine Piston

3.6 Million Hours12%

Turboprops1.5 Million Hours

5%

Turbojets1.2 Million Hours

4%

Piston Rotorcraft (0.6 Million) and Other Aircraft (0.5 Million)

1.1 Million Hours4%

Turbine Rotorcraft2.2 Million Hours

7%


• Single-engine piston aircraft made themost landings, nearly 34 million, with74 percent of the landings in local flightand 26 percent in cross-country flight.

• Turbojets and turboprops, which are usedprimarily for long, cross-country fly-ing, made 91 percent and 69 percent,respectively, of their landings in cross-country vs. local flight.

• Rotorcraft had 6.7 million landings, with81 percent in local flight.

Most Aircraft FlewVMC Conditions

A sample from the survey also presents statis-tics on the meteorological conditions underwhich the general aviation fleet flew. This in-cludes the number of hours flown by visualflight rules (VFR)/daylight visual flight rules(DVFR) flight plan, no flight plan, and other/unknown flight plan, in addition to hours flownunder IFR.

Other

2 Engines, 7+ Seats


1 Engine, 4+ Seats


18.83

2.32

1.24

0.01

Fixed-wing Piston

Other


2 Engines, 1-12 Seats 1.2

0.47

0.37


Other

2 Engines 0.93

0.04

Turbine

Piston

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

2.02

4.73

Fixed-wing Turbojet

Rotorcraft

Other Aircraft

Number of Landings (millions)

Air

cra

ft T

yp

e

0.74


Figure 3

1991 General Aviation Landings by Aircraft Type The data cover the number of active gen-eral aviation aircraft and total hours flownby aircraft type during the day and night,by aircraft type under visual meteorologi-cal conditions (VMC) and by aircraft typeunder IFR in instrument meteorologicalconditions (IMC), respectively. Additionaldata provide breakdowns by manufacturer/model (M/M) group; the number of ac-tive general aviation aircraft and total hoursflown during the day and night by M/Mgroup, and the number of active generalaviation aircraft and total hours flown underIMC (based on IFR flight plan hours) andVMC (based on total hours flown) byM/M group.

Figure 4 (page 17) depicts the findings ofthe above data, showing the number ofhours flown under VMC and under IFRflight plan in IMC conditions by day andby night. Figure 5 (page 17) shows thenumber of hours flown by IFR flight plan,VFR/DVFR flight plan, no flight plan, orother/unknown flight plan.


• Approximately 87 percent of generalaviation flying took place during theday.

• Overall, 88 percent of VMC flying tookplace during the day.

• IMC flying under IFR took place 73 per-cent of the time during the day.

• Overall, 81 percent of the general avia-tion fleet’s total hours were flown inVMC conditions during the day. Theremainder of the total hours flown bythe general aviation fleet were dividedas follows: 11 percent night VMC, sixpercent under IFR in day IMC, and twopercent under IFR in night IMC.

• The results show that 45 percent of thehours flown by the general aviation fleetwere flown with no flight plan, and anadditional seven percent of the hoursflown were under some other/unknown


flight plan. Only 25 percent of the hourswere flown VFR/DVFR flight plan, and23 percent were flown IFR.

Aircraft Consume More Jet Fuel

The general aviation aircraft fleet consumed930 million gallons of fuel – 577 million gal-lons of jet fuel and 354 million gallons of avia-tion gasoline. Although data on propane fueluse were collected, they are not included be-cause the data collected were not sufficient toprovide reasonable estimates.

Figures 6 and 7 (page 18) show the fleet’s fuelconsumption rates and estimated fuel consump-tion by aircraft type, respectively. Figure 8 (page18) depicts the percentage fuel consumptionof the general aviation fleet by fuel grade.


• Of the 930 million gallons of fuel con-

1991 General Aviation Total Hours Flown byFlight Plan

1991 General Aviation Hours Flown by Weather and Light Conditions

Source: U.S. Federal Aviation Administration Source: U.S. Federal Aviation Administration

Figure 4 Figure 5

IMC-Night0.4 Million Hours

2%

IMC-Day1.2 Million Hours

6%

VMC-Night2.1 Million Hours

11%

VMC-Day16.0 Million Hours

81%

DayNight(instrument meteorological conditions)(visual meteorological conditions)

IMCVMC

Note: These estimates are based on 19.8 millionhours since data was not provided by all surveyrespondents. IMC hours listed represent IMChours flown under an IFR flight plan.

Note: These estimates are based on 28.4 million hourssince data was not provided by all surveyrespondents.

Other/Unknown Flight Plan

1.9 Million Hours7%

VFR/DVFR Flight Plan7.1 Million Hours

25%

IFR Flight Plan6.5 Million Hours

23%

No Flight Plan12.9 Million Hours

45%

(instrument flight rules)(visual flight rules)(Day VFR)

IFRVFR

DVFR

sumed by the fleet, 38 percent was avia-tion gasoline and 63 percent was jetfuel.

• Fixed-wing piston aircraft, with a lowaverage fuel consumption rate of 14 gal-lons per hour, nevertheless accountedfor approximately 37 percent (341 mil-lion gallons) of the total fuel consumedby the general aviation fleet due to theirlarge numbers. This aircraft type alsoaccounted for 97 percent of the avia-tion gasoline consumed.

• Turbojet aircraft had the highest ratesof fuel consumption – 412.7 gallons perhour for “other” turbojets, and 273.6gallons per hour for twin-engine tur-bojets. In contrast, fuel consumption ofsingle-engine piston aircraft averaged11.2 gallons per hour.

• Turbojets, which accounted for 38 per-cent of active turbine-engine aircraft in


Average Fuel Consumption Rates(Gallons per Hour) by Aircraft Type


Figure 6

Figure 8

1991 General Aviation Estimated FuelConsumption by Aircraft Type

1991 General Aviation Estimated FuelConsumption by Fuel Grade


Other

2 Engines, 7+ Seats


1 Engine, 4+ Seats


11.2

27

35.4

177.1

Other


2 Engines, 1-12 Seats 82

112

114.6


Other

2 Engines

0 50 100 150 200 250 300 350 400 450 500

273.6

412.7

Fixed-wing Turbojet

Turbine

Piston 12.4

37.6

Rotorcraft

Other Aircraft 17.3

Gallons per Hour

Fixed-wing Piston

Air

cra

ft T

yp

e

Aviation Gasoline

Jet Fuel

Note: Propane fuel data were collected but are notincluded because the data collected were notsufficient to provide reasonable estimates.


Figure 7

100 Octane66 Million Gallons7.1%

80 Octane14 Million Gallons1.5%

Jet Fuel579 Million

Gallons62.4%

Auto Gasoline17 Million Gallons

1.9%

100 Octane Low Lead251 Million

Gallons27.1%

the 1991 general aviation fleet, con-sumed 61 percent of all jet fuel usedby the general aviation fleet.

• Averaging 90 gallons per hour, tur-boprops consumed 133 million gal-lons of jet fuel (23 percent of the totaljet fuel consumed). Overall, turbopropsconsumed 14 percent of the aviationfuel.

• Of the 354 million gallons of aviationgasoline consumed by fixed-wing pis-ton aircraft, approximately 14 milliongallons were 80 octane gasoline, 66million gallons were 100 octane gaso-line, 251 million gallons were 100 oc-tane low lead gasoline and 17 milliongallons were automobile gasoline. ♦

Single-engine Piston

231 Million Gallons24.8%

Multi-engine Piston


Turboprops133 Million

Gallons14.3%

Turbojets351 Million

Gallons37.8%

Piston Rotorcraft (8 Million) and Other Aircraft (5 Million)


Turbine Rotorcraft92 Million Gallons

9.9%

Aviation Gasoline

Jet Fuel


Reports Received at FSFJerry Lederer Aviation Safety Library

Springfield, Virginia, U.S. Available throughthe National Technical Information Service,[1992]. 69 p. in various pagings.

Keywords1. Aviation Medicine — United States —

Indexes.2. United States — Office of Aviation Medi-

cine — Bibliography.

Summary: This reference source is an index toCivil Aeromedical Research Institute Reportsfrom 1961-1963 and Office of Aviation Medi-cine Reports from 1964-1991. It is intended asa guide to those engaged in aviation medicineand related activities. Its three indexes list allpublished FAA aviation medicine reports chro-nologically, alphabetically by author and al-phabetically by subject. The chronological in-dex also provides document numbers for NTIS.A supplement for 1992 reports is given in chro-nological order.[modified abstract]

TCAS Incident Reports Analysis/ National Aero-nautics and Space Administration (NASA), Avia-t ion Sa fe ty Repor t ing Sys tem (ASRS) ,Mountain View, California. Available throughthe National Aeronautics and Space Adminis-tration, Ames Research Center, [1992]. 123 p.in various pagings.

Keywords1. Airplanes — Collision Avoidance.

Winter Operations Guidance for Air Carriersand Other Adverse Weather Topics/ preparedby Flight Standards Service. Washington,D.C. U.S. Federal Aviation Administration,Office of the Administrator, 1992. 376 p.in various pagings. ill .

Keywords1. Airplanes — Cold Weather Operation.2. Airplanes — Ice Prevention.3. Airplanes — Climatic Factors.

Summary: This second update to the FAA winteroperations publication includes current advi-sory circulars, maintenance bulletins, air car-rier bulletins as well as articles from manu-facturers’ and air carriers’ publications. Thiscomprehensive guide covers general winterweather operating for ramp operations, air-craft ground deicing, fuel system icing pre-cautions, inflight considerations, takeoff andlanding considerations, helicopters and ad-verse weather, such as wind shear and thun-derstorms. This 1992 edition includes FAAadvisory circular Pilot Guide/Large Aircraft GroundDeicing, and a comprehensive report, TailplaneIcing. Sources for associated material and anindex are included.

Index of FAA Office of Aviation Medicine Reports:1961 through 1991/ William E. Collins, MichaelE. Wayda. Washington, D.C. U.S. Federal AviationAdministration, Office of Aviation Medicine;

Winter Operations Updated by U.S.

byEditorial Staff


tent is to identify how communication is ac-complished within the maintenance organiza-tion. Chapter three offers a study of the main-tenance technician in inspection. Its approachis to determine typical human/system mis-matches to guide both future research and short-term human factors implementation by sys-tem participants. Chapter four documents astudy of advanced technology for maintenancetraining. This study reports the status of aproject to support the application of advancedtechnology systems for aircraft maintenancetraining. The final study, chapter five, pre-sents research on job performance aids. Thisresearch was designed to provide informationfor government and industry managers in theirefforts to access the utility and implementa-tion of job-aiding technology. An executivesummary, chapter one, is also provided andincludes summaries of subsequent phase plansfor each study, information on other researcha c t i v i t y i n h u m a n f a c t o r s a n dreferences.[Modified executive summary]

Aviation Safety: Progress on FAA Safety Indica-tors Program Slow and Challenges Remain: Re-port to the Chair, Subcommittee on Govern-ment Activities and Transportation, Committeeon Government Operations, House of Repre-sentatives General Accounting Office. Wash-ington, D.C. General Accounting Office, [1992].Report No. GAO/IMTEC-92-57. 40 p.: ill. In-cludes bibliographical references.

Keywords1. Aeronautics — United States — Safety

Measures.2. Air Traffic Control — United States.

Summary: This report gives the backgroundand an assessment of the FAA’s Safety Indica-tors Program. Two components of the programare described: development of safety indica-tors, or categories of safety measurements forthe aviation system; and development of a de-cision support system, or computer analysistool, to obtain information from numerous stand-alone databases for sophisticated analysis andpresentation. While the program’s goal was todevelop a consistent set of safety measuresand computer analysis capability to presentquickly and vividly the state of aviation safety,

2. Aeronautics — Safety Measures — Air TrafficControl.

3. Traffic Alert and Collision-avoidanceSystem (TCAS).

Summary: At the request of the FAA Office ofAviation Safety and the National Transporta-tion Safety Board (NTSB) for data from ASRS,this study of ASRS reports provides a generaloverview of traffic alert and collision-avoid-ance system (TCAS) data in the ASRS data-base for the period of January 1, 1988, to March1, 1992. This study further provides a morefocused analysis of a random sample of TCASincident reports relating to the same time pe-riod. The study is presented in three sections:a review of the 1,124 TCAS-related reports inthe ASRS database, an analysis of 170 ran-domly sampled TCAS reports and a discus-sion of key findings. An appendix containingreports illustrative of the kinds of TCAS-re-lated events reported to the ASRS is also in-cluded. While the data collected from thesevoluntarily submitted incident reports reflectreporting biases, the authors believe that thereal power of the ASRS lies in the report nar-ratives where pilots, controllers and others dis-close aviation safety incidents and situationsin detail.[modified synopsis and caveat]

Human Factors in Aviat ion Maintenance :Phase 1, Progress Report/ William T. Shepherd... [et al.] Washington, D.C. U.S. Federal Avia-tion Administration, Office of Aviation Medi-cine; Springfield, Virginia, U.S. Available throughthe National Technical Information Service,[1991]. 158 p.: charts. Includes bibliographicalreferences.

Keywords1. Av i a t i o n M e c h a n i c s ( P e r s o n s ) —

Psychology.2. Airplanes — Maintenance and Repair.3. Aeronautics — Human Factors.

Summary: This interim report on human fac-tors research in aviation maintenance includesfour studies, each dedicated to factors associ-ated with the aviation maintenance techni-cian and other personnel supporting the main-tenance system goals. Chapter two presents astudy of the maintenance organization. Its in-


ineffective user involvement and unclear man-agement commitment have contributed to theagency’s inability to complete the program.Because a wide difference of opinion as towhat constitutes acceptable measures of airtraffic safety resulted between those develop-ing the indicators and many of those who weretargeted to use them, progress in developingthe key safety indicators has been slow.

According to the report, clear top-level FAAmanagement backing of the program, throughstatements on the participants responsibilitiesand authority, is essential in fostering the co-operation needed to complete the developmentof the safety indicators. The report further statesthat although the National Aviation Safety DataCenter (NASDC) has made some progress inthe development of the automated decisionsupport system intended to import, validateand integrate data imported from the FAA’sdiverse collection of existing safety-relateddatabases, development of the capability toimport data from outside of NASDC’s owndivision has yet to begin. According to thereport, this capability is critical to the successof the decision support system since NASDCestimates that information will be extractedfrom as many as 70 aviation data sources. Theunreliability of many of FAA’s safety-relateddatabases is also a significant challenge. Ac-cording to the report, many of these databasesare inaccurate, inconsistent and often incom-patible. In light of this testimony and previ-ous recommendations, no further recommen-dations were offered. However, the report saidsuccess in the development of the safety indi-cators program and decision support system

depend heavily on the FAA’s effectively ad-dressing the issues of source data reliability,refinement of safety indicators and user in-volvement in and management commitmentto the development of both the safety indica-tors and the decision support system.[modifiedresults in brief and conclusions]

New Reference Materials

Advisory Circular 21-32, 10/14/92, Control of PartsShipped prior to Type Certificate Issuance. Wash-ington, D.C. U. S. Federal Aviation Adminis-tration (FAA), 1992. 3p.

Summary: This advisory circular provides in-formation and guidance concerning the con-trol of parts to be shipped by manufacturerswith an approved production inspection sys-tem (APIS) or production certificates (PC) inadvance of type certification of a new aircraft,aircraft engine or propeller (product).♦

*U.S. Department Of CommerceNational Technical Information Service (NTIS)Springfield, VA 22161 U.S.Telephone: (703) 487-4780

**U.S. General Accounting Office (GAO)Post Office Box 6012Gaithersburg, MD 20877 U.S.Telephone: (202) 275-6241

Updated Reference Materials (Advisory Circulars, U.S. FAA)

Numbers Month/Year Subject

21.17-1A September 1992 Type Certification — Airships (Cancels AC 21.17-1, dated9/20/87).

91-62 October 1992 Use of Child Seats in Aircraft (Cancels AC 91-62, dated2/26/85).

21.303-2H October 1992 Parts Manufacturer Approvals 1992 (Microfiche listing ofparts manufacturer approvals valid through August 1992;cancels AC 21.303-2G, dated 10/21/88).


Accident/Incident Briefs

This information is intended to provide an aware-ness of problem areas through which such occur-rences may be prevented in the future. Accident/incident briefs are based on preliminary informa-tion from government agencies, aviation organiza-tions, press information and other sources. Thisinformation may not be entirely accurate.

Strong Gusts Add ToGo-around Woes

Airbus A310. Substantial damage. No injuries.

The Airbus was on final approach to an islandoff the coast of Greece when tower controllersissued a gust warning.

Winds at the time were reported to be 090degrees through 140 degrees, with 20-knot windsgusting to 32 knots. Approach plates for theairport cautioned pilots about landing withwinds from the south to southeast at windspeeds above 15 knots.

The first officer, who was flying the aircraft,continued the approach. Wind conditions re-quired a firm touchdown without excessiveflare. Because of a higher than normal approach

speed, the aircraft entered the flare about eightfeet off the runway. The captain called for ago-around and took the controls.

After the go-around was initiated, the A310was caught by a violent wind gust and itsright wing struck the ground. The impact dam-aged the wingtip, slats, flap fairings and thewing underside.

The crew recovered from the gust and contin-ued the go-around. The climbout and subse-quent landing were uneventful.

After the aircraft landed, the airport was closedbecause of excessive wind conditions. It tookfive days to repair the aircraft.

Water Leak Causes EmergencyLanding

Boeing 737. Minor damage. No injuries.

Shortly after takeoff, the cabin crew reportedwater streaming out of the forward toilet.

A few moments later, several aircraft systemsfailed, including autopilots A and B, altim-eters, mach trim, transponder and automaticpressurization. Warning flags appeared on thecaptain’s instrument panel and on the firstofficer ’s as well.

The aircraft returned safely to its departureairport, but it took several minutes to openthe cabin doors because of a faulty pressuriza-tion system. It was determined that electricalsystem damage was caused by a cracked wa-ter fountain in the forward toilet.

Air Carrier

Liquor Bottles Cripple Aircraft

byEditorial Staff

Air Carrier


Stashed Liquor Cripples Controls,Cancels Flight

McDonne l l Douglas MD-80 . No damage .No injuries.

The MD-80 with 77 passengers on board waspreparing to depart from a European airportwhen the controls jammed.

Mechanics summoned to the aircraft foundthree bags of bottled whiskey and vodka hid-den under a hatch. It was determined that thebottles were smuggled aboard by the pilot,who was subsequently grounded for 10 months.The bags had prevented the controls frommoving freely. The flight was canceled andthe passengers were transported to their des-tination by other operators.

Icing Forces Aborted Takeoff

Cessna 402. Substantial damage. No injuries.

The takeoff was begun following deicing, taxiingand runup procedures. Acceleration and en-gine indications were normal.

Rotation at 95 knots, however, resulted in astall warning. The pilot determined that thetwin-engine aircraft would not remain airborneand aborted the takeoff. The crew was unableto stop the aircraft before it reached the endof the runway, and it slid down a slope andacross a ditch. The landing gear was shearedoff, but passengers and crew were able to evacu-ate successfully.

It was determined that minor deposits of snowon the top side of the wings and water frommelting snow refreezing on the lower sidesdisturbed the airflow. A contributing factor in

the daylight incident was the crew’s lack ofprecise information on the actual glycol/wa-ter mixture of the deicing fluid.

Takeoff Without Elevator ControlEnds Abruptly

Mitsubishi Diamond. Substantia l damage.No injuries.

The night takeoff was aborted after the twin-turbofan aircraft experienced elevator controlproblems. The pilot was not able to stop theaircraft on the runway and elected to turn rightinto a recently sown field. The nose gear col-lapsed, but the crew and two passengers wereable to exit the aircraft safely.

It was determined that the crew had forgottento remove the elevator gust lock during thepreflight. Records indicated the crew had per-formed poorly together on earlier flights andduring check rides. An investigation concludedthat poor crew coordination and unsatisfac-tory training were contributing factors in theaccident.

Unstabilized ApproachDooms Commuter

Antonov An-24. Aircraft destroyed. Forty-one fa-talities.

About two kilometers from the runway, thetwin-engine turboprop began to deviate to theright of its proper approach course. A towercontroller queried the crew about the devia-tion but received no response.

At a distance of .93 miles (1.5 kilometers) fromthe runway threshold, the controller orderedthe flight to go-around. A few moments later,the aircraft crashed about 2,640 feet (800 meters)from the runway, about 1,980 feet (600 meters)to the right of the center line.

The four-man crew and 37 passengers werekilled in the evening crash. A post-crash in-vestigation found a layer of ice up to .6 inch(15 mm) thick on the horizontal stabilizer.

Air Taxi/CommuterAir TaxiCommuter


Darkness, Fatigue Disorient Pilot

Beech 90 King Air. Aircraft destroyed. Five fatali-ties.

Shortly after the night takeoff, the twin-turboprop aircraft collided with a line of treeslocated about 1980 feet (600 meters) from theend of the runway and slightly left of the cen-ter line. The King Air struck the trees in awings-level attitude in a very shallow descent.

After colliding with the trees, the aircraft im-pacted the ground and caught fire. The pilotand four passengers were killed.

A post-crash inquiry determined that the nightwas very dark with no moon and no visiblehorizon. Evidence suggested that the pilot mayhave been suffering from fatigue due to a heavywork schedule. The pilot had received no for-mal human factors instruction during his in-strument training, and there was evidence thatthe pilot may have experienced visual illu-sions and vertigo after takeoff. All aircraft sys-tems were functioning normally at the time ofthe crash.

Hard Landing Follows TwoMissed Approaches

Cessna 340. Substantial damage. No injuries.

The twin-engine Cessna 340 was attemptingto land under instrument meteorological con-ditions at night with a 7/8 cloudbase at 200feet (60 meters).

The airport was not equipped with instru-ment approach facilities, only low-intensityrunway lights. The pilot made a hard land-

ing after two missed approaches. The air-craft was covered by clear ice and after landing,cracks were found in the main beam of theleft wing and other structural parts of bothwings.

Missed Approach Spells Tragedy

Cessna 421. Aircraft destroyed. Five fatalities.

The twin-engine Cessna was on a daylight in-strument approach in meteorological condi-tions below minimums.

Following a missed approach, the aircraft struckthe ground in a left spiral and in a 80-degreenose-down attitude. The pilot and four pas-sengers were killed in the crash, which wasattributed to pilot spatial disorientation dur-ing the go-around.

Turbulence Knocks Jodel Down

Jodel D120A. Substantial damage. Two injuries.

The pilot of the single-engine Jodel was prac-ticing touch-and-go maneuvers.

During the takeoff phase, violent turbulencegenerated by wind and trees at the end of therunway caused the aircraft to stall. A wingtouched the ground and the aircraft cartwheeledonto its nose. The front of the cockpit sus-tained severe damage, and the wings weretorn from the fuselage. The pilot and passen-ger suffered serious injuries. The private pilothad logged a total of 265 flying hours.

Cessna Loses Contest withStrong Crosswind

Corporate ExecutiveCorporateExecutive

Other GeneralAviation

OtherGeneralAviation


Oil Loss Leads to Crash Landing

Bell 47J2. Aircraft destroyed. Four fatalities.

The pilot of the Bell 47J had begun executingan emergency landing following loss of en-gine power.

Witnesses observed the helicopter flying at 500feet (150 meters) above ground level when theengine appeared to backfire and begin sput-tering. The aircraft descended rapidly in a leftturn and struck trees before impacting theground. The helicopter then rolled down ahill. The pilot and three passengers were killed.

A post-crash investigation revealed that theengine oil dip stick was not secure. A streak ofengine oil extended from the filler neck to thetail boom. Two quarts of oil were drained fromthe engine oil system.

Disassembly of the engine revealed damageindicative of oil starvation on all bearing sur-faces, but the engine had not seized. The acci-dent occurred in daylight visual meteorologi-cal conditions. There was no fire.

Earthly Ties Too Strong forSchweizer

S c h w e i z e r 2 6 9 C . S u b s t a n t i a l d a m a g e .No injuries.

The helicopter was attempting to take off froma platform on a fishing vessel in the PacificOcean for a routine fish-spotting mission.

After the takeoff was initiated, the aircraft sud-denly rolled over on the deck.

It was determined that the pilot had attemptedto take off with one of the skids tied to theship’s deck. The pilot escaped injury. ♦

RotorcraftRotorcraftCessna 172. Substantial damage. Two seriousinjuries.

The pilot of the single-engine Cessna had beenforced to execute two missed daylight approachesbecause of high winds.

On the third attempt, an increased power set-ting was used to counter wind effect. As theaircraft turned on final approach, it stalled atan altitude of 300 feet (90 meters). There wasinsufficient altitude for recovery, and the air-craft impacted the ground in a wings-level,nose-down attitude. The nose cowling, rightwing, wheels and elevators were severely dam-aged. The pilot and a passenger were seri-ously injured.

A post-crash investigation concluded that theaccident was attributed to the pilot attempt-ing to land in a crosswind greater than accept-able for the aircraft type. The pattern flownalso contributed to the accident by increasingthe ground speed on the base leg/final ap-proach turn, which required an increased bankangle to execute the turn. The private pilothad logged 650 flying hours.

Wind Shift Adds Drama toShort Field Landing

C e s s n a 1 8 2 . A i r c r a f t d e s t r o y e d .Three injuries.

The single-engine Cessna encountered a sud-den wind shift while landing at a short ruralairfield.

The aircraft failed to reduce speed sufficientlyand overran the runway, coming to rest in apheasant coop. The aircraft was destroyed be-yond economic repair. The pilot and two pas-sengers suffered serious injuries. The privatepilot had logged 444 hours flying time.

Rejected Takeoffs: Causes, Problems and Consequences · controller as a safety factor in RTO events...

Documents

Transcript of Rejected Takeoffs: Causes, Problems and Consequences · controller as a safety factor in RTO events...