JANUARY–MARCH 2010

Reducing the Risk of Runway Excursions (Page 4) Closing the Gap Between Accident Investigation and Training (Page 8) How Significant Is the Inflight Loss of Control Threat? (Page 12) Bringing the Worldwide Helicopter Accident Rate Down by 80% (Page 16) Sifting Lessons from the Ashes (Page 22) CONTENTS

Volume 43, Number 1 Publisher Frank Del Gandio Editorial Advisor Richard B. Stone FEATURES Editor Esperison Martinez Design Editor William A. Ford 4 Reducing the Risk of Runway Excursions Associate Editor Susan Fager By Jim Burin, Director of Technical Programs, Flight Safety Foundation— Annual Report Editor Ron Schleede During the past 14 years, the number of runway excursion accidents is more ISASI Forum (ISSN 1088-8128) is published than 40 times the number of runway incursion accidents, and more than quarterly by International Society of Air 100 times the number of runway confusion accidents. Safety Investigators. Opinions expressed by authors do not necessarily represent official 8 Closing the Gap Between Accident ISASI position or policy. Editorial Offices: Park Center, 107 East Investigation and Training Holly Avenue, Suite 11, Sterling, VA By Michael Poole and Lou Németh, CAE Flightscape—Full flight simulators 20164-5405. Telephone (703) 430-9668. are used to train pilots to fly airplanes and to carry out emergency procedures Fax (703) 430-4970. E-mail address isasi@ erols.com; for editor, espmart@comcast. fastidiously. But are pilots, and should pilots, be trained to prevent accidents? net. Internet website: www.isasi.org. ISASI Forum is not responsible for unso- 12 How Significant Is the Inflight Loss licited manuscripts, photographs, or other materials. Unsolicited materials will be Of Control Threat? returned only if submitted with a self-ad- By Capt. John M. Cox and Capt. Jack H. Casey, Safety Operating Systems, L.L.C.— dressed, stamped envelope. ISASI Forum At all times, manufacturer recommendations for proper aircraft operation are reserves the right to reject, delete, sum- controlling. marize, or edit for space considerations any submitted article. To facilitate editorial production processes, American-English 16 Bringing the Worldwide Helicopter spelling of words will be used. Accident Rate Down by 80% Copyright © 2010—International Society of By Jack Drake, Helicopter Association International—The International Air Safety Investigators, all rights reserved. Publication in any form prohibited without Helicopter Safety Team (IHST) takes the proactive attitude that anyone’s permission. ISASI Forum registered U.S. helicopter accident belongs to all of us. Patent and T.M. Office. Opinions expressed by authors do not necessarily represent of- 22 Sifting Lessons from the Ashes ficial ISASI position or policy. Permission to By Ludwig Benner, Jr., Principal, Starline Software Ltd., and Ira J. Rimson, reprint is available upon application to the editorial offices. Forensic Engineer—Current processes for identifying, defining, communicating, Publisher’s Editorial Profile: ISASI and acting on lessons to be learned are inadequate to take advantage of the Forum is printed in the United States opportunities offered by investigated accidents. and published for professional air safety investigators who are members of the In- ternational Society of Air Safety Investiga- tors. Editorial content emphasizes accident DEPARTMENTS investigation findings, investigative tech- 2 Contents niques and experiences, regulatory issues, industry accident prevention developments, 3 President’s View and ISASI and member involvement and 26 ISASI RoundUp information. 30 ISASI Information Subscriptions: Subscription to members 32 Who’s Who—A brief corporate member profile of is provided as a portion of dues. Rate for nonmembers (Domestic and Canada )is Australian and International Pilots Association US$28.00; Rate for nonmember interna- tional is $US30.00. Rate for all libraries and schools is $US24. 00. For subscription bout the over information, call (703) 430-9668. Additional A C or replacement ISASI Forum issues: Do- A very typical excursion accident: i.e., unstable approach, land long and fast, then mestic and Canada US$3.00; international engine issues happen. Narrative: Boeing 747-F-GITA operated Flight AF072, Paris- member $US4.00; domestic and Canada Los Angeles-Tahiti. Weather on Sept. 12, 1993, was good when the flight approached nonmember $US 6.00; and international nonmember $US8.00. Tahiti, French Polynesia, at night. The aircraft positioned for a Runway 22 VOR-DME approach. At 21:05 the aircraft touched down at a speed of 168 knots. Two seconds later, No. 1 engine power increased to 107% N1. Because of this, the spoilers did not deploy and the automatic brake disarmed. Reverse thrust was used on all remaining engines. Because of difficulties due to thrust asymmetry, the No. 4 engine thrust reverser was cancelled. The aircraft overran the runway and ended up in a lagoon.

2 • ISASI Forum January–March 2010 PRESIDENT’S VIEW

Looking Back and Ahead By Frank Del Gandio, ISASI President On this page of the January–March 2000 issue fund balance (net worth) was $31,205. At the close of 2008, it of the Forum, I wrote of what “we might face was $168,512. The increase is greatly owed to the purchase of in the coming decade.” Today, I look back at and mortgage pay off of the ISASI office condominium and some of those thoughts and let you judge their to the great successes of the annual international seminars germaneness. Also, I pen a few thoughts on conducted by the Society. our Society’s present status and its future However, not every year or every seminar is a financial suc- well-being. cess. The recent seminar in Orlando, Fla., is a case in point. It “A new millennium,” I wrote, “tempts us to speak of un- was, from the prospective of the delegates, highly successful. imaginable levels of aviation safety…[but] accidents will not The papers presented were the best to date. This was attested become a thing of the past.” I noted that we, as investigators, to me by at least a half dozen delegates. The keynote speeches would encounter new as well as some of the usual suspects: “… controlled aircraft flying into high terrain or into level ground, s your president, I want you to know approach and landing accidents, poor communication between that your Society is financially solvent crewmembers or between pilots and ATC, judgment mistakes A by pilots, maintenance issues, and small errors of omission or and that you, as a member, will not see commission that invite other small errors and eventually the any reduction in services or benefits. crew is behind the airplane and never catches up.” given by the Honorable Deborah Hersman, Chairman of the I also noted, “We also will face new challenges or issues National Transportation Safety Board, and Paul-Louis that, at most, are only emerging today. For example, as air- Arslanian, Director of BEA, were well received and very ap- craft become ever more automated and integrated, intellectu- propriate. Disney’s Coronado Springs Resort met all our needs, ally there is a loss in the understanding of how all systems in including lodging and food. Of course, almost everyone got to an aircraft interact…. Other challenges…include the need to visit the main attraction, Disney World. ISASI was supplied strike a balance with the rapid advances in media technology with passes for the delegates to use at night. I received numer- that speeds reporting…. We need a working understanding ous accolades regarding the tutorials, which were attended by with the press.…Yet, we must ensure the continued integrity 135 delegates. The companion’s tours were memorable, exciting, of complete and thorough investigations….” and informative and well-received by the attendees. Lastly, I wrote “We can also expect more direct and instant However, the financial aspects did not work out as well. involvement by national political officials in many countries…. Traditionally ISASI seminars generate funds in excess of The political pressures that news media coverage creates for expenses. It is this overage of funds that helps keep ISASI fi- visible public officials suggest that the political complexity of nancially solvent. The Orlando seminar resulted in a loss of ap- air safety investigation may accelerate even faster than the proximately $39,000. There were many contributing factors to technical and organizational complexity.” this loss. The main factor was the economy. When the contract Now let me turn to Society matters of today and the future. for the hotel was signed in March 2006, the economy was very We have become a much more viable professional society in robust and the anticipated attendance was higher than actu- the past 10 years. Our international Reachout Workshop pro- ally realized. The loss resulted from rooms not utilized and, of gram has instructed more than 2,010 persons in 22 countries. course, the food associated with the missing attendees. We have become a voice in ICAO’s AIG meetings. Our Kapus- That said, I want to assure you that ISASI will never be tin Memorial Scholarship program has been highly successful. confronted with this situation again. We have taken initiatives We have placed all our historical library records into a digital to revise the seminar manual and other associated policies. library managed by Embry-Riddle Aeronautical University, Room guarantees will be realistic and reduced to levels below and most recently, we published and placed on our website a the anticipated attendees. The ISASI Executive Council has 30-page booklet titled ISASI Guidelines for Investigation of implemented a number of cost-saving measures to reduce our Human Factors in Accidents or Incidents, developed in con- overall expenses for the next year. Additionally, I encourage cert with the Transport Safety Board of Canada. The booklet each of you to assist in the recruitment and retention of poten- is designed as an aid for accident investigators-in-charge spe- tial and current corporate and individual members. cifically on the subject of human factors. These advancements As your president, I want you to know that your Society is are but a few of the Society’s accomplishments. financially solvent and that you, as a member, will not see any In the past decade, the Society placed itself on a firm mem- reduction in services or benefits. My best wishes to you and bership and financial footing. At the close of 1999, our final yours for a very fruitful New Year. ◆

January–March 2010 ISASI Forum • 3

(continued on page 30) During the past 14, years the number of runway excursion accidents is more than 40 times the number of runway incursion accidents, and more than 100 times the number of runway confusion accidents. Reducing the Risk of Runway Excursions

By Jim Burin, Director of Technical Programs, Flight Safety Foundation.

(This article is adapted with permission from the author’s paper entitled Reducing the Risk of Runway Excursions presented Table 2. Runway-Related Accidents for at the ISASI 2009 seminar held in Orlan- Table 1. Total Commercial Transport Commercial Turbojet and Turboprop do, Fla., Sept. 14-18, 2009, which carried Accidents, 1995-2008 Aircraft the theme “Accident Prevention Beyond Investigations.” The full presentation, reviewed the three areas of runway safety: have averaged one accident per year. including cited references to support the runway incursions, runway confusion, and Figure 1 shows that the largest portion points made, can be found on the ISASI runway excursions. Ultimately, the RSI of runway-related accidents is, by far, website at www.isasi.org.—Editor) Group determined that it would be most excursion accidents. effective to focus its efforts on reducing Forty-one of the 431 runway accidents he Flight Safety Foundation, the risk of runway excursions. All data involved fatalities. Excursion accidents at the request of several in- used in this document are taken from the accounted for 34 of those 41 fatal acci- ternational aviation organiza- RSI report Flight Safety Foundation dents, or 83% of all fatal runway-related tions, initiated the Runway Reducing the Risk of Runway Excursions Safety Initiative (RSI) project (May 2009). to address the challenge of RSI defined a runway excursion as Only a small Trunway safety. This was an international when an aircraft on the runway surface effort with participants representing the departs the end or the side of the runway percentage of runway full spectrum of stakeholders from the surface. Runway excursions can occur on aviation community. The effort initially takeoff or on landing. They consist of two excursion accidents types of events: 1. A veer-off is a runway are fatal. However, Jim Burin has 33 years excursion in which an aircraft departs the of experience in the avia- side of a runway. 2. An overrun is a runway since the overall tion safety field. He holds excursion in which an aircraft departs the number of runway an MS degree in end of a runway. systems analysis and is During the 14-year period from 1995- excursion accidents a 30-year career retired 2008, commercial transport aircraft expe- is so high, that small U.S. Navy captain who rienced a total of 1,429 accidents involving has commanded an attack squadron major or substantial damage (see Table 1). percentage accounts and a carrier air wing. Prior to joining Of those, 431 accidents (30%) were runway for a large number of the Flight Safety Foundation, he was the related. The specific RSI focus on excur- director of the School of Aviation sion accidents was driven by the fact that fatalities. Safety in Monterey, Calif., where he was of the 431 runway-related accidents, 417, responsible for the safety training of or 97%, were runway excursions. 650 Navy, Marine, Coast Guard, and The number of runway excursion ac- international safety officers each year. cidents is more than 40 times the number As the FSF director of technical pro- of runway incursion accidents, and more grams, he organizes and oversees safety than 100 times the number of runway con- committees and manages safety-related fusion accidents (see Table 2). During the conferences and research. He is also the past 14 years, there has been an average of chairman of the Foundation’s interna- almost 30 runway excursion accidents per tional ALAR (approach and landing year for commercial aircraft, while runway accident reduction) effort. incursion and confusion accidents combined

4 • ISASI Forum January–March 2010 During the past 14, years the number of runway excursion accidents is more than 40 times the number of runway incursion accidents, and more than 100 times the number of runway confusion accidents. Reducing the Risk of Runway Excursions

Figure 1. Proportions of Runway-Related Figure 5. Landing Excursions, By Type Accidents for Commercial Turbojet and Figure 3. Runway Excursions, By Type Turboprop Aircraft Air traffic management/air traffic con- trol has two primary roles in reducing the risk of runway excursions: to provide air traffic services that give flight crews the opportunity to fly a stabilized approach and to ensure that aircrews are given the best available information on environ- mental and runway conditions in a timely manner. Finally, the regulator needs to encour- Figure 2. Proportions of Fatal and Non- Figure 4. Takeoff Excursions, By Type age and provide more approaches with Fatal Runway Accidents vertical guidance, since these assist in enabling stabilized approaches. They accidents. In general, the likelihood of The manufacturers do a great job pro- also need to be sure that aircrews are fatalities in a runway-related accident is viding the operators with safe and reliable given the best possible information from greater in incursion and confusion acci- aircraft. They also provide data and proce- the manufacturers for operations under dents. However, the much greater number dures that the crews will need in most day- all conditions. Regulators should also re- of runway excursion accidents results in to-day operations. However, without good quire some universal system for measur- a substantially greater number of fatal data on how the aircraft will perform under ing and reporting runway conditions. excursion accidents (see Figure 2). all runway conditions, some landings have Only a small percentage of runway ex- the potential of becoming physics experi- In-depth study cursion accidents are fatal. However, since ments. Operators must have and monitor An in-depth data study was conducted the overall number of runway excursion stabilized approach criteria, since the data of all runway excursion accidents from accidents is so high, that small percentage clearly show that unstabilized approaches 1995 through March 2008 to investigate accounts for a large number of fatalities. are a primary risk factor in landing excur- the causes of runway excursion accidents During the 14-year period, 712 people died sions. Operators also need to have a true no- and to identify the high-risk areas. Fol- in runway excursion accidents, while run- fault go-around policy because the leading lowing are some of the basic data from way incursions accounted for 129 fatalities risk factor in landing excursions is failure the study. and runway confusion accidents accounted to go around when warranted. Landing excursions outnumber take- for 132 fatalities. Airports have many issues to address off excursions approximately 4 to 1 (see in the runway excursion area, including Figure 3). Who is responsible? airport design, lighting, approach aids, Almost two-thirds of the takeoff excur- Who is responsible to address the chal- runway design (e.g., crowned, grooved, sions are overruns (see Figure 4). lenge of runway excursions? The answer porous), markings and signage, runway Landing excursion overruns and veer- is everyone aircraft manufacturers, opera- cleaning and clearing, runway condition offs occur at nearly the same rate (see tors (both aircrews and management), air- measurement, runway end safety areas, Figure 5). ports, air traffic control, and regulators. and aircraft rescue and firefighting. Among aircraft fleet types, turboprops

January–March 2010 ISASI Forum • 5 Reducing the Risk of Runway Excursions

tion, and ineffective braking (e.g., hydroplaning, con- taminated runway). Three of the top 5 risk factors deal with elements of a stabilized approach.

Risk factor interactions The risk of a runway excur- sion increases when more Figure 6. Takeoff Excursions, By Fleet than one risk factor is pres- Composition ent. Multiple risk factors Figure 8. Takeoff Excursion Risk Factors create a synergistic effect (i.e., two risk factors more than double the risk). Risk factor interactions present the possibility of many as- sociations between various contributing factors, but determining whether any pair of associated factors has a causal connection Figure 7. Landing Excursions, by Fleet would require more detailed Composition study and analysis. As an example, risk fac- are involved in the largest percentage of tor interactions for land- takeoff excursions, followed closely by jet ing overruns show strong Figure 9. Landing Excursion Top Risk Factors transports (see Figure 6). associations between “go- For landing excursions, the proportions around not conducted” and other factors responsible for the low number of runway between jet transports and turboprops such as “unstabilized approaches,” “long/ incursion accidents. The specific goal of were approximately reversed—jets were fast landings,” “runway contamination,” the RSI team was to provide data that involved in more excursions than turbo- and “hard/bounced landings.” Logically, highlight the high-risk areas of runway props (see Figure 7). these factors may have a causal connection excursions and to provide interventions The data were analyzed to identify the to each other that significantly increases and mitigations that can reduce those most common risk factors, both in takeoff the probability of a runway excursion risks. The RSI effort brought together excursions (see Figure 8) and landing ex- accident. In looking at some risk factor multiple disciplines that included aircraft cursions (see Figure 9). More than one risk interactions for takeoff excursions, the manufacturers, operators, management, factor could be assigned to an accident. risk factor interactions suggest that there pilots, regulators, researchers, airports, The most common risk factor in takeoff might be interesting associations between and air traffic management organizations. excursions was a rejected takeoff (RTO) engine power loss and aborts initiated It used the expertise and experience of all initiated at a speed greater than V1. Loss above V1, as well as an association between the stakeholders to address the challenge of pilot directional control was the next these high-speed aborts and the presence of runway excursions. most common, followed by rejecting the of runway contaminants. takeoff before V1 is reached. This is of 1. A mishandled rejected takeoff (RTO) concern since aborting prior to V1 should Conclusions and recommendations increases the risk of takeoff runway result in a successful RTO. The RSI team fully supports the many out- excursion: For landing excursions, the top risk standing activities being conducted around • Operators should emphasize and train factors were go-arounds not conducted, the world by organizations like the FAA, for proper execution of the RTO decision. touchdown long, landing gear malfunc- ICAO, and Eurocontrol that have been • Training should emphasize recognition

6 • ISASI Forum January–March 2010 of takeoff rejection issues. runway condition reporting should be —Sudden loss or degradation of thrust. developed to reduce the risk of runway —Tire and other mechanical failures. excursions. The Runway Safety —Flap and spoiler configuration issues. • Manufacturers should provide ap- Initiative effort brought • Training should emphasize directional propriate operational and performance control during deceleration. information to operators that accounts for together multiple • CRM and adherence to SOPs are es- the spectrum of runway conditions they disciplines that included sential in time-critical situations such as might experience. RTOs. aircraft manufacturers, 6. Thrust reverser issues increase the operators, management, 2.Takeoff performance calculation er- risk of runway excursions: rors increase the risk of a takeoff runway • Flight crews should be prepared for pilots, regulators, excursion: mechanical malfunctions and asymmetric researchers, airports, • Operators should have a process to deployment. ensure a proper weight and balance, in- • Flight crew application of reverse and air traffic manage­ cluding error detection. thrust is most effective at high speeds. ment organizations. It • Operators should have a process to en- sure accurate takeoff performance data. 7. Combinations of risk factors (such as used the expertise and abnormal winds plus contaminated run- experience of all the 3. Unstable approaches increase the risk ways or unstable approaches plus thrust of landing runway excursions: reverser issues) synergistically increase stakeholders to address • Operators should define, publish, and the risk of runway excursions: the challenge of runway train the elements of a stabilized ap- • Flight crews should use a runway ex- proach. cursion risk awareness tool for each land- excursions. • Flight crews should recognize that fast ing to increase their awareness of the risks and high on approach, high at threshold, that may lead to a runway excursion. and fast, long, and hard touchdowns are major factors leading to landing excur- 8. Establishing and adhering to standard sions. operating procedures (SOPs) will en- • ATC/ATM personnel should assist hance flight crew decision-making and aircrews in meeting stabilized approach reduce the risk of runway excursions: criteria. • Management and flight crews should mutually develop SOPs. 4. Failure to recognize the need for and to • SOPs should be regularly reviewed execute a go-around is a major contribu- and updated by a management and flight tor to runway excursion accidents: crew team. • Operator policy should dictate a go- • Aircraft rescue and firefighting (ARFF) around if an approach does not meet the 9. The survivability of a runway excur­ personnel should be trained and available stabilized approach criteria. sion depends on the energy of the aircraft at all times during flight operations. • Operators should implement and sup- as it leaves the runway surface and the port no-fault go-around policies. terrain and any obstacles it will encoun- 10. Universal standards related to the • Training should reinforce these policies. ter prior to coming to a stop: runway and conditions, and comprehen- • All areas surrounding the runway sive performance data related to aircraft 5. Contaminated runways increase the should conform to ICAO Annex 14 speci- stopping characteristics, help reduce the risk of runway excursions: fications. risk of runway excursions: • Flight crews should be given accurate, • All runway ends should have a certified • Regulators should develop global, useful, and timely runway condition in- runway end safety area (RESA) as re- uniform standards for runway condition formation. quired by ICAO Annex 14 or appropriate measuring and reporting, and aircraft • A universal, easy-to-use method of substitute (e.g., an arrestor bed). performance data. ◆

January–March 2010 ISASI Forum • 7 Closing the (This article is adapted with permission analysis of crew behavior, is now extended from the authors’ paper entitled Closing to all crewmembers through flight data the Gap Between Accident Investigation animation visualization and analysis tools. Gap Between and Training presented at the ISASI 2009 These technologies can communicate seminar held in Orlando, Fla., Sept. 14-18, subtle causal factors effectively and con- 2009, which carried the theme “Accident sequently enable instructors to improve Accident Prevention Beyond Investigations.” The scenario-based training. full presentation, including cited refer- In addition to using flight data to de- ences to support the points made, can be velop enhanced scenario-based training, Investigation found on the ISASI website at www.isasi. applying FOQA concepts to the full flight org.—Editor) simulator will enable airlines to cross- reference problems encountered in simu- And Training ntimacy with accident investigations lator sessions with problems encountered and/or serious FOQA events results in in daily flight operations. Using flight data Full flight simulators are used to intimacy with many of the often subtle from the simulator session to objectively train pilots to fly airplanes and to factors and human factors issues that measure and report on the training pilot’s Ican ultimately culminate in a catastrophic performance allows the instructor pilot carry out emergency procedures outcome. While written accident reports to focus on the subjective human factors fastidiously. But are pilots, and supply a wealth of information, their aspects of the flight operation. should pilots, be trained to shortcoming is that the reports are time Understanding how seemingly benign prevent accidents? consuming to read. More importantly, events can lead to catastrophic situations people often read the same sentence yet is paramount to changing attitudes and By Michael Poole and Lou Németh, have a very different understanding of the vigilance in the cockpit. Augmenting CAE Flightscape sentence. There is arguably a gap between simulator training to replicate real-world accident investigation and simulator train- situations based on improved intimacy recorded on the aircraft’s CVR. ing in that the problems we typically see in with the sequence of events beyond the The Swiss AAIB IIC originally re- training are often not the same problems investigation report and beyond the statis- quested a “readout” of the recorders. tics of FOQA programs promises to bring Consequently, the TSB prepared printouts accident prevention to a new level. and graphs of the flight data along with The following accidents are examples a transcript of the audio, which the IIC cases in which one of the authors (Poole) intended to take back to Zurich to conduct was directly involved in the flight recorder the analysis. People intimate with the pro- analysis when employed at the TSB. cess of recovering flight data realize that Hence, he has firsthand knowledge of the the original data are a sea of binary 1s and details of the accident sequence beyond 0s that need to be converted into meaning- what is typically ascertainable from the ful engineering units. Many investigators written reports. The authors believe that believe that flight data are “factual,” but this level of intimacy can be more readily the process to convert the data is fraught gleaned through the use of flight anima- with the opportunity for error. Engineer- tions and the use of flight data to develop ing conversion formulas, documentation, Figure 1. Saab 340 accident site. full flight simulator training scenarios. wiring, acquisition unit programming, software used to convert the data, timing that cause accidents or that we see in daily Saab 340 accident issues, resolution issues, replay options, FOQA program results. This article will On Jan. 10, 2000, a Saab 340, HB-AKK etc., will all affect the quality of the out- explore a few accidents to demonstrate operated as Crossair 498, crashed shortly come. In fact, if the same source binary how improved intimacy about what hap- after takeoff from Zurich’s Runway flight data are replayed with two different pened, based on objective flight data, may 28 during night IMC. The aircraft was replay systems, it is highly unlikely that benefit flight safety. destroyed and all 10 persons on board the same results will be produced. As expert observers of cockpit behavior, died (see Figure 1). The Swiss Aircraft Flight 498’s data revealed that shortly instructor pilots have the unique skill of Accident Investigation Board (AAIB) after takeoff the aircraft, in night IMC, reliably predicting the outcome of even requested the assistance of the Transpor- entered an increasing right turn appar- small omissions or lapses in procedures. tation Safety Board of Canada (TSB) with ently consistent with control inputs. As a This instructor skill comes naturally as the readout and analysis of the FDR and flight data analyst, when you see this type a function of observing crews practicing CVR. Approximately 50 parameters were of data, you immediately start to question skills over and over again. This same in- recorded on the aircraft’s solid-state FDR, if the data are being processed properly or structor skill, acquired through persistent and 30 minutes of good quality audio were are working properly (in this case if sign

8 • ISASI Forum January–March 2010 Closing the conventions are correct) because, on the would be OK if the investigation team surface, the sequence does not appear to was brought to Ottawa to analyze the Gap Between make sense. The TSB started to work on a data interactively using the animation as flight animation immediately in an effort to opposed to trying to analyze printouts understand if the data were properly pro- and plots. Indeed, in this particular in- Accident cessed, because animations are an excellent vestigation, the early animation with the means to validate the correct behavior of audio and transcript synchronized was numerous interdependent parameters. very useful to conduct the analysis of the Investigation The level of “validation” of any given data and greatly expedited a common parameter should be proportional to what understanding of what likely happened you intend to conclude. If you are putting (see Figure 2). And Training a lot of weight on a given parameter, it is The Swiss team came to the TSB and natural that you would check its validity spent a few very fruitful days developing Full flight simulators are used to more so than if it were a less important and studying the animation. Animations train pilots to fly airplanes and to parameter. The IIC noticed the TSB was have two very distinct purposes—one is Figure 2. Flight animation developed by working on an animation and asked if it to assist in the analysis process and one TSB to support the investigation team carry out emergency procedures is to communicate the findings. Often the (courtesy the TSB). fastidiously. But are pilots, and Michael Poole, execu­ display choices are different for each of should pilots, be trained to tive director and chief these purposes. Some authorities still view prevent accidents? investigator of CAE animations as having little or no analytical Flightscape, is a pro- value and use them only for communica- By Michael Poole and Lou Németh, fessional engineer tion purposes. But in the case of this CAE Flightscape with a current pilot’s accident, the animation had tremendous license and is recog- analytical value. Without the animation, nized as an expert in the field of understanding the sequence of events flight data analysis. He represent- and gaining confidence in the data quality ed Canada as the national expert would have been much more difficult. panel member to the Inter­national Key points of this accident related to the Civil Aviation Organiza­tion’s subject of this material are as follows. Figure 3. Planned route, instructed Flight Recorder Panel. He started The pilot flying (commander) became route, and flight path to crash site. in the field of aircraft accident disoriented, essentially believing he was investigation in 1977 and worked in a left turn when he was in a right turn. the animation with the CVR transcript for more than 20 years with the During the standard instrument departure integrated it becomes relatively easy to Transportation Safety Board of (SID ZUE 1Y), ATC issued a change in the understand how the commander could Canada. For the last 15 years of clearance, essentially cutting the SID short become disoriented and roll the aircraft his career at TSB, he was the head and instructing a turn direct to VOR ZUE. into a right turn into the ground. of the flight recorder and perfor- The first officer confirmed by radio stating To further validate this early theory mance laboratory, which “turning left to Zurich East.” The SID calls in the investigation, the TSB derived the he developed for the Board. for a left turn as shown in Figure 3. theoretical behavior of the command bars The first officer reprogrammed the (since this was not a recorded parameter) Lou Németh is the LRN (long range navigation system) from and displayed them in the animation, Photo chief safety officer its present position to ZUE. At this point which further supported the supposition not for CAE’s ab-initio, in the flight, the aircraft was more or less that the commander became disoriented. available helicopter, business, 180 degrees in the opposite direction of The Swiss AAIB report makes several and commercial ZUE. When reprogramming the LRN, if excellent safety recommendations to pre- aviation training the operator does not explicitly select left vent a recurrence. While the Swiss report business. He joined or right, the LRN will choose the turn di- is very thorough and filled with excellent CAE after a distinguished 26-year rection offering the shortest distance. The safety information, it is still questionable career at US Airways, where he aircraft was a few degrees closer to a right as to whether this accident or similar ac- served as a line pilot, instructor turn at this point, and it was apparent that cidents in which the crew is essentially pilot, pilot training manager, the first officer inadvertently programmed “tricked” into a situation by a series of and courseware developer. He has a right turn by not explicitly selecting left. seemingly harmless events will effectively accumulated more than 25,000 With both crewmembers believing they be prevented in the future. hours of flight time and is type were to turn left and both crewmembers To the authors’ knowledge, there is no rated in nine commercial and believing the flight director was pro- simulator training (brief, debrief, or in the business aviation jet aircraft. grammed for a left turn, when watching actual simulator) whereby crews are ex-

January–March 2010 ISASI Forum • 9 posed to the sequence of events identified in The French BEA wrote a detailed report this accident. Even worse, there are many on this accident, but the question again flight crews who have no knowledge of this comes up—Have we done enough or what accident or the specifics of what caused it. more can be done to ensure there is not a re- The safety community has benefited from peat of this accident? As with the previous the lessons learned from this accident; but example, the authors know of no scripted are flight crews flying similar equipment simulator training in which crews are given benefiting from the lessons learned? The a false stall warning on liftoff in night IMC current approach is for the safety com- to see how they react to this real-life known munity to learn the lessons and implement situation. Given the same circumstances, it changes by way of recommendations. It is probable that many crews would react Figure 4. Undercarriage from A310 would arguably be much better if the crews the same way as the crew in question did, accident. could learn the lessons directly by expos- so it is arguably only a matter of time before ing them to the sequence in the simulator this accident repeats itself. execute an escape maneuver. How many training environment. crews would do the same thing in the same B-727-200 accident circumstances? Is this accident also a mat- Airbus A310 accident On July 7, 1999, a B-727-243F, registration ter of time before it repeats itself? Simula- On Jan. 30, 2000, an Airbus A310 regis- VT-LCI, crashed in Kathmandu into the tor training replicating this sequence for tered as 5Y-BEN crashed shortly after Champadev hills at 7,550 feet approxi- pilots frequently flying into airports with takeoff from Runway 21 in Abidjan in mately 5 minutes after takeoff in IMC. The mountainous terrain might go along toward night IMC (see Figure 4). The aircraft’s Ministry of Tourism and Civil Aviation of reinforcing the need to carry out an escape flight data recorder and cockpit voice re- the government of Nepal investigated the maneuver in all cases. corder were brought to the TSB Canada accident. No report from the government for readout and analysis. The aircraft’s of Nepal could be found in a search of the Closing the gap FDR recorded alternating streams of Internet; but the following was found on All three of these cases exhibit similar hu- steady 1s and 0s indicating the flight data the NTSB website: man factors problems: The crewmembers acquisition unit (FDAU) had malfunc- “The investigation determined that did not correctly diagnose the problem tioned and was sending erroneous data the probable cause of the accident was the and/or did not respond in a way that they to the FDR. The CVR was of good quality failure of the flight crew to adhere to a were trained due to mitigating factors and, although cryptic to determine the standard instrument departure (SID) and that led them to believe there was a better sequence of events, eventually enabled the the failure of the controllers to warn the course of action. In all cases, the crew- investigation team to piece together what flight of terrain. Contributing factors were members were competent, well trained, happened. Although there was no flight determined to be an incomplete departure and representative of the industry. It can data available for this investigation, it is briefing, unexpected airspeed decay during be argued, however, that their response perhaps a good example of a case where the initial climb, inadequate intra cockpit was understandable, which means another the crew was essentially “tricked” and flew crew coordination and communication, crew confronted with the same scenario the aircraft into the ground without ever and the slow response to the premonition may well respond the same way. understanding the problem. given by the air traffic controller.” There have been numerous similar The aircraft’s stall warning system The flight recorders were replayed at accidents in which crewmembers did not activated on liftoff, which surprised the the TSB Canada. In this accident, the respond the way they were trained. It is flight crew. As they attempted to diagnose crewmember was a little late in carrying the authors’ opinion that this is in part the problem, the flying pilot instinctively out a right turn as required by the SID and because the training environment does pushed forward on the control column to was consequently flying toward mountain- not replicate real-world scenarios such as eliminate the stall condition. However, the ous terrain. When they realized they were the three examples presented. One reason aircraft was not in a true stalled condition, late, they immediately began a right turn that the training environment does not and in less than one minute, the aircraft to regain where they were supposed to replicate real-world scenarios like this is was essentially flown into the sea. It was be. During the right turn, they received a because the people developing the training concluded that one of the angle of attack GPWS warning―Pull up. Instead of execut- simply do not know the intimate details vanes must have been damaged, causing ing the escape maneuver in response to of the accident sequence, not having been the aircraft stall system to trigger as soon the GPWS, which requires a wings level involved in the investigations. The same as the weight on wheels logic went to “air.” maximum climb, they increased their turn logic can apply to serious FOQA events. It No amount of forward control input could radius to the right. In this case, given they really does not matter if the aircraft hits avert the stall (stickshaker) condition. knew they had made a mistake and had just the ground or not in the end. FOQA events Simulator tests confirmed that the only corrected the mistake, it is understandable with a high potential for safety action need way to reach the crash site was to fly the that when confronted with a GPWS their to be investigated and well understood and aircraft in an “unstalled” condition. instinct was to tighten the turn rather than ideally used to develop simulator training

10 • ISASI Forum January–March 2010 scenarios if we really want to prevent them objective common base between the two from becoming an accident. communities. Coincidently, IATA within Flight animations have the ability to dis- its ITQI (IATA Training and Qualifications seminate complex information in a highly Initiative) is actively exploring flight data intuitive and entertaining manner in a from FOQA programs from volunteering fraction of the time it takes to read a re- airlines in an effort to change the regula- port. Like any good movie, you tend to pick tions regarding simulator training to allow out details that you did not see before each for evidence-based training. The following time you watch the movie. Written reports is an extract from the ITQI 2008 report also do not lend themselves to assessing from IATA’s website: timing issues while animations provide an “Progress in the design and reliability of immediate sense of timing, which can be modern aircraft has prompted an indus- important to the overall understanding of try review of pilot training and checking the accident. requirements. In addition to the wealth of Figure 5: Simulator session animation Finally, flight animations are an excel- accident and incident reports, flight data replay using the same core technology as lent means to communicate what hap- collection and analysis offer the possibility the accident investigation replay (cour- pened to a wide cross-section of people. to tailor training programs to meet real tesy Oxford Aviation). Without consensus as to what happened, risks. The aim is to identify and train there is little point of trying to understand the real skills required to operate, while tion, they would undoubtedly conclude why it happened. Further, what happened addressing any threats presented by the that this could happen to them as well, is exclusive in that there is only one set of evidence collected. The IATA best-practice since they begin to appreciate the subtle- facts. The why, on the other hand, is not document will facilitate regulatory change ties of the sequence. Any pilot who works exclusive. For every what, there are many and enable more efficient safety-driven and for a year at a safety investigation author- opinions as to why; there is not necessarily cost-effective training.” ity comes out of that experience with a a right answer. Despite the best efforts of One of the ways that CAE Flightscape real appreciation for what really causes the investigation community, many people is actively exploring to bring the FOQA accidents and is a safer pilot for it. simply do not know the intimate details of and training communities together is We cannot afford to send all the worlds’ the accident unless involved in the inves- through the development of our simulator pilots for a one-year sabbatical at an tigation, as it is impractical to glean this brief debrief technology and FOQA tech- investigation agency. What we can do is level of intimacy from a written report. nology. By using the same core technology give these same pilots and instructor pilots Flight animations have a unique ability to study and debrief a simulator flight as easy access to flight data from accidents to quickly communicate what happened, well as study and debrief a FOQA event and serious FOQA events ideally in the which greatly facilitates determining why from an actual aircraft, the potential for form of interactive flight animations so it happened and, more importantly, how to FOQA results to influence the training that they can appreciate the intimate de- prevent it from happening to you. environment is significantly increased. tails of what went wrong. We can include Simulator training today largely fo- simulator brief and debrief using actual cuses on how to fly the aircraft and how Summary flight data as an integral part of the train- to respond to an emergency. It has not Many people in the accident investigation ing process, not a nice-to-have option. progressed to “evidence based” training business see the same core human factors We can train instructors to leverage the in which we use objective flight data to issues over and over again. A combination technology to the benefit of the safety of develop explicit scenarios from known of individually benign events led to a situ- flight. This will facilitate the creation of accidents, incidents, and FOQA events. ation “outside the box” of current simula- evidence-based training and allow the If you ask a simulator instructor pilot for tor training. It is, of course, impossible to industry to better correlate problems a list of problems training pilots experi- train for every scenario possible, but it is identified through investigation and ence in the simulator, you will discover technically possible to train using objec- FOQA programs to problems encountered that there is little or no correlation to the tive aircraft flight data from past accidents during flight simulator sessions. The main list of problems that are known to cause and serious FOQA events. problems in the simulator are typically not accidents. This suggests that there is a gap Evidence-based training scenarios need related to reasons why airplanes crash. between the flight safety community and to be developed using objective flight data This is because we still train to regulatory the training community and that there is to ensure pilots appreciate the need for requirements and to carry out emergency benefit from a much closer relationship vigilance, communication, and a strong procedures. This is not to say we should than exists today in many airlines. safety ethic. Many pilots read the accident no longer do this. The more the real air- It is timely for the industry to look headline and conclude that this would not craft data and the simulator data match in at ways to improve the ability for the happen to them—that the pilots in ques- terms of problem areas, the more we will training community to exploit lessons tion were not doing a good job. If these know that we are closing the gap between learned by using actual flight data as the same pilots participated in the investiga- accident investigations and training. ◆

January–March 2010 ISASI Forum • 11 (This article is adapted with permission activity, rulemaking, and expense. A glance Airplane upset defined from the authors’ paper entitled How at the objective data establishes that while While specific values may vary among air- Significant Is the Inflight Loss of Control runway incursions are a concern as an issue plane types, the unintentional conditions Threat? presented at the ISASI 2009 it pales compared to loss of control. Indus- seen below generally describe an airplane seminar held in Orlando, Fla., Sept. 14-18, try statistical analysis shows 22 inflight, upset (Airplane Upset Recovery Train- 2009, which carried the theme “Accident loss-of-control accidents between 1998 and ing Aid, Revision 3). For our purposes, Prevention Beyond Investigations.” The 2007. These accidents resulted in more than airplane upset is defined as an airplane full presentation, including cited refer- 2051 fatalities. (Airplane Upset Recovery unintentionally exceeding the parameters ences to support the points made, can be Training Aid, Revision 3 [2009]). normally experienced in line operations or training. • Aircraft pitch attitude greater than 25 degrees, nose up. • Aircraft pitch attitude greater than 10 How Significant Is degrees, nose down. • Aircraft bank angle greater than 45 degrees. • Within the above parameters, but flying The Inflight Loss of at airspeeds inappropriate for conditions. Significantly, these flight conditions of- ten occur in combination. Loss of control, Control Threat? flight upset (LOC-I) is established as the potential event demanding immediate and At all times, manufacturer recommendations for decisive attention by the aviation industry to avoid further loss of life, vast financial proper aircraft operation are controlling. losses, and decline in the public’s confi- By Capt. John M. Cox and Capt. Jack H. Casey, dence. In past years, several developments Safety Operating Systems, L.L.C., Washington, D.C. in technology and improved training have resulted in significant safety improve- ments for the industry. Generally accepted found on the ISASI website at www.isasi. Data also suggest an even larger num- developments that have increased safety org.—Editor) ber of “incidents” have occurred in which have been airplanes experienced near or actual loss • the jet engine. irplane manufacturers, airlines, of control and qualified as upsets. There • improved and operator friendly avionics. pilot associations, flight training are several reasons such events occur: • improved training. organizations, and regulatory flight control problems, environmental • proactive, not reactive, safety programs. agencies are increasingly con- dangers, equipment, and pilot inattention • technological improvements, cerned with the incidence of loss or inaction. Investigation of pilot actions —weather radar, Aof control events. Accidents resulting from during these events suggest pilots require —TCAS, loss of airplane control are, and continue to specialized training to cope with airplane —TAWS (terrain awareness warning be, major contributors to fatalities in the upsets. Research indicates most airline system). commercial aviation industry. In fact, since pilots rarely experience airplane upsets The emergence of technology as the the decline of controlled flight into terrain during their flying careers. It also indi- leading contributor to safety improve- (CFIT) accidents due to technological cates that many pilots have never been breakthroughs, loss of control has become trained in maximum-performance air- the No. 1 cause of hull losses and fatalities plane maneuvering, such as aerobatic Authors’ Note in the worldwide air carrier fleet. maneuvers. Additionally, those pilots who Resources are finite in any business. have been exposed to aerobatics lose their This article reflects the groundbreak- Industry safety professionals are tasked skills as time passes unless such flying is a ing work contained in the Federal with determining the primary issues of con- consistent hobby or a second career. Aviation Administration’s Flight Up- cern and addressing them in a planned and This does not suggest training in aero- set Recovery Document, Revision 2. forthright manner with objective data and batics, although such training does im- It is consistent with the content and professional guidance. This is very difficult prove an assortment of pilot skills. Indeed, recommendations of that document, within a larger society that often decides for our purposes aerobatic training may in addition to industry best-practice issues with subjective data at best, “feel- be counterproductive, producing negative standards. The Royal Aeronautical ings” at worst. Data clearly establish loss of training outcomes, and possibly, as we will Society will soon release a document control flight upset (LOC-I) as the primary see, implanting incorrect technique. The on upset recovery intended to assist danger today in flight operations. Compare aircraft in question, transport-category pilots who encounter an unexpected this with the news media interest regarding aircraft, are not designed nor intended upset condition. ◆ runway incursions that has driven FAA for such flight.

12 • ISASI Forum January–March 2010 ments is obvious. For example, CFIT Faced with such a chal- reigned for years as the No. 1 cause of lenge, a crew, whose hull losses and loss of life. The indus- individual flight skills try has responded in a variety of ways, might have atrophied including increased training, at least due to reliance on auto- in emphasis, and regulatory agencies mation, then must deal have issued directives and regulations with an unfamiliar flight to companies and pilots regarding the situation it has not pre- seriousness of the matter. Such results pared for. This “shock” produced some mitigation of the problem, or “stun” factor must but CFIT did not cease to be the major be understood as part cause of accidents and loss of life until of the solution. the advent of TAWS, and its mandatory By necessity, flight installation and use. upset becomes a train- However, technology offers little as- ing question because of sistance (with the possible exception of the technology-resistant protection offered by fly-by-wire technol- nature of the problem. The solution should program is no longer sufficient. Anything ogy providing hard envelope protection) demand a practical approach, using al- less creates an equation of when the in- with the challenges inherent in inflight ready existing training aids, while remain- evitable will happen. Training for flight upset. Technology, especially autoflight, ing within the regulatory guidance of the upset should be as much a business model has not reached the point where it can re- Airplane Upset Recovery Training Aid, as anything else related to training and act and control flight actions at or beyond Revision 3. The need is established by a safe operations. the parameters of LOC-I. In fact, in an string of deadly accidents that illustrate No airline or operator expected the fol- era when regulators encourage crews to the problem. lowing accidents to occur with their crews utilize autoflight and other sophisticated One psychological barrier should be and aircraft, yet they did. flight aids to the maximum degree pos- examined and dispensed with. Belief that USAir Flight 427 (Boeing 737), Sep- sible, pilots are facing a situation in which “it won’t happen here,” because it has not tember 1994 the parameters of flight upset result in happened, is meaningless. Anything less American Airlines Flight 587 (A300), the disconnection of those same systems. than a professional and active training November 2001 Pinnacle Airlines Flight 3701 (CRJ200), Capt. John Cox, retiring holds an air safety certificate from the October 2004 from the airlines after University of Southern California. West Caribbean Airways Flight 708 flying 25 years, founded (MD-82), August 2005 Safety Operating Jack H. Casey spent the These tragedies were selected for brief Systems, a Washington, 8 years prior to becom- examination because they illustrate the D.C.-based aviation ing a partner at Safety problem in clear and unambiguous terms. safety consulting firm. Operating Systems as There are others. In fact recent tragedies, He has more than 14,000 flying hours the pilot liaison for while investigations continue, show signs with more than 10,000 in command of Embraer Aircraft Hold- within regulatory public statements of jet airliners. He holds an airline trans- ings, Customer Services possible crew control mismanagement and port pilot certificate with type ratings in USA, achieving a reputation as a lack of awareness of actual flight condi- the Airbus A320 family, the Boeing 737 worldwide expert in Embraer aircraft tions without autoflight. family, the Fokker F28, and the Cessna and their operations in line service. Citation. Cox is a Fellow of the Royal Since leaving military service in 1975, USAir 427 Aeronautical Society. He served as an he has been employed with airlines At approximately 19:03 Eastern Daylight air safety representative for the Air and has flown the BE18, DC-3, B-737, Time on Sept. 8, 1994, USAir Flight 427 Line Pilots Association for more than and FK27. In addition, Casey has (ORD-PIT) descended out of control 20 years, rising to the position of Execu- developed expertise in the Mitsubishi and crashed killing all aboard outside of tive Air Safety Chairman, ALPA’s top MU-2 family of aircraft. He possesses Pittsburgh, Pa. safety job. ALPA awarded him its high- five type ratings from DC-3 to B-737, In the executive summary of the final est safety award in 1997. The Guild of and has flown more than 11,000 hours report of this accident, the NTSB states Air Pilots and Air Navigator presented and 2.8 million miles without accident. the following: “The National Transpor- him with the Sir James Martin Award During his career, Casey has served in tation Safety Board determines that the for aviation safety in 2007. He is an all air carrier required positions, been probable cause of the USAir Flight 427 experienced accident investigator, hav- a designated examiner, type rating accident was a loss of control of the air- ing been involved in 13 major investiga- examiner, and line check airman and plane resulting from the movement of the tions (the best known being the US Air gained considerable experience with rudder surface to its blowdown limit. The 427 accident in Pittsburgh in 1994) and the FAA, the NTSB, and numerous rudder surface most likely deflected in numerous smaller investigations. He foreign regulators. a direction opposite to that commanded

January–March 2010 ISASI Forum • 13 American Airlines designed to allow him to do it. 587 For our purposes, the Board concentra- Barely one month tion on the Advanced Aircraft Maneuver- following the attacks ing Program is significant and troubling. of Sept. 11, 2001, It is clear that American Airlines made a New York City again strong commitment to address the clear faced tragedy from dangers present in loss of control flight the sky. This time it upset events. The airline created an ag- was not the madness gressive program designed to improve a of terrorism, but hu- pilot’s awareness and skills. The design man error. of the program was an “in house” training At 09:16 East- department effort that, at least initially, ern Standard Time, sought input from manufacturers and American Airlines regulators. However, as time passed, Flight 587, an Airbus disagreements on basic aerodynamic A300-605 N14053, theory and technique began to surface. crashed into a resi- While the program had the very best by the pilots as a result of a jam of the dential area of Belle Harbor, N.Y. Flight intentions, it came under question within main rudder power control unit servo 587 was a regularly scheduled flight from the company’s operations management. valve secondary slide to the servo valve JFK to Las Americas International Air- The NTSB Final Report discussed one housing offset from its neutral position port, Danto Domingo, Dominican Repub- significant issue. and overtravel of the primary slide. The lic. Fatalities totaled 260 passengers and “On Feb. 6, 2003, American Airlines safety issues in this report focused on crew aboard the aircraft, and 5 people on provided the Safety Board with a copy of Boeing 737 rudder malfunctions, includ- the ground. a May 27, 1997, memorandum from the ing rudder reversals, the adequacy of Once fears of terrorism were eliminated, company’s managing director of flight the 737 rudder system design; unusual it was evident the aircraft impacted the operations technical to the company’s attitude training for air carrier pilots; ground in an ominous fashion. The location chief pilot and vice-president of flight. and flight data recorder (FDR) param- of the vertical stabilizer and rudder in Ja- The memorandum stated that the manag- eters.” (p. ix). maica Bay was proof positive of a structural ing director of flight operations technical After the building of thousands of B-737 breakup while in flight. The first officer was had ‘grave concerns about some flawed series aircraft, a significant danger to the pilot flying the aircraft. aerodynamic theory and flying techniques flight, existing since the original design, During the investigation it became that have been presented in the AAMP.’ was discovered under tragic circum- apparent that the aircraft encountered The memorandum also stated that it was stances. The B-737 had been in service in wake turbulence a few minutes after wrong and ‘exceptionally dangerous’ to various versions for years. The fault could takeoff from a B-747 that departed the teach pilots to use the rudder as the pri- not have been planned for in a standard same runway immediately before. This mary means of roll control in recoveries business plan or normal testing. However, encounter started the trouble. Wake tur- from high AOAs.” pilot training mitigation for such nasty bulence around busy traffic areas mixing The memorandum continued to request surprises can be a mitigating factor. aircraft of various sizes and capability is a review of a number of concerns regarding The report includes commentary on the hardly unknown; in fact, it is a frequent the program, some raised by manufacturer inadequacy of air carrier pilot training occurrence. test pilots. In addition to the propensity to address this type of unexpected flight The executive summary of the NTSB of the first officer to use excessive rudder, upset. This resulted in remodeled training final report on the loss of Flight 587 says such instruction created a toxic combina- for crews flying the B-737, and increased the following: “The National Transporta- tion that, under demands of the event, knowledge of more sophisticated aero- tion Safety Board determines that the stressed the Airbus vertical stabilizer and nautical issues like cross-over speeds. probable cause of this accident was the rudder beyond design limits. Although this crew had little chance inflight separation of the vertical stabilizer This chain of events illustrates key is- given its previous experience and train- as a result of the loads beyond ultimate sues regarding training for inflight upsets. ing, future B-737 crews should fare better. design that were created by the first of- This event brought the entire concept Prominent in the NTSB recommendations ficer’s unnecessary and excessive rudder into question in some minds. Carriers were suggestions for more sophisticated pedal inputs. Contributing to these rudder developing such programs ceased their training of flight crews so that following pedal inputs were the characteristics of development. The fact that American such training, an event like USAir 427 the Airbus A300-600 rudder system design Airlines increased exposure and liability could be prevented, but if encountered, and elements of the American Airlines through such a program was not lost on be recoverable. Advanced Aircraft Maneuvering Pro- the industry. It also provided additional Of interest to operators is that this in- gram.” (p. xi) (authors’ italics) rationalization for those opposed to such vestigation required more than 4 years to In effect the report states that the fly- training for various reasons, such as cost, complete and cost the industry more than ing pilot excessively loaded the rudder the effort involved, or simple resistance $1.5 billion in direct and indirect losses. beyond design limits and the system was to change. By any estimation, Flight 587

14 • ISASI Forum January–March 2010 ing program, containing did not cease until ground impact. an inflight upset section, The events discussed here demonstrate is a useful tool for de- the challenge ahead for the industry. Is- tecting and, if need be, sues include the proper use of technology, removing pilots from preserving, and enhancing non-automated the system who cannot pilot flying skills, corporate commitment, or will not improve their regulatory understanding and oversight, performance. and significantly “buy in” by the pilot The key issue es- groups. tablished by Pinnacle The post American 587 syndrome is fi- 3701 is that regard- nally waning under the pressure of events less of the behavior and acceptance of the problem. A growing and the predicament number of operators are developing and that resulted, the crew implementing pilot training programs, could have probably including academic and simulator training. recovered sufficiently Regulatory agencies are again encourag- stands as a classic inflight upset event with to save their lives and the aircraft with ing airlines to provide education and train- unintended consequences. knowledge contained in inflight upset ing in the subject. Airplane manufacturers training programs. responded to the challenge by leading Pinnacle Airlines 3701 an industry team formed to develop the On Oct. 14, 2004, at approximately 2215:06 West Caribbean Airways 708 Airplane Upset Recovery Training Aid, Central Daylight Time, Pinnacle Airlines On Aug. 16, 2005, West Caribbean Air- Revision 3 with the FAA and other in- Flight 3701, a repositioning flight from ways Flight 708, an MD-82 (HK-4374X) dustry experts. This aid provides basic, Little Rock, Ark., to Minneapolis-St. charter flight from Panama to Martin- but useful, guidance and templates for a Paul International Airport in Minnesota, ique, descended from cruise altitude in a training program as well as sample train- crashed near the Jefferson City, Mo., airport, killing the crew of two, the only souls aboard the aircraft. The NTSB final report was scathing: ccident data are clear—the greatest risk “The National Transportation Safety Board determines that the probable Ato the fleet of transport aircraft is loss causes of this accident were (1) the pilots, unprofessional behavior, deviation from of control in flight. Through proper training standard operating procedures, and poor airmanship, which resulted in an inflight and education, this risk can be reduced. emergency from which they were unable to recover, in part because of the pilots’ inadequate training; (2) the pilots’ failure nose-up flight attitude and crashed near ing manual revisions and lessons to begin to prepare for an emergency landing in a Machiques, Venezuela, killing all 160 per- the process on the correct footing. timely manner, including communicat- sons aboard. As we have seen, airplane upsets hap- ing with air traffic controllers immedi- Investigation by the CIAA of Venezuela pen for a variety of reasons. Some events ately after the emergency about the loss showed the following: are more easily prevented than others. of both engines and the availability of • Ground scarring indicated impact in a Improvement in airplane design and landing sites; and (3) the pilots’ improper nose-up, slight right roll attitude. aerodynamic simplicity and equipment management of the double engine failure • Wreckage was distributed over a trian- reliability continues to be a goal. Auto- checklist, which allowed the engine cores gle-shaped area approximately 205 meters mation may have a result counter to its to stop rotating and resulted in the core long by 110 meters at the widest point. intention: we have arrived at the point that lock engine condition. Contributing to • Both engines exhibited indications of when airplane upsets occur pilots discover this accident were (1) the core lock engine high-speed compressor rotation at the degradation in basic flying skills. condition, which prevented at least one time of ground impact. In too many recent accidents, pilot engine from being restarted, and (2) the • The engine inlets, empennage, and inability to recover from an unintended airplane flight manuals, which did not wing leading edges showed no sign of pre- inflight condition (upset or stall) has communicate to pilots the importance of impact damage. resulted in the loss of the aircraft and oc- maintaining a minimum airspeed to keep • The horizontal stabilizer was found at cupants. The number of this accident type the engine cores rotating.” about the full airplane nose-up position can, and should, be reduced. Accident data Professionalism and aircraft knowl- (12 units). are clear—the greatest risk to the fleet edge, as well as basic aerodynamics, can be Additionally, the FDR showed that of transport aircraft is loss of control in trained. Additionally a well-designed and the aircraft had slowed while at cruise flight. Through proper training and educa- appropriately taught and monitored train- altitude before beginning a descent that tion, this risk can be reduced. ◆

January–March 2010 ISASI Forum • 15 (This article is adapted, with permission, rate by 80% in 10 years. It was recognized developed in the U.S. by the IHST. from the author’s paper entitled Bringing by the government-industry partnership At the writing of this paper, the Eu- the Worldwide Helicopter Accident Rate that constituted the IHST that its goal ropean Aviation Safety Agency (EASA) Down by 80% presented at the ISASI 2008 could not be achieved by looking at gen- and Canada are continuing the initiative seminar held in Nova Scotia, Canada, eral aviation and helicopter accidents and by examining more closely their re- Sept. 8-11, 2008, which carried the theme investigative reports in the same way. gional accidents while refining the IHST “Investigation: The Art and the Science.” The taskforce formed to address and methodology to optimize the ability to The full presentation, including cited ref- try to correct the problems contributing identify corrective actions there. Several erences to support the points made, can be to the helicopter accident rate was led other countries and regions have begun found on the ISASI website at www.isasi. initially by the U.S. FAA, airframe and to organize similar teams to determine org.—Editor) engine original equipment manufacturers how the process can reduce accident risk (OEMs) in the United States, and by op- elsewhere. The methodology of the IHST his presentation was motivated by a erators and associations, such as the He- was adapted from the U.S. Commercial personal desire to spread the word licopter Association International (HAI), Aviation Safety Team (CAST), which set Tabout an exciting new approach to which represented U.S. operators. As the out to substantially reduce worldwide accident prevention that has particular initiative has grown internationally, other commercial airline fatal accidents by ad- applicability to air safety investigators, countries have stepped up or expressed dressing proactively common accident safety researchers, and aviation safety interest in using or adapting the model causes. Many of us have seen some of the agencies. Traditionally, if we were par- ticipants in a government-led aircraft accident investigation, the objectives The International Helicopter Safety Team (IHST) would be to determine the factors that takes the proactive attitude that anyone’s led to the accident, to define the “prob- helicopter accident belongs to all of us. able cause” (or causes), and to initiate corrective actions to prevent future ac- cidents. The safety recommendations Bringing the that followed would usually advocate regulatory change to require sometimes narrowly defined corrective actions. This Worldwide Helicopter methodology is frequently addressed as the reactive approach or “preventing the last accident.” Accident Rate Down That’s the way fatal air carrier acci- dents have been addressed in many coun- tries during the past 40 years, and despite by lots of criticism, the approach has helped to bring the air carrier accident rate down substantially. Unfortunately, general avia- % tion accidents, especially those that did not involve mechanical failures of critical components, although much more numer- ous, usually do not result in the issuance of safety recommendations—either because the accidents were not investigated to suf- ficient depth to support recommendations or because regulatory change could not be justified. Studies of similar accidents are infrequently conducted, and we continue to experience accidents just like we did before. A more proactive approach and a de- sire to improve the worldwide helicopter accident rate, which was perceived to be unacceptably high and negatively influ- encing the safety image of all helicopter operations, led to the formation of the International Helicopter Safety Team (IHST) in late 2005 and its commitment to reduce the worldwide helicopter accident By Jack Drake, Helicopter Association International, USA 16 • ISASI Forum January–March 2010 80 by-products that addressed controlled to initiate an effort to flight into the ground, approach and land- improve the flight hour ing accidents, and uncontained engine measurement that was failures. The IHST set a similar but higher critical to accurate acci- goal—to reduce the rate of all helicopter dent rate calculations. accidents, worldwide, by 80%—and prog- Bell Helicopter has ress is already being made. taken the lead on this and is collecting flight Measuring our progress hour data for all helicop- While it must be admitted that the 80% ter models worldwide. goal was borrowed from CAST, the expec- The process will use tation from the beginning was that the goal data points from aircraft could be achieved—because of the early sales, public records of success of CAST in bringing down the aircraft registrations, fatal air carrier accident rate and a belief service difficulty re- that the helicopter community could be porting, maintenance convinced that a systematic approach to data, and other sources. helicopter safety was both overdue and The data will allow us needed. The IHST hoped for and saw an to calculate accident early reduction in the U.S. helicopter ac- rates more accurately cident count and rate (in 2006) that was in the future and (we apparently the result of enthusiasm for believe) will allow us to the program and improved industrywide show that flight hours safety awareness. are higher than previ- While that improvement was gratifying, ously estimated and it also became apparent that we didn’t that accident rates are actually know the U.S. and worldwide lower than previously helicopter accident rates because we didn’t indicated by industry have a good handle on the denominator in data. the equation, helicopter operating hours. Measuring future helicopter accident rates The IHST process (and our progress) would require more ac- The process whereby curate measures of helicopter flight hours, the IHST is analyzing accident data to The analysis function and production which had traditionally been estimated achieve a higher safety goal is illustrated of recommended interventions is the based on limited operator surveys and by the following charts and described function of the IHST’s Joint Helicopter were notoriously inaccurate. FAA flight briefly here. First, IHST is a government- Safety Analysis Team (JHSAT). The hour estimates were used by the helicopter industry partnership that seeks to bring vetting, prioritizing, and selling of the % industry to calculate accident rates. The about safety change without increasing best of the recommendations is the work IHST decided it would start its program regulation. It is not a U.S. program (al- of the IHST’s Joint Helicopter Safety using industry-accepted accident rate data though it was dominated by U.S. partici- Implementation Team (JHSIT). Figure from the years immediately preceding the pants in its first 2 years). It is an interna- 1 illustrates the IHST organization and start of its programs, but it also committed tional safety program that hopes to grow the initial makeup of the U.S. JHSAT and to a worldwide effort to reduce the risk of JHSIT teams. Not surprisingly, the roles Jack Drake is currently helicopter accidents worldwide. of the U.S. teams have evolved, partly an independent aviation Further, there is no reason its principles because of the expansion of the process safety consultant and co- couldn’t be applied proactively to all seg- internationally and partly because of the chair of the Internation-­ ments of aviation and to other industries. evolution of the tasking. One notewor- al Helicopter Safety Very simply, it uses the combined talents thy change is that the U.S. JHSAT has Team’s Accident Analy­ represented by regulators, manufactur- gradually assumed a regional role in a sis Group. He is a former ers, operators, and other safety specialists coordinated effort with other state or director of safety for the Helicopter As- to examine accident reports to identify the regional JHSATs while it has also refined sociation International and director of events that contributed to accident causes its accident analysis process. operations safety for America West Air- (root cause analysis) and to find interven- Figure 2 illustrates the analytical pro- lines. Before that, he was an investi-­gator tions to address each of those factors. cess used by the JHSAT. The charter of and manager at the National Transpor­ ­ Recommendations that would come from the organization established the ground tation Safety Board for 26 years; and he the analysis process would be based on rules, and the CAST-based process en- was a pilot, flight instructor, and safety those interventions that are considered sured that the result would be data-based, officer in the U.S. Navy. Drake has been most feasible and economically acceptable objective, and acceptable to the majority of a member of ISASI since 1977. to a cost-conscious industry. the team. The team was selected from ap-

January–March 2010 ISASI Forum • 17 public by free download from the IHST Figure 1. U.S. JHSAT and JHSIT Stakeholders website (http://www.ihst.org/Default. aspx?tabid=1797&language=en-US). It is interesting to note that while the report discusses the accidents from the prospec- tive of 15 different mission categories, the majority of the safety issues and interven- tion recommendations were substantially the same across mission categories. Most of the problems defined were operational, and most of the interventions addressed better risk management, operational oversight by operators, and training. The presenta- tion of the JHSAT report to the IHST, in September 2007, officially transferred the result of the first year of JHSAT effort to the JHSIT, which was tasked with deciding which recommendations to prioritize and how they should be implemented. That work is on-going, even as the JHSAT is now concluding its second year of acci- dent data analysis. (A later IHST report summarizing the U.S. JHSAT analysis of CY2001 accidents from the respective of “occurrence” types is now available on the IHST website, and the U.S. JHSAT, having plicants with accident investigation, safety tions for the SPSs. When data were in- now analyzed more than 500 helicopter ac- research, and proactive aviation safety sufficient to define exactly why an event cidents, is currently preparing its analysis experience, who also represented a cross- occurred, we still attempted, based on of the CY2006 accidents.) section of the industry. There was a delib- our collective experience, to determine erate effort to achieve balance between how the problem or accident might have We know more than we know! members who would represent regulators, been prevented. We did not rely on “prob- An in-depth review of a large number of manufacturers, and operators and to bring able cause” determinations and typically accident reports and their probable causes to bear real-world experience. arrived at more problem statements and reveals many things, but not all we’d like It was necessary to select a dataset that interventions than would be found in the to know—particularly if our task is to was several years old because it takes conclusions of the NTSB reports. determine responsibility for the accident a few years before entire calendar year The identified SPSs and interventions (our tasking was to look for prevention accident sets are completed by investi- were “scored” based on how well the group opportunities, not for causes). We’d like to gation authorities and recent accidents might still be in litigation (precluding some members from participating in the he JHSAT examination, even of investigation reports JHSAT group analyses). Thus, the U.S. Tthat were lacking, found lots of fertile ground for JHSAT decided to start (in early 2006) interrupting the causal chain with proactive action to with the U.S. NTSB calendar year 2000 reduce the accident frequency and rate. accident dataset, which consisted of 197 accident reports and about 4,000 docket felt the SPS or linked intervention defined know more about the pilot’s training and items (reports, statements, photos, and the problem or would eliminate it in a real- decision-making, whether undocumented supporting documents). world setting. In the roll up of the data, human factors were involved, the exact Having established what we would ex- there were about 1,200 SPS-intervention sequence of events, what the pilot saw and amine, we developed a methodology that pairs. Those interventions that became the what actions were taken, how the aircraft included development of standardized recommendations of the JHSAT (number- systems responded, and how company problem statements (SPS) corresponding ing about 135) were those that appeared standards (or lack thereof) and pressures to the events or links in the safety chain most frequently in the roll up of all of the affected crew performance on the accident that were considered to have contributed data, as the frequency of occurrences was flight. Without detailed documentation to the causes of the accidents or our ability found to outweigh the qualitative analysis of these things and digital data from the to define those causes (missing data). Hav- of individual interventions. aircraft, many of those questions remain ing defined the problems, we developed a The U.S. JHSAT report summarizing unresolved in too many cases. set of corresponding interventions that its first year of accident data analysis `However, the accident reports, espe- were thought to be appropriate mitiga- and recommendations is available to the cially when examined in the context of

18 • ISASI Forum January–March 2010 • FAA installation of ADS-B in GOM Figure 2. JHSAT Process Overview to facilitate IFR operations in adverse weather and at night. • Establish company SOP that disallows flying in adverse weather at night except under IFR. • Company SOP to eliminate onerous flight schedules and reduce risk of fatigued pilots. • Incorporate non-punitive fatigue call-in protocol. • Company risk assessment/management program. • Cockpit recording device with under- water locator.

Case Study #2: EMS Positioning Flight— ”Fire-Radio Dispatch” directed the launch of an EMS helicopter to a landing zone (LZ) on a residential city street at night, although the LZ was only a few minutes reports of similar events, do reveal a great there was no FAA communications or radar away and an ambulance was already on deal about how such accidents might have flight following available near the presumed scene. Patient transport was intended been prevented. We can read between crash site. Analysis showed that the aircraft to end at a hospital in another city. The the lines to some extent, especially when was about 25 miles from its destination and launch decision by a non-aviation dis- we’ve seen the same set of circumstances well off course when the pilot last spoke to patcher was not questioned by the pilot/ described in a variety of reports. The a company dispatcher. The aircraft never operator. Operator guidance (to ground JHSAT examination, even of investigation arrived, and the wreckage was not found. emergency response crews) had previous- reports that were lacking, found lots of As with all the other cases examined, the ly established minimum LZ requirements, fertile ground for interrupting the causal aircraft was not, nor was it required to be, but a non-conforming LZ was selected and chain with proactive action to reduce the equipped with cockpit or digital flight data was accepted by the pilot. The pilot’s high/ accident frequency and rate. recorders. Nonetheless, the NTSB report low recon did not detect obstructing wires, The following case studies are offered was rich with information that provided and ground personnel incorrectly advised to illustrate the process and the potential an understanding of the circumstances the pilot that there were no wires. Ground value of the JHSAT process and data. that may have led to the accident, and the personnel did not adequately protect the JHSAT elected to consider this information LZ from approaching automobile traffic, Case Study #1: Offshore, Part 135 Cargo— as it examined how the accident might have which necessitated a go-around on final ap- The helicopter went missing and presum- been prevented. Probable cause: Undeter- proach. The approach was not stabilized— ably crashed on a night overwater flight mined. NTSB recommendations: None. the pilot carried insufficient power to in support of oil industry operations over clear obstructing wires—necessitating a the Gulf of Mexico (GOM) in deteriorating Problem statements controlled crash under the wires. The pilot weather. The accident flight was one of • Pilot experience lead to inadequate had only 28 hours in make/model. NTSB several that day for the 61-year-old pilot, planning with regard to weather. probable cause: Pilot failed to detect the who transported personnel and cargo be- • IFR system incompatible with mis- presence of the wires and his misjudgment tween shore bases and oil rigs, and from sion. of clearance from the ground during the rig to rig. The pilot’s duty day began at • Management policies/oversight inad- evasive maneuver. NTSB recommenda- 0800 and was drawing to a close 13.5 hours equate. tions: None. later, after about 7 hours of flight time, as • Management disregard of human per- he tried unsuccessfully to find the oil rig formance factors (i.e., duty/flight time, Problem statements that was his final destination. fatigue). • Improper launch decision by non-avia- Meteorological data indicated that the • Pilot disregarded cues that should have tion dispatch/communications center. flight most likely encountered adverse led to termination of course of action. • Customer pressure to complete mis- weather (high winds, low ceilings, rain, and • Darkness, fog, and rain. sion. reduced visibility) in darkness. Although he • Data/information not available to in- • Flight crew decision-making inad- was highly experienced in VFR operations, vestigators. equate. the pilot was not IFR-rated and the aircraft • Management policies/oversight inad- was not equipped for instrument flight. The Proposed interventions equate. pilot relied on a GPS receiver, a telephone, • Company risk assessment/management • Improper landing site selection by and DR navigation to find his destination, as program. ground emergency personnel.

January–March 2010 ISASI Forum • 19 • Pilot unaware of obstructing wires near agement program. ing and aeronautical decision-making the LZ. • Develop hiring/screening criteria for (ADM) with regard to autorotation, loss • Evasive maneuver required—pilot pilot applicants. of tail rotor effectiveness (LTE), aircraft inexperienced in make/model. • Conduct procedural intentional non- performance capabilities and limitations, compliance (PINC) training. emergency procedures, inadvertent flight Proposed interventions • Implement a non-punitive safety event into instrument meteorological conditions • Mission-specific (EMS) risk assessment reporting system. (IIMC), make/model transitions, model- training for dispatch/communications • Use crash-resistant fuel systems (when specific power and energy management, center personnel. available). quick-stop maneuvers, landing practice • Assertiveness/AMRM training for • Install cockpit data recording devices. on platforms and in unimproved areas, flight crew. • Include helicopter data in PRIA (Pilot and pinnacle approaches. • Mission-specific (EMS) operational risk Records Improvement Act). • Provide extensive initial and recurrent assessment program. emergency training as described in the Ro- • Establish/assert operational control/ U.S. JHSAT fleetwide torcraft Flying Handbook and as outlined in oversight by operator. recommendations the OEM rotorcraft flight manual to address • Reinforce the purpose and importance The proposed interventions in the case autorotation, vortex ring state (settling of landing site recon. studies above illustrate the kinds of safe- with power), dynamic rollover, systems and • Establish preapproved landing sites for ty recommendations being considered by equipment malfunctions, and LTE. EMS activities. the IHST. Together they demonstrate • Infrastructure: Ensure that crews are • Risk assessment/training for LZ per- there are many ways of preventing the aware of adverse or deteriorating weather sonnel. • Risk assessment program that ad- dresses night LZ operations. hose of us on the JHSAT hope you’ll find our process • Establish EMS mission-specific op- Tuseful and employ it elsewhere to improve accident erational risk assessment standards/ investigations and make better use of the information controls. contained in those reports. Case Study #3: Returning from Electron­ same or similar accidents. The roll up conditions by expanding availability of ic News Gathering (ENG) Flight—Two of the similar interventions sometimes weather data needed for preflight planning ENG aircraft were flying several hundred resulted in combining, rewording, or and for inflight decision-making. Improve feet from one another when one pilot radi- dropping some potential recommenda- the Automated Weather Observing System oed the other, “Hey, watch this.” The pilot tions when others seemed more viable. (AWOS) infrastructure and other weather subsequently lost control and crashed; These case studies and the roll up of the reporting sources to provide greater access there was a post-crash fire, and both oc- intervention recommendations show that to weather information. Share weather cupants were killed. The investigation re- there is greater potential for prevention information, both reporting and receiv- vealed that the accident pilot had a history based on the analyses of general avia- ing, through PIREP, the helicopter EMS of aggressive flying and risk taking, and tion or helicopter accident reports than (HEMS) weather tool, and other systems. he had a criminal record. Probable cause: we have tapped in the past. While the • Companies operating in the same lo- The pilot’s decision to perform an abrupt resultant recommendations addressed cal areas should formalize agreements low-altitude aerobatic maneuver. NTSB mission groups separately, ISASI mem- to share weather data, especially when recommendations: None. bers may be more interested in the kind weather considerations result in refusing of data-based recommendations that to accept or canceling flight operations. Problem statements were proposed across mission groups • Improve maintenance by better inte- • Improper pilot decision-making and (the fleetwide recommendations). Some gration of quality assurance systems and actions. of these are summarized below. ensured adherence to instructions for • Pilot disregarded rules and SOPs. • Develop and use formalized Safety continued airworthiness (ICA). Push strict • Inadequate company SOP and opera- Management Systems (SMSs) to reduce adherence to ICA, including improved tional oversight. risk and to improve individual and or- regulatory oversight of maintenance. • Inadequate crew hiring/screening ganizational decision-making. Establish • Regulatory: Hold public-use military criteria. and use non-punitive safety event report- surplus helicopter maintenance to civil • Absence of threat-free safety event ing systems to address employee safety ICA (or equivalent) standards. Require reporting system. concerns. that public-use operators comply with Part • Data unavailable to analyze the LOC • Identify and manage risks associated 91 operating rules. GSA: Take stronger maneuver. with mission, low/slow and other higher action to minimize unapproved part use in • Crash-resistant fuel system had been risk maneuvers, and flight in close proxim- public-use operations. Defense Reutiliza- removed. ity to obstacles. tion and Marketing Services (DRMS)— • Promote increased use of simulators, develop an easily accessed database to Proposed interventions training aids, and training devices to identify released military surplus aircraft, • Establish an operator safety/risk man- reduce training risk and improve train- engines, and critical parts.

20 • ISASI Forum January–March 2010 • Regulatory: Make Pilot Records Im- was the first FAA-approved helicopter Safety is an attitude provement Act of 1996 (PRIA) information FOQA program. Without digital or video I’d like to refer to some anonymous ram- more readily available to employers for recording equipment installed, it is virtu- blings (author unknown) that attempt to background checks by helicopter opera- ally impossible to accurately reconstruct define what we are trying to do. tors. Expand PRIA to include helicopter the events leading to many helicopter acci- • Safety is not an activity to be engaged pilots and FAA disciplinary actions and dents. With such recorders, it is possible to in only when one is being watched or make this data available for aviation em- improve accident investigations and also to supervised. ployer background checks. use digital data to reconstruct training or • Safety is not posters, slogans or rules; • Implementation of health and usage operational flights as a means of improving nor is it movies, meetings, investigations, monitoring systems (HUMS) or engine training and flight crew performance. or inspections. monitoring systems (EMS) capability • Safety is an attitude, a frame of mind. for early detection of impending failure. Accident investigating • It is the awareness of one’s environment Regulatory: Provide regulatory flexibil- and reporting and actions all day, every day. ity for installation of (HUMS/EMS) data The IHST sought improvement in accident • Safety is knowing what can injure or recording systems. investigation reporting so that reports by damage, knowing how to prevent the in- • Improve crash survival by making the investigating authority would be more jury or damage, and acting to prevent it. greater use of available crash-resistant fuel useful for identifying root causes and The International Helicopter Safety systems and personal locator devices. implementing appropriate and responsive Team (IHST) takes the proactive attitude • Make greater use of helicopter terrain safety actions. To facilitate this, in June that anyone’s helicopter accident belongs avoidance warning systems (HTAWS), 2007 members of the JHSAT and JHSIT to all of us. Accidents affect our collective obstacle proximity detection and pro- met with two NTSB members, the deputy reputation as the providers of air trans- tection equipment, radar altimeters, director for regional operations, regional portation and the suppliers of air services synthetic/enhanced vision systems (SVS/ directors, and helicopter experts to discuss that don’t exist elsewhere. We don’t need EVS), video recording systems (including IHST findings and to offer suggestions, in- to accept accidents or a high industry ac- rearward-facing cameras), and wire strike cluding a checklist seeking documentation cident rate, and it affects our profitability protection systems (WSPS) as applicable of the planning that preceded the accident if we do so. to aircraft mission. flight, weather data available to the pilot, a The U.S. JHSAT attitude is that interven- • Encourage development and use of op- description of any inflight emergency and tions can be identified and mitigated for all tional aircraft warning systems to include how the pilot responded to it, a description accidents, even when the exact causes are low rotor speed, low fuel quantity, and of the size and complexity of company not known to the operator or the investiga- dynamic rollover alert systems. operations, the operator’s program for tors. We don’t have to sit back any longer • Utilize HOMP to monitor and provide managing risk and safety, the pertinent and wait for a probable cause determination safety oversight of flight operations. operator SOPs and operational oversight, before we initiate risk reduction measures. • Install cockpit/data recording de- the pilot’s pertinent (mission) training and We can use the data-based solutions derived vices appropriate to mission and aircraft experience, company hiring criteria, the by a team with broad helicopter safety model—such as cockpit image/information availability and usage of safety/mission expertise to reduce risk and the helicopter recorders (CIR), low-cost flight opera- equipment (including recording devices), accident rate, and we can use that process tions data recorders, GPS positional flight and other information that would aid the to learn from other accidents. recorders, CVR/DFDR, and FOQA quick investigator in determining root causes Other groups of helicopter safety experts access recorders (QAR). of the accident. The NTSB subsequently representing other countries and regions responded that it would use the IHST are using similar processes to examine acci- Note: Although cockpit and flight data suggestions to improve its accident in- dent data from other parts of the world. All recorders were not installed on any of vestigations. of us are working hard to deliver to you un- the 197 helicopters in the calendar year The IHST participants were very en- biased and data-based solutions to prevent 2000 accident dataset, it is not farfetched couraged by the NTSB response in 2007 the problems that show up in the accident to be asking for their installation in many and met with the NTSB Board and senior reports. But what is your attitude? Are helicopter make/models. Sikorsky (as staff in 2009 to provide further data to you ready to use that data to bring about a part of its Safety Enhancement Program) justify NTSB action. The IHST is very more proactive safety culture, better opera- is currently installing CVR, FDR, and encouraged by the interest shown by tional oversight, better mission-specific op- current generation TAWS as standard NTSB and have high hopes that the result erational training, better risk-based launch equipment in its commercial helicopter will include better accident investigation and inflight decision-making, and the models and offering retrofit kits for its documentation in the future, allowing installation of equipment that will reduce S-76 helicopters that were produced more in-depth root cause analysis and pilot workload, reduce accident risk, and before 2005. Bristow/Air Logistics has better safety recommendations in the fu- better define why accidents occur? Those installed low-cost flight operations data ture. If there is similar progress in getting of us on the JHSAT hope you’ll find our recorders in its Bell 206 and 407 aircraft more helicopters equipped with recording process useful and employ it elsewhere to and has recently received FAA approval devices in the future, a quantum leap in improve accident investigations and make for its flight operations quality assurance credible accident investigation findings better use of the information contained in (FOQA) program, and ERA Helicopters can be expected to follow those reports. ◆

January–March 2010 ISASI Forum • 21 Sifting Lessons from the

(This article is adapted with permission from the authors’ paper entitled Sifting Lessons from the Ashes: Avoiding Lost Learning Opportunities presented at the ISASI 2009 seminar held in Orlando, Fla., Sept. 14-18, 2009, which carried the theme “Accident Prevention Beyond Investiga- Current processes for identifying, defining, tions.” The full presentation, including she cited references to support the points A s communicating, and acting on lessons to be made, can be found on the ISASI website at www.isasi.org.—Editor) learned are inadequate to take advantage of the opportunities offered by investigated accidents. ecent high-visibility accidents demonstrate that processes for By Ludwig Benner, Jr., Principal, Starline Software Ltd., learning costly lessons that should and Ira J. Rimson, Forensic Engineer have been identified by investiga- Rtions continue to underperform expecta- tions fulfilled the expectations of ISASI’s Paris. Facts are not yet adequate to support tions. The accident scenarios of the crash founders. What happened? any of the many hypotheses, at least two of of a Continental-Colgan de Havilland What happened has been the recurrence which have happened before— Dash 8-Q400 in Buffalo, N.Y.; and the of accidents that bear striking similarities • air data inertial reference unit (ADIRU) crash of a FedEx MD-11 in Narita, Japan, to those that have happened before. We call faults resulting from errant input signals, a month later, reflect missed opportunities these recurrences “retrocursors.” Unlike with resulting reversion of control laws to learn the lessons from similar previous “precursors,” which presage events to come from (normal) computer control to one accidents, or analyses by those who might in the future, retrocursors reenact behavior of three degraded levels demanding im- have used that knowledge successfully patterns that have led to accidents in the mediate manual control by the crew in to avoid the latest crashes. Current pro- past. At the time of this paper’s writing in an ambiguous situation. Out-of-envelope cesses for identifying, defining, communi- late June 2009, the most recent of these airspeed signals could have resulted from cating, and acting on lessons to be learned was the loss of Air France Flight 447 over pitot tube icing in severe thunderstorm; are inadequate to take advantage of the the equatorial Atlantic enroute from Rio to • overstress separation of the airplane’s opportunities offered by investigated accidents. The official motto of ISASI is “Safety Authors’ Note through Investigation.” ISASI was in- corporated 45 years ago, and its official After we submitted the paper to munication alternatives. motto was adopted at that time. Its Code ISASI’s seminar committee, our con- 3. Digitizing operational data has of Conduct has been in effect for more than tinuing analysis revealed more insights opened the capability for capturing 25 years. It states about successfully identifying and ap- real-time aviation information and new “5. ACCIDENT PREVENTION… plying lessons that should be learned opportunities for instant access to the Each member shall: “5.1 Identify from from accidents and incidents. As behavioral data needed to achieve les- the investigation those cause-effect re- reported during our seminar presenta- sons learning system results. lationships about which something can tion, we realized that 4. Evaluating the occurrence of “ret- be done reasonably to prevent similar 1. Standardizing input data structure is rocursor” accidents, which repeat the accidents;” and “5.3 Communicate facts, essential to identifying and communi- lessons that should have been learned analyses, and findings to those people or cating behavior sets that were signifi- historically, provides a metric both for organizations which may use such infor- cant during the mishap process. the insightfulness and accuracy of in- mation effectively….” 2. Employing social networking to vestigations and their reports, and the Recurrence of accidents from similar transmit accident data is both quicker effectiveness of lessons learned users’ sources should have been reduced sub- and more accurate than current com- applications of the lessons of history. ◆ stantially, if not eliminated, had investiga-

22 • ISASI Forum January–March 2010 vertical stabilizer and subsequent loss of internally or made public in various media, sessions, codes or regulations, and books. control; or as databases, reports, articles, papers, Deriving lessons from the data depend • a combination of both. books, stories, graphics, training materi- on someone recognizing the value of the Continental-Colgan Flight 3407, a als, checklists, etc. Published data are content and generating and communicat- Bombardier Dash 8-Q400 that crashed on stored in organizational files or databases ing the lessons. approach to Buffalo, N.Y., on Feb. 12, 2009, for retrieval and use. They may also find Reported investigation data may also and Turkish Airlines Flight 1951, which their way eventually into revised proce- be used for research to develop lessons crashed on approach to Amsterdam’s dures, standards, and regulations. learned in the form of historical trends or Schiphol Airport 13 days later, were high- Dissemination practices vary, but statistical correlations, using statistical profile retrocursors. In both cases minor include electronic dissemination in com- analyses or data mining techniques. Data anomalies distracted the crews from the puterized databases, e-mails, and Internet are frequently abstracted or characterized principal airmanship rule: “First fly the sites. Non-electronic dissemination may to generate “taxonomies” of causes and airplane.” Crew distraction accidents have include hard-copy investigation reports, causal factors referenced in investiga- been a bane for decades. tables, checklists, on-the-job training, tion report databases, safety digests, and A third retrocursor was the FedEx safety meetings, standardization, training investigation software. MD-11 landing crash at Narita, Japan, on We analyzed contemporary “lessons March 22, 2009, that duplicated a similar Ludwig Benner joined learning” practices, focusing on how data FedEx MD-11 at Newark, N.J., in 1997. A the NTSB in 1970 and are analyzed to isolate and describe the China Airlines MD-11 crash at Kai Tak in headed its Hazardous lessons that should be learned. Major August 1999 exhibited similar operational Materials Division inadequacies we observed include behavior. Why hadn’t the lessons that from 1976-1982. He has • Authors variously define lessons as should have been learned from earlier published more than 50 causes, cause factors, findings, conclu- accidents been communicated well enough articles and a book on sions, recommendations, issues, state- to the crews and internalized sufficiently accident investigation process research. ments, or scenarios in texts of narrative to prevent the retrocursors? He has taught accident investigation and reports. led development of investigation support • Authors often obscure lesson data Contemporary lessons software. He is a correspondent member within excessive wordiness and jargon. learned practices of the European Safety and Reliability • Authors do not explicitly list lessons Are there formal contemporary lessons Data Association’s Accident Investiga- learned as such. learned “systems,” and, if so, why don’t tion Project Group, chairman of ISASI’s • Analysts rarely categorize investiga- they maximize learning from lessons Board of Fellows, a Fellow of the Inter- tion data to facilitate end-users’ retrieval generated by accidents? national System Safety Society, and a and use. Historically, investigators acquire, retired professional safety engineer. • Analysts assume that proposed changes document, and report factual data in many alter system behavior favorably, without forms and formats, by many diverse and Ira Rimson completed a testing. often-isolated systems. These data are Navy career as director • Lessons are “pushed” to pre-estab- used by investigators and analysts to piece of safety of the Naval lished recipients, but must be “pulled” by together a description and explanation Air Systems Command. other users. of what happened, usually in narratives He joined senior FAA What inadequacies of current lessons or on preexisting forms, using natural retirees founding Avia- learned practices have already been language. These accident data comprise tion Resources Group in reported? Dr. Paul Werner and Richard the bases for cause-oriented conclusions 1981, heading its safety services. He has Perry cited the following barriers to effec- from which findings and recommendations consulted and performed private safety tively capture and apply lessons learned are derived. Causes, findings, and recom- investigations since 1977. He is a Fellow by investigators: mendations rarely specify the lessons of ISASI, the American Academy of • Data are not routinely identified, col- learned from an investigation. Analysts Forensic Sciences, and the American lected, and shared across organizations abstract, code, characterize, aggregate, College of Forensic Examiners, and is and industries. or otherwise refine or condense the data. a senior member of the International • Unsystematic lessons are too difficult They are then “published”: disseminated System Safety Society. to use because

January–March 2010 ISASI Forum • 23 —there is too much material to search, —they are formatted differently in differ- ent reports, or —they are not readily available. • Applications are unplanned and hap- hazard. • “Taxonomy” categories obscure data searches. We observed two categories of inhibi- tions to developing lessons learned within the investigation process itself. The more fundamental is a mindset of unquestioned acceptance of “how things have always been done,” and can include • archaic accident “causation” models, • unwillingness to share investigation data, Figure 1. User’s Component of Lessons Learning Model • language barriers that obscure identi- fication of relevant behaviors, do not prioritize information needed by guage. Common grammar, structure, and • data loss from software obsolescence end-users who initiate behavioral changes. format for investigation input data should and lack of standardization, and Investigations focus on determining thoroughly and objectively describe • concerns for legal liability. “causes”: cause factors, multiple causes, behaviors that constituted the mishap A secondary category frequently de- “root” causes, and other easily labeled process. Investigators must test behav- rives from the obstacles above and occurs actions from which investigators and ana- ioral data sequencing, coupling, and logic at the levels of individual investigators and lysts infer lessons and propose corrections. during investigations. That will ensure the analysts. It includes missing data, biased Investigation report authors typically do identified, needed data will be developed scope and data selection, logic errors, not provide data in forms from which end- and delivered to end-users in formats they misinterpretation or misrepresentation users can derive the behavioral changes can internalize readily and directly, and of observations, flawed assumptions, and they need to prevent recurrence. Instead, provide them with unambiguous reasons premature conclusions during investiga- the “expert” investigating agencies select for changing the behaviors that produced tions. Each inhibits development of useful changes they deem desirable and direct the unwanted outcomes. lessons. them to target audiences of their choice A third major challenge is to define in the form of recommendations. the structure and content of the lessons Clarification of terms learning system. It must satisfy end-users’ Lessons learned are often considered to be Developing lessons learning systems needs and, at the same time, support ma- new knowledge obtained from experience, The challenges to lessons learning sys- chine documentation, processing, remote applied to benefit future performance. tems are to collect accurate mishap-based access, interoperability among users, and The questions arise: knowledge about data and communicate them quickly and easy access. Its goal should be timely and what? And how can we put it to beneficial efficiently to end-users that can develop efficient identification of the behavioral use? We find it helpful to think of the new and implement changes. changes needed to effect the lessons that knowledge generated by investigations The first challenge is to define the end- need to be learned, and their delivery to as clarification of what happened, and users of lessons from investigations and those who need to learn them. why it happened. That new knowledge how they would use them. End-users are can be applied to change behaviors of all entities that can change behaviors that A Lessons learning system people, systems, or energies. This concept led to an undesired outcome, or initiate We developed a model of a comprehensive distinguishes between the lessons and the new avoidance behaviors, in their opera- lessons learning system from investiga- learning, identifying the tasks required of tions, in objects or systems they design tions by tracking the functions and tasks those documenting the lessons to describe or operate, or in energies they manage. required to achieve changed behaviors. and communicate them so that end-users Current investigation data are designed to The system begins with capturing the can apply them to initiate desired behav- fulfill the needs of the agency conducting lessons-to-be-learned data during the ac- ioral changes. the investigation. The investigation com- cident process and ends with an archive What data are needed to develop lessons munity would better serve its prevention of lessons and responses that have been to be learned? goal by devoting priority attention to tested and shown to produce effective Mixed perceptions of the investigation fulfilling the lessons learning data needs results. data that need to be acquired and dis- of end-users that can apply that new Users’ components of the learning seminated as lessons may be the greatest knowledge to changing behaviors. system model are shown in Figure 1. The obstacle to learning. Accident causation A second challenge is to systematize model assumes that lessons learned are and investigation models influence those investigation data inputs and outputs by new knowledge developed by investigators perceptions. Current investigation goals standardizing and applying scientific lan- about behaviors that interacted during the

24 • ISASI Forum January–March 2010 accident process. Each task can be decom- • timely repository updating. • objective verification and validation to posed further for specific applications. From a developer’s perspective, shown ensure quality before dissemination. What should users expect from a lessons in Figure 2, investigation components • provisions to modify and update col- learning system (LLS)? LLS users deal of LLS should support development of lected data with new knowledge. with dynamic processes. LLS documenta- lessons-to-be-learned source data with During the study of lessons learned tion must be behaviorally consistent with such attributes as processes; we noted two other significant dynamic processes to enable comparing • establish investigation goals to provide observations— behavior sets, defining alternative changes lessons that can change future behaviors. • Special investigating bodies appointed to behavioral relationships, and predict- • establish an input-output framework for to inquire into specific accidents often ad- ing effects that changes might introduce. defining what happened by LLS data sets dress lessons learned explicitly in their The system should enable translating that describe behaviors in non-judgmental reports. Yet the reports we surveyed LLS response options into some form of and logically verifiable terms. by traditional government investigation change management analysis, and into • focus on behavior data acquisition and bodies lack a discrete section addressing, instructions to incorporate the changed processing. documenting, or summarizing the lessons behavior in the targeted person, object, • specify a structure for input data docu- found during the investigation. No stan- energy, or process. Therefore, LLS must mentation that ensures data consistency and dardized guidance exists for doing so. For describe behavioral interactions among economy, and facilitates data coupling and example, ICAO Annex 13 does not define people, objects, and energies, rather than support for documenting output LLS. or otherwise mention lessons learned. linear “causes” or abstracted “factors.” • machine supportable input data man- Lack of standardized methodology for reporting “lessons” burdens prospective end-users by requiring them to search and interpret voluminous data with little assurance of finding what they need to initiate changed behaviors. • LLS requires designers to make strategic choices about investigation process frameworks, purposes, scope, and data structures; LLS content, form and language; and appropriate choices of repositories, distribution, updating, and metrics. Traditional (or inadvertent) strategic system design choices have ad- versely affected the utility of current LLS processes, operation, and performance. Figure 2. Developers’ Components of Lessons Learning Model Conclusions Ideal LLS attributes include agement, display, and expansion to reduce Contemporary investigation-based LLS • open to multiple change options. latency. has not prevented recurrence of accidents • inclusive of context identification. • objective quality assurance and valida- from known behaviors that produced un- • accessible expeditiously to all potential tion processes. desired outcomes. Their primary weakness users. LLS documentation derived from inves- lies in neglecting the knowledge require- • backward compatibility with legacy tigation descriptions must fulfill end-users’ ments of users capable of changing those data repositories. needs. System attributes should include behaviors. Current reports are too often • minimize elapsed time (latency) be- • requirements that behavioral data inconsistent, ambiguous, and vague. In- tween the occurrence that generates data outputs provide context, minimize inter- vestigating agencies should design LLS to for LLS, and when the lesson becomes pretive and analytical workload, maximize identify and report all the lessons that can available to end-users. signal-to-noise ratio, and reduce latency. be learned from each mishap, record them • maximize “signal to noise” ratio; i.e., • provisions for machine processing explicitly for ready access and retrieval, maximizing relevant content. support, interoperability, and reposi- oversee their application where they can • enhanced determination of relevance. tory uploading capabilities to accelerate contribute to avoiding retrocursors,” and • enhanced assimilability. documenting and distributing lessons to measure the results. The first steps needed • scalability: the ability to increase data all collections. to improve lessons learning practices are quantity without sacrificing quality. • establish accessible Internet LLS out- • redesigning the form and substance • cost sensitivity: the value of the system put data libraries and end-user notification of lessons-to-be-learned source data to in terms of results it produces. to support both “push” and “pull” data improve their usefulness for users, and • improved acceptance and more actions distribution and minimize latency. • redefining investigation data product initiated by end-users. • easy repository access, with search and specifications to require that lessons • performance metrics for behavioral filter capability to minimize end-user ac- learned be an explicit documented output changes. cess time, cost, and workloads. of the investigation processes. ◆

January–March 2010 ISASI Forum • 25 ISASI ROUNDUP

ISASI 2010: Sapporo, Japan

The International Society of Air Safety with its Odori Park and many shopping enter the country for the seminar. To see Investigators 41st annual international districts. Seminar details are now in the if your country has such an arrangement conference on air accident investigation, reconciliation stages and will be posted go to website http:///www.mofa.go.jp. “ISASI 2010,” will be held in Sapporo, in the next issue of ISASI Forum, as On the upper tool bar click “visa” then Japan, September 6-9. well as on the seminar website, which is “Other nationalities, if visa not required” The event is being hosted by Japan’s also nearing completion. and scroll down to the 63-country local Seminar Committee under the listing. ◆ auspice of the Japan Transport Safety General information Board (JTSB). The Seminar Commit- Sapporo is the capital of the island of tee consists of Chairperson: Mamoru Hokkaido, which is located at the north Kapustin Scholarship Sugimura; Technical Program: Yukiko end of Japan, near Russia. The city’s Applications Now Being Kakimoto and Robert Matthews; Spon- “birth” is recorded as being 1868, and sorship: Koichi Saito and Ron Schleede; its name can be translated to “dry, great Sought Companion Program: Hideyo Kosugi; river.” Sapporo is located in the south- The ISASI Rudolf Kapustin Memorial Website: Hiroshi Itokawa; and Regis- west part of Ishikari Plain and the allu- Scholarship Fund administrators, Rich- tration: Sharon Morphew and Masaru vial fan of the Toyohira River, a tribu- ard Stone and Ron Schleede, urge all Chiba. tary stream of the Ishikari River. The members to quicken their search for stu- The decision to host the conference western and southern part of Sapporo dents to apply for the memorial scholar- is the planners’ acknowledgement that is occupied by a number of mountains ship offered by ISASI. The deadline for in recent years the environment sur- including, Mount Teine, Maruyama, and applications is April 15. Full application rounding air accident investigators has Mount Moiwa. details and forms are available on the dramatically changed. From the world- Today it is a city of 2 million and ranks ISASI website, http://isasi.org. wide perspective, with the introduction as the fifth largest city in Japan. Many Given the lead time to the applica- of bigger aircraft into airline fleets and will recall that it hosted the Winter tion deadline, the fund administrators increased influx of Asian passengers, Olympics in 1972. Seminar attendees can encourage all ISASI societies, chapters, more and more people will be on the expect the weather to range from a high working groups, and individual members move by air around the world. From the of 22C/72F to a low of 14C/55F, or a daily to promote the availability of the ISASI investigators’ standpoint, it is imperative mean of 18C/64F. The airport serving scholarship and its application proce- to understand local culture, customs, and Sapporo is New Chitose Airport. It is dures to students, student groups, and peoples’ sentiments to overcome cultural the island’s major airport and is serviced education centers whenever the oppor- differences and language barriers to bet- directly from a number of Asian cities. tunity presents itself. Fund administra- ter cope with severe accident situations. Travel to the hotel from Chitose is by tors stress the need for applicants to Taking such a trend into consider- taxi, rail, or bus. adhere to the deadline date and to not ation, the conference theme selected is Only the Japanese yen is acceptable exceed the word limit of the required “Investigating ASIA in Mind-Accurate, at regular stores and restaurants. Coins 1,000-word essay. Speedy, Independent, and Authentic.” are in denomination of 1, 5, 10, 50, 100, To date, ISASI has awarded 15 The theme embodies the seminar goal and 500 yen. Bill denominations are scholarships since the inception of the of presenting material that reflects the 1,000, 2,000, 5,000, and 10,000 yen. Credit program in 2002. Continued funding for latest trends and practices in accident cards are widely used in the urban areas. the Memorial Fund is through dona- investigation and prevention, with Tipping is not customary in restaurants, tions, which in the United States are particular emphasis on Asia overcom- hotels, taxies, or anywhere. Instead tax-deductible. An award of US$2,000 ing cultural and language problems. hotels impose a service charge, and a 5% is made to each student who wins the The 5-day program consists of a day of consumption tax for goods and services competitive writing requirement, meets tutorial workshops, a 3-day technical exists throughout Japan. Electricity is the application requirements, and who program, and an optional day of touring. 100V/50Hz (60Hz in western Japan) for registers to attend the ISASI annual The tutorial and technical program will appliances with a two-prong plug. seminar. The award will be used to cover be held at Royton Sapporo (http://www. Citizens of countries that have a recip- costs for the seminar registration fees, daiwaresort.co.jp/English/29_royto. rocal visa exemption arrangement with travel, and lodging/meals expenses. Any html), located in the center of Sapporo Japan are not required to have a visa to expenses above and beyond the amount

26 • ISASI Forum January–March 2010 ISASI 2010 CALL FOR PAPERS

ISASI’s 41st annual seminar will be may address any segments of the air held in Sapporo, Japan September carrier industry or general aviation. 6-9. Our Japanese hosts have chosen An expression of interest in deliver- the theme of “Investigating ASIA in ing a paper should be sent by e-mail Mind-Accurate, Speedy, Independent, no later than February 1 to bob. and Authentic.” Papers are invited [email protected] or to kakimoto- that address methods, issues, or past [email protected]. Please include a Nominations Sought for findings from accident investigations working title for your presentation. and analyses that address the issues Abstracts must be submitted by April The Jerome F. Lederer of timeliness, technical competence, 1, from which final papers and pre- Award and intellectual integrity that are sentations will be invited for submis- free of political constraints. Topics sion by June 15. ◆ Nominations for this prestigious award are open until May 31. Chairman Gale Braden urges ISASI members to look of the award will be borne by the re- will be advised if donations are made in for deserving candidates in the various cipient. ISASI corporate members are honor of a particular individual. fields of aircraft accident investigation encouraged to donate “in kind” services Students who wish to apply for the and nominate those meeting the criteria. for travel or lodging expenses to assist scholarship may acquire the application Year 2009 awardees were Capt. Richard student scholarship recipients. form and other information at the ISASI B. Stone and the Australian Transport Students granted a scholarship also website: www.isasi.org. Students may Safety Bureau (ATSB). It was only receive also request applications by sending an the second time that the Award was • a one-year membership to ISASI. e-mail to [email protected]. The ISASI of- bestowed on two recipients. • tuition-free attendance from the fice telephone number is 1-703-430-9668. Each year, at our annual seminar, we Southern California Safety Institute recognize positive advancements in the (SCSI) to any regularly scheduled Application requirements art and science of air safety investigation SCSI course. This includes the 2-week • Applicants must be enrolled as full- through the Jerome F. Lederer Award. Aircraft Accident Investigator Course or time students in a recognized (note The criterion for the award is quite any other investigation courses. Travel ISASI-recognized) education program, simple. The Lederer Award recognizes to/from the course and accommodations which includes courses in aircraft outstanding contributions to technical are not included. For more information, engineering and/or operations, avia- excellence in accident investigation. go to http://www.scsi-inc.com/. tion psychology, aviation safety and/or Any member of the Society may submit • a tuition-free course from the Trans- aircraft occurrence investigation, etc., a nomination, and the nominee may be portation Safety Institute. Travel to/ with major or minor subjects that focus anyone in the world. The Award may be from the course and accommodations on aviation safety/investigation. given to a group of people or an organi- are not included. More information is • The student is to submit a 1,000 (+/- zation, as well as an individual, and the available at http://www.tsi.dot.gov/. 10 percent) word paper in English ad- nominee does not have to be a Society • tuition-free attendance from the dressing “The Challenges For Air Safety member. The Award may recognize Cranfield University Safety and Acci- Investigators.” (Adherence to length a single event, a series of events, or a dent Investigation Centre for its 5-day requirements is important.) lifetime of achievement. The ISASI Accident Investigation Course, which • The paper is to be the student’s own Awards Committee considers such traits runs as part of its masters degree pro- work and must be countersigned by the as duration and persistence, standing gram at the Cranfield campus, 50 miles student’s tutor/academic supervisor as among peers, manner and techniques of north of London, UK. Travel to/from authentic, original work. operating, and, of course, achievements. the course and accommodation are not • The papers will be judged on their Each nominee competes for 3 years included. Further information is avail- content, original thinking, logic, and unless selected. If not selected during able at www.csaic.net/. clarity of expression. that time, the nominee can be nominated The Fund is administered by an • The student must complete the appli- after an intervening year for another appointed committee and oversight cation form and submit it to ISASI with three-year period. (See “Lederer Award of expenditures is done by the ISASI the paper by April 15. Selection Process,” Forum, October- treasurer. The Committee ensures that • Completed applications should be for- December 2009, page 18). the education program is at an ISASI- warded to ISASI, 107 Holly Ave., Suite This is a prestigious award usually recognized school and applicable to the 11, Sterling, VA 20164-5405 USA. E-mail resulting in good publicity for the recipi- aims of the Society, assesses the applica- address: [email protected]. Telephone: ent, and might be beneficial in advancing tions, and determines the most suitable 703-430-9668. a recipient’s career or standing in the candidate(s). Donors and recipients • The decision of the judges is final. ◆ community.

January–March 2010 ISASI Forum • 27 ISASI ROUNDUP Continued . . .

In Memoriam

Michael J. Baker (LM3240), Wellington, Nomination letters for the Lederer New Zealand, July 30, 2009 help field investigators improve compre- Carlos Jose Bondio (LM2120), Pica Award must be limited to a single page. Cordoba, Argentina, May 29, 2009 hension of human factors in accident in- Nominations should be mailed, or e- Robert R. Crispin (MO0919), Higley, vestigation by using human factor tools mailed, to the ISASI office or directly Ariz., USA to understand, guide, and report the Berry M. Sweedler (MO4862), Lafay- to Awards Committee Chairman Gale ette, Calif., USA ◆ role the human played in the accident/ Braden, 13805 Edmond Gardens Drive, incident. Unfortunately, the effort’s goal Edmond, OK 73013, USA. E-mail ad- was not reached. However the need to dress: [email protected]. ◆ Committee by April 1. Any member in provide human factors information rela- good standing may submit a nomination. tive to accidents and incidents remained ISASI Executive Each potential candidate whose name a priority for the ISASI Human Factors is submitted to the Nominating Commit- Working Group. Nominations Due April 1 tee must have consented to the submis- To fill that need, Chairman Stone Nominations for election to the ISASI sion. The nominator must submit a short secured permission from the Transport offices of president, vice-president, biographical sketch of the nominee. Safety Board of Canada and Leo Donita, secretary, treasurer, U.S. councillor, and Nominees must be at least a full member manager of human factors, to borrow international councillor for the term in good standing to be eligible for office. from material the TSB uses in its Hu- 20011-2012 are due to the Nominating Nominations should be sent to ISASI, man Factors Course taught in Canada attention Nominating Committee, Park and attended by every investigator asso- Center, 107 East Holly Ave., Suite 11, ciated with the Board. The result is the Sterling, VA, USA, 20164-5405. ◆ 30-page booklet that highlights issues of human factors or performance in acci- Moving? dents or incidents. It is intended for use Please Let Us Know ISASI Issues Guidelines by accident investigators-in-charge (IIC) Member Number______For Human Factors specifically on the subject of human factors. ISASI especially wants to pro- Fax this form to 1-703-430-4970 or mail to Investigation vide these basic concepts to IIC’s who ISASI, Park Center 107 E. Holly Avenue, Suite 11 ISASI’s International Council members haven’t had an opportunity to attend Sterling, VA USA 20164-5405 adopted the newly developed ISASI formal courses in accident investigation Guidelines for Investigation of Human or human factors. However, users should Old Address (or attach label) Factors in Accidents or Incidents book- bear in mind that the guidelines are not Name______let prepared by Richard Stone, chairman a substitute for specialists in human fac- Address______of the ISASI Human Factors Working tors. The booklet is posted on the ISASI City______Group and ISASI executive advisor. website: http://isasi.org, under the tabs, The booklet is an outgrowth of a prior About ISASI, General, Guidelines. State/Prov.______2-year-long effort to prepare material to The following is excerpt from the Zip______Guidelines. Country______New Address* Correction Background Accident statistics show that issues Name______The Past Lederer Award Winners associated with human performance Address______listing in the October-December are major contributors to incidents and City______2009 Forum, page 19, inadvertent- accidents in commercial aviation and State/Prov.______ly omitted the following recipients: can become the subject of much contro- 1977-Samuel M. Phillips, 1978-Al- versy. Worldwide, investigations vary Zip______len R. McMahan, 1979-Gerard M. significantly with respect to the beliefs Country______Bruggink, and 1980-John Gilbert about the role of humans and appropri- E-mail______Boulding. Our apologies to these ate methods for investigating human *Do not forget to change employment and dedicated safety advocates. ◆ factors. Ideally, an investigation will e-mail address. seek to understand the context of human

28 • ISASI Forum January–March 2010

NEW MEMBERS

Corporate Civetti, Chad, R., Wilbraham, MA, USA Monteith, David, J., Toronto, ON, Canada Australian and International Pilots Association Cowell, Steven, R., Denver, CO, USA Nicholson, Eric, S., Burkburnett, TX, USA (AIPA), Australia Curran, Thomas, F., Malahide, Co. Dublin, Odle, William, D., Saint Charles, MO, USA Angela Williams, Research & Technical Ireland Pantas, Lee, J., Leesburg, VA, USA Support Officer Dunn, Leigh, M., Leighton Buzzard, Pellegrino, Gina, T., Rumford, RI, USA Capt. Barry Jackson, President Bedfordshire, United Kingdom Poole, Michael, J., Frankston, VIC, Australia AVISURE, Australia Ellis, Clyde, N., Hoover, AL, USA Rajnicek, Mary, J., Roseville, MI, USA Jeff J. McKee, Principal Research Scientist Evans, Zoe, J., Christchurch, New Zealand Rivera, Edwin, F., Holly Hill, FL, USA Phillip E. Shaw, Managing Director Granger, Scott, C., Powell Butte, OR, USA Robbins, Brian, L., Columbus, NJ, USA Nova Aerospace, Australia Grounsell, Colin, P., Lower Hutt, New Zealand Ross, Stewart, G., Fraser, ACT, Australia Brett Martin, Systems Engineer Hanson, John, M., Prescott, AZ, USA Satran, Maximilian, F., Melbourne, FL, USA Seamus Miller, Systems Engineer Henegar, Kristopher, J., Melbourne, FL, USA Schoenberg, Daniel, I., Melbourne, FL, USA Hodges, Nancy, O., Panama City Beach, FL, Schwyzer, Walter, Zell, Switzerland Individual USA Sedor, Joseph, M., Ellicott City, MD, USA Abdul Wahed, Ismaeil, Dubai, United Arab Hoepel, Jonathan, G., Melbourne, FL, USA Sevillian, Dujuan, B., Wichita, KS, USA Emirates Horton, Curtis, H., Melbourne, FL, USA Simen, Nelly, A., Melbourne, FL, USA Aziz, Imran, Karachi, Pakistan Huysmans, Roel, J.M., Koksijde, Belgium Stacy, Phillip, E., Sunbury, Vic, Australia Bailey, Mark, N., Surrey, BC, Canada Jackson, Lincoln, B., Kingston, Jamaica Staszel, Michael, M., Park Ridge, IL, USA Barriere, Brendan, J., Melbourne, FL, USA Jones, Richard, A.D., Kingston, Jamaica Steinberg, Markus, HB, Hamburg, Germany Bartenstein, Jennifer, A., Manassas, VA, USA Lattimore, David, Auckland, New Zealand Telya, Murtaza, Palmerston North, Beyer, Deanna, M., Golden Valley, MN, USA Lomas, Stuart, W., Port Melbourne, VIC, Manawatu, New Zealand Bledsoe, Justin, M., Anchorage, AK USA Australia Thorpe, Ricki, K., Yukon, OK, USA Bolton, Michelle, A., Deland, FL, USA Marinkovich, Leslie, J., South Auckland, New Timmermans, Linda, Den Haag, Byrne, Alan, J., Vincentia, NSW, Australia Zealand Netherlands Carl, Cecil, R., Pembroke Pines, FL, USA McCarthy, Michael, E., Sanford, FL, USA Van Geffen, John, T., San Francisco, CA, USA Carl, Katherine, B., Pembroke Pines, FL, USA McGregor, Andrew, B., Auckland, 0620, New Villela, Bruno, T., Port Orange, FL, USA Cepus, Elvis, Vancouver, BC, Canada Zealand Weselak, Marissa, K., Melbourne, FL, USA Chopin, Robert, C., Bridgeman Downs, McKinnon, Joshua, D., Jacksonville, FL, USA White, Benjamin, M., Tolland, CT, USA OLD, Australia Mee, Jeffrey, J., Portsmouth, NM, USA Wulber, Mark, S., Independence, KY, USA ◆ performance and how it contributes to occurred rather than who was at fault the actions taken or decisions made. the observed behaviors and decisions. will yield more vital and accurate infor- • Every accident or incident investiga- Whether designing equipment, plan- mation. It is currently true that accident tion should initiate human performance ning training programs, integrating new investigators can begin an investigation data collection, as soon as possible, for aircraft into a fleet, or finding out what by sifting through pieces of aircraft data that are easily lost or tainted with went wrong in an accident or incident, wreckage and have no presumption of a the passage of time. understanding the human component is mechanical fault. It should also be true • That appropriate human factors ex- critical to developing and maintaining a that investigators can gather data on hu- pertise is brought to bear on all investi- safe air operation. man performance and the conditions of gations of human performance issues. Accident and incident investigation human performance without presuming • Accident and incident databases presents a real opportunity to examine human error or negligence. worldwide share a common taxonomy the interactions between the human and • The identification of a human for identifying and listing human perfor- the other system components. While “error”—which simply refers to a devia- mance issues so that the databases can human factors expertise is available to tion between the behavior observed or be used to track trends over time. inform investigations, this expertise is decisions made by a human (e.g., pilot) • Investigations to assess criminal not uniformly applied. By developing and the behavior or decision that, in behavior alleged to have occurred in an new guidelines, ISASI intends to en- hindsight, seemed most appropriate—is aircraft accident should be carefully con- hance existing guidance documents now the starting point of the investigation ducted so as not to impact negatively the available to investigators. ISASI hopes into the precursors of human perfor- air safety investigation. States that have these guidelines will highlight critical mance contributions to an accident or attempted to conduct both an air safety areas that affect human performance. incident. The identification of “human and criminal investigation concurrently, error” is not a stopping point. particularly where human performance General statement of suggested policies • The collection of human performance is involved, have found negative effects To provide for optimum investigation of data should not be seen as implying that to both investigations. ◆ the role played by humans in accidents human error is a working hypothesis for or incidents we suggest the investigation. Initial interviews of Toulouse, France, • That all agencies involved in accident operational personnel involved in the acci- investigation endorse the policy that dent or incident (e.g., pilots, air traffic con- Is Site of ESASI Seminar the investigation of human performance trollers, maintenance technicians) should The European Society of Air Safety proceeds without the presumption of hu- be conducted in a way to maximize the Investigators has selected Toulouse, man error or negligence. An investiga- retrieval of information about the event; France, as the site of its third air safety tive process that seeks to ascertain what they should not focus on finding fault with seminar to be held on April 29-30, with

January–March 2010 ISASI Forum • 29 ISASI ROUNDUP ISASI Information Continued . . .

OFFICERS an optional technical visit to the Airbus Air Canada Pilots Association’s (ACPA) President, Frank Del Gandio ([email protected]) A380 assembly line followed by a meal technical and safety division jointly Executive Advisor, Richard Stone on April 29. The event will emphasize hosted International Winter Operations ([email protected]) Vice-President, Ron Schleede current European issues in the investi- Conference held in Ontario, Canada, on ([email protected]) gation and prevention of accidents and Oct. 7-8, 2009. CSASI’s president, Bar- Secretary, Chris Baum ([email protected]) incidents. Presentations will address bara Dunn, said the effort carried the Treasurer, Tom McCarthy current issues in the European environ- theme “Winter Operations: Safety ([email protected]) ment and the challenges of modern air Is No Secret.” safety investigations. “It was an inaugural event at which COUNCILLORS Australian, Lindsay Naylor Toulouse, also known as la ville rose participants learned how to cope with ([email protected]) (Pink Town), is the fourth largest city winter weather in all its forms,” said Canadian, Barbara Dunn ([email protected]) in France. Located in the southwest Dunn. European, Anne Evans of France, within a few hours of the The first day’s program began with ([email protected]) International, Caj Frostell Atlantic Ocean, the Mediterranean Sea, a welcome from Capt. Paul Strachan, ([email protected]) and the Spanish border, Toulouse is a ACPA president. Capt. Robert “Hoot” New Zealand, Peter Williams ([email protected]) beautiful town filled with friendly people Gibson (USN, Ret.), mission commander United States, Toby Carroll and interesting architecture. It is also aboard the space shuttles Challenger, ([email protected]) the home of the European aerospace Columbia, Atlantis, and Endeavor industry and, in particular, Airbus gave the keynote address and spoke of NATIONAL AND REGIONAL The 2-day program (details to be con- the challenges and high points of space SOCIETY PRESIDENTS Australian, Lindsay Naylor firmed) will be held at the École Nation- travel. ([email protected]) ale de l’Aviation Civile (ENAC), situated Additional speakers included Chris Canadian, Barbara M. Dunn ([email protected]) in the south of Toulouse. Hotel accom- St. Clair of the Weather Network and European, David King modation has been arranged at the representatives from the Toronto and ([email protected]) Latin American, Guillermo J. Palacia Mercure Toulouse Saint Georges with Chicago area airports. Chief Robert (Mexico) a discounted room rate of 130.90 Euros Donahue of the Massachusetts Port New Zealand, Peter Williams ([email protected]) (taxes and breakfast included). The Authority Fire Rescue Department dis- Russian, Vsvolod E. Overharov hotel is located in the Toulouse city cen- cussed airport emergency planning, and ([email protected]) SESA-France Chapter,Vincent Fave ter, 500 m from the capitol and 800 m other excellent presentations on deicing ([email protected]) from the Musée des Augustins. Hotel and runway contamination completed United States, Toby Carroll ([email protected]) bookings should be made directly with the first day’s program. That evening the hotel. Telephone: (+33)5/62277979. found everyone enjoying a welcome UNITED STATES REGIONAL Fax: (+33)5/62277900. E-mail: H0370@ reception while catching up with old and CHAPTER PRESIDENTS accor.com. new friends. Alaska, Craig Bledsoe For seminar bookings and further The same level of excellent presen­ ([email protected]) Arizona, Bill Waldock details, contact ESASI Councillor tations continued throughout the second ([email protected]) Anne Evans. Telephone: +44 (0) 1252 day. Dallas-Ft. Worth, Tim Logan ([email protected]) 510300, e-mail: [email protected]; or Representatives from Bombardier, Great Lakes, Matthew Kenner contact ESASI Secretary John Dunne. Airbus, Embraer, and Boeing presented ([email protected]) Los Angeles, Inactive Telephone: +44 (0) 7860 222266, e-mail: aspects of winter operations, including Mid-Atlantic, Ron Schleede [email protected]. ◆ airframe and engine icing, both on the ([email protected]) Northeast, David W. Graham ground and in flight, runway contami- ([email protected]) nation, and takeoff and landing per- Pacific Northwest, Kevin Darcy CSASI, ACPA Score ([email protected]) formance. Transportation Safey Board Rocky Mountain, David Harper and NTSB representatives discussed ([email protected]) High with Winter Ops San Francisco, Peter Axelrod challenges, past and present, related to ([email protected]) Conference flying in winter conditions. Southeastern, Robert Rendzio More than 200 persons from 13 countries It is hoped that the Conference will ([email protected]) attended the Canadian SASI and the become a biannual event. ◆

30 • ISASI Forum January–March 2010 Allied Pilots Association Int’l Assoc. of Mach. & Aerospace Workers COMMITTEE CHAIRMEN Interstate Aviation Committee Audit, Dr. Michael K. Hynes American Eagle Airlines ([email protected]) American Underwater Search & Survey, Ltd. Irish Air Corps Award, Gale E. Braden ([email protected]) AmSafe Aviation Irish Aviation Authority Ballot Certification, Tom McCarthy Aramco Associated Company Japan Airlines Domestic Co., LTD ([email protected]) ASPA de Mexico Japanese Aviation Insurance Pool Board of Fellows, Ludi Benner ([email protected]) Association of Professional Flight Attendants Jeppesen Bylaws, Darren T. Gaines ([email protected]) Atlantic Southeast Airlines—Delta Connection JetBlue Airways Code of Ethics, Jeff Edwards Australian Transport Safety Bureau Jones Day ([email protected]) Aviation Safety Council KLM Royal Dutch Airlines Membership, Tom McCarthy Aviation Safety Investigations, UK Korea Air Force Safety Ctr. ([email protected]) Avions de Transport Regional (ATR) Korea Aviation & Railway Accident Investigation Nominating, Tom McCarthy BEA-Bureau D’Enquetes et D’Analyses Board ([email protected]) Board of Accident Investigation—Sweden Kreindler & Kreindler, LLP Reachout, John Guselli Boeing Commercial Airplanes L-3 Communications Aviation Recorders ([email protected]) Learjet, Inc. Seminar, Barbara Dunn ([email protected]) Bombardier Aerospace Bundesstelle fur Flugunfalluntersuchung—BFU Lockheed Martin Corporation CAE-Flightscape, Inc. Lufthansa German Airlines WORKING GROUP CHAIRMEN Cathay Pacific Airways Limited MyTravel Airways Air Traffic Services, Scott Dunham (Chair) Cavok Group, Inc. National Aerospace Laboratory, NLR ([email protected]) Centurion, Inc. National Air Traffic Controllers Assn. Ladislav Mika (Co-Chair) ([email protected]) Charles Taylor Aviation, Singapore National Business Aviation Association Cabin Safety, Joann E. Matley ([email protected]) China Airlines National Transportation Safety Board Corporate Affairs, Erin Carroll Cirrus Design NAV Canada ([email protected]) Civil Aviation Safety Authority Australia Nigerian Ministry of Aviation and Accident Flight Recorder, Michael R. Poole Investigation Bureau ([email protected]) Colegio De Pilotos Aviadores De Mexico, A.C. Comair, Inc. Northwest Airlines General Aviation, Randall S. Mainquist Parker Aerospace ([email protected]) Continental Airlines Government Air Safety, Willaim L. McNease Continental Express Phoenix International, Inc. ([email protected]) COPAC/Colegio Oficial de Pilotos de la Aviacion Comercial Pratt & Whitney Human Factors, Richard Stone Cranfield Safety & Accident Investigation Centre Qantas Airways Limited ([email protected]) Curt Lewis & Associates, LLC Qatar Airways Investigators Training & Education, DCI/Branch AIRCO Qwila Air (Pty), Ltd. Graham R. Braithwaite Defence Science and Technology Organization (DSTO) Raytheon Company ([email protected]) Delta Air Lines, Inc. Republic of Singapore Air Force Unmanned Aerial Systems, Tom Farrier Directorate of Aircraft Accident Investigations— Rolls-Royce, PLC ([email protected]) Namibia Royal Netherlands Air Force Directorate of Flight Safety (Canadian Forces) Royal New Zealand Air Force Directorate of Flying Safety—ADF RTI Group, LLC CORPORATE MEMBERS Dombroff Gilmore Jaques & French P.C. Sandia National Laboratories Dutch Airline Pilots Association SAS Braathens AAIU Ministry of Transport Bulgaria Dutch Transport Safety Board Saudi Arabian Airlines Accident Investigation Board, Finland EL AL Israel Airlines SICOFAA/SPS Accident Investigation Board/Norway Embraer-Empresa Brasileira de Aeronautica S.A. Sikorsky Aircraft Corporation Accident Investigation & Prevention Bureau Embry-Riddle Aeronautical University Skyservice Airlines, Ltd. Administration des Enquêtes Techniques Emirates Airline Singapore Airlines, Ltd. (Luxembourg) Era Aviation, Inc. SNECMA Moteurs Aeronautical & Maritime Research Laboratory European Aviation Safety Agency South African Airways AeroVeritas Aviation Safety Consulting, Ltd. EVA Airways Corporation South African Civil Aviation Authority Aerovias De Mexico, S.A.De C.V. Exponent, Inc. Southern California Safety Institute Air Accident Investigation Bureau of Singapore Federal Aviation Administration Southwest Airlines Company Air Accident Investigation Unit—Ireland Finnair Oyj Southwest Airlines Pilots’ Association Air Accidents Investigation Branch—U.K. Finnish Military Aviation Authority Star Navigation Systems Group, Ltd. Air Astana (Kazakhstan) Flight Attendant Training Institute at Melville College State of Israel Air Canada Pilots Association Flight Safety Foundation Transport Canada Air Line Pilots Association Flight Safety Foundation—Taiwan Transportation Safety Board of Canada Air New Zealand, Ltd. Galaxy Scientific Corporation U.K. Civil Aviation Authority Airbus S.A.S. General Aviation Manufacturers Association UND Aerospace Airclaims Limited GE Transportation/Aircraft Engines University of NSW Aviation Aircraft Accident Investigation Bureau—Switzerland Global Aerospace, Inc. University of Southern California Aircraft Mechanics Fraternal Association Gulf Flight Safety Committee, Azaiba, Oman Volvo Aero Corporation Aircraft & Railway Accident Investigation Commission Hall & Associates, LLC WestJet ◆ Airservices Australia Hellenic Air Accident Investigation AirTran Airways & Aviation Safety Board Airways New Zealand Honeywell Alaska Airlines Hong Kong Airline Pilots Association Alitalia Airlines—Flight Safety Dept. Hong Kong Civil Aviation Department Allianz Aviation Managers, LLC, USA IFALPA All Nippon Airways Company Limited Independent Pilots Association

January–March 2010 ISASI Forum • 31 Non-Profit ISASI Org. U.S. Postage 107 E. Holly Ave., Suite 11 PAID Sterling, VA 20164-5405 USA Permit #273 CHANGE SERVICE REQUESTED ANNAPOLIS, MD

WHO’S WHO Australian and International Pilots Association (Who’s Who is a brief profile prepared active stake in the Australian aviation eration of the aircraft, are secondary to by the represented ISASI corporate industry by participating in a wide range the pilot’s primary mission of safety. member organization to enable a more of government, legislative, and regula- AIPA, through its Safety and Tech- thorough understanding of the organi- tory inquiries and development pro- nical Subcommittee, is committed to zation’s role and functions.—Editor) cesses. Internationally, AIPA members protecting and advancing aviation safety are recognized as being among the most standards and operations and ensuring stablished in 1981, the Australian experienced flight crews in the world, and that the views of Australia’s professional and International Pilots Association AIPA is an active member of the global airline pilots are considered in important E(AIPA) is the professional associa- pilot body, the International Federation of safety and technical matters. Recently, tion representing pilots employed by Air Line Pilots’ Associations (IFALPA). the Association has worked with the Qantas Airways Limited and its wholly Civil Aviation owned subsidiaries in domestic and Safety Authority international airline operations. to draft regula- AIPA represents more than 2,600 air- tions in relation line transport-category flight crews and to multicrew pilot is the largest professional body of license, aeronauti- airline pilots in Australia. AIPA’s cal information membership comprises training cap- A number of AIPA pilots hold IFALPA publications, unmanned aerial systems, tains, captains, first officers, and second senior executive positions. IFALPA links fatigue risk management systems, and officers flying aircraft ranging from upward to the International­ Civil Avia- alcohol and other drugs. AIPA is also regional turboprops to 569-ton Airbus tion Organization (ICAO). International committed to consulting with the De- A380s. Its offices are located in Sydney airspace, airports, accident investigation, partment of Infrastructure, Transport, and Melbourne, Australia. and flight crew licensing are among some Regional Development, and local gov- The Association considers flight crews of the matters of interest that are dealt ernment on aviation security matters. as an essential part of a quality-control with through ICAO. Some recent achievements include the process that ensures safety remains at Both the Australian civil aviation signing of a memorandum of under- the center of aviation decision-making— legislation and Chicago Convention 1944 standing with the Australian Transport an independent role that AIPA believes standards clearly define the pilot-in- Safety Bureau in April 2009 to test an to be increasingly important within command’s role as being responsible expertise sharing arrangement in the liberalized aviation settings. for the safe conduct of a flight. All other interests of improving aviation safety In assuming this role, AIPA takes an considerations, such as the efficient op- through accident investigation. ◆

32 • ISASI Forum January–March 2010