1
Good day. I’m Rod Rogers from Embry-Riddle Aeronautical University in Florida.
An airplane upset is any unusual or unexpected attitude or kinetic energy level that threatens loss of control inflight and subsequent ground impact. LOC-I is the abbreviation for loss-of-control inflight.
LOC-I for the past decade has been the major cause of hull losses and passenger/crew fatalities in air-transport operations worldwide.
In response, many air-transport operators provide employee pilots with upset prevention and recovery training–UPRT for short. UPRT typically involves classroom academic training as well as simulator-based training. A very few companies offer on-aircraft training in an aerobatic airplane.
Despite this training, LOC-I accidents continue to occur, for example
· the crash of Colgan Air Flight 3407 near Buffalo, NY in February 2009 with 50 fatalities
· the crash of Air France Flight 447 in the Atlanticf Ocean in June of the same year with 228 fatalities.
These two preventable accidents and others like them support suspicions that air transport pilots may be ill-prepared to recover from an in-flight upset—in short, that UPRT may not be very effective.
One way to assess the effectiveness of UPRT is to query the pilots who receive it. What follows reports preliminary results of an ongoing online survey to determine air-transport pilot attitudes about UPRT. 106 self-selected pilots have responded to date.
The slides that follow are organized into three parts. The first part describes the survey.
The second reports and analyzes collected data. It has 4 subsections:
· Pilot demographics
· Hours and type of company-provided UPRT
· Pilot confidence in their own and their peers’ upset-maneuvering skills
· Pilot attitudes about the UPRT they receive
The third part draws conclusions and implications from survey data.
Much more extensive information on the survey is available in the conference paper.
The survey has 23 questions:
· 9 on Demographics
· 6 on the UPRT pilots received
· 8 on pilot attitudes about UPRT and whether it improves pilot maneuvering skills.
With a sample size of 100, assuming a target pilot population of 50,000, the survey’s margin of error computes to plus or minus 8% at a confidence level of 0.95.
These figures mean that the opinions respondent pilots expressed differ at most by 8% from opinions of the pilot population the survey targets, with a 95% probability that if the survey were repeated, responses would differ from the target population by no more than 8%.
The age of pilot respondents was distributed fairly evenly over the range 26-60 years, with fewer respondents younger than 26 or older than 60. Numbers at the top of graph bars—one is highlighted by an enclosing red circle—indicate the total number of respondents selecting that response.
22% of respondent pilots had served between 6 and 10 years as air transport pilots. 58% had served 11 years or more. One-third had over 20 years of experience.
Almost three-quarters of respondent pilots had between 5000 and 20,000 hours of total flight time. Average flight time was 9,612 hours.
Just over one-third of respondents received their initial pilot training at a degree-granting aviation university. 44% were trained at non-degree pilot schools or by individual flight instructors. 15% were trained by the military.
Two-thirds of the respondents were employed by a legacy or a regional airline, with nearly equal distribution between the two. One-seventh worked for cargo carriers and one-seventh for other type companies.
Recall that UPRT consists of academic training, simulator-based training, and more rarely on-aircraft training.
Initial training occurs when a pilot is first employed by an air carrier or is retrained to fly a new type of airplane. Recurrent training is periodic refresher training as mandated by regulatory authority.
During initial training, 18% of respondents received no academic UPRT. The figure is 29% if—as seems likely—pilots who were unsure received none as well. An additional 33% of all pilots received less than an hour of academic training. Only 13% received more than two hours.
During initial training, 8% of respondent pilots received no simulator-based UPRT—15% if those who were unsure received none. An additional 46% received less than an hour. Only one-fifth received more than an hour.
During recurrent training, 18% of respondent pilots received no academic UPRT—26% if those who were unsure received none. An addition 58% received less than one hour. Only 16% reported receiving more than one hour of academic UPRT.
During recurrent training, 9% of respondent pilots received no simulator-based UPRT—15% it those who were unsure received none. 70% received less than an hour of training. Only 15% reported receiving more than an hour of simulator-based UPRT.
On-aircraft training in an aerobatic airplane is considered an important component of UPRT by many authorities. However, only 4% of pilots worked for companies that provided such training. 23% of respondents had never been in control an aerobatic airplane—and presumably had not been upside down in an airplane even once.
Computers fly air-transport airplanes most of the time today. The result is what has been called pilot automation dependency—deteriorating pilot skills in hand-flying an airplane when automation fails. This issue received international media attention when Air France Flight 447 crashed in the Atlantic Ocean after automation failed and the flight crew was unable to hand-fly the airplane successfully. Everyone knows the joke about the ideal cockpit crew—a pilot and a dog: the pilot feeds the dog and the dog bites a pilot who touches the airplane’s controls.
Pilots were asked to agree or disagree with the statement
I am well-practiced and feel confident to hand-fly an air transport airplane when automation fails….
Respondent pilots expressed strong confidence in their ability to hand-fly an airplane: 84% agreed or strongly agreed with the statement.
Pilots were asked to respond to the statement
I am confident beyond doubt that I could safely recover from any upset I might encounter…
Two-thirds of all pilots expressed confidence in their ability to recover from an in-flight upset. However, one-third failed to do so.
Pilots were asked to respond to the statement
The typical U.S. air transport pilot is well prepared to recover an air transport airplane from an in-flight upset.
Pilots rated their peers far lower than they rated themselves: less than one-quarter of pilots agreed, and two-fifths disagreed.
How can one understand this disparity?
When I was young, I flew the Navy jets off aircraft carriers. Fatal shipboard crashes then were far more common than they are now. When a death occurred, fellow aviators assembled in squadron ready rooms to express reverence and share their sorrow
However, a second response quickly surfaced. As more information about the cause became available, one heard pilots explaining to each other how the accident might have been prevented. Had one of them been faced with the same circumstances, the accident likely would not have been fatal. There was always something a pilot in extremis could do to forestall death.
Their dubious attitude after a fatality was: “Other pilots may not always be able to deal with an in-flight threat, but I can.” This of course is the opinion expressed by survey respondents who doubted their peers’ ability to recovery from an upset but were confident in their own ability to do so.
Pilots were asked to respond to the statement
My company provides adequate training to prepare me to recover from an upset…
Response was mixed. 40% agreed; 42% disagreed.
Pilots were asked to respond to the statement
My company should provide its pilots more extensive and/or better upset-recovery training...
Lack of confidence in company-provided UPRT was quite strong: 55% percent agreed with the statement; only 15% disagreed.
To sum up, respondent pilots believed they should receive improved UPRT.
They reported low training hours. During initial training, over one-half received less than an hour of academic or simulator-based UPRT. During recurrent training, over three-quarters received less than an hour of either.
They expressed low confidence in the ability of peers to recover from a serious upset.
They felt less than confident that company-provided UPRT was adequate.
Finally, they believed that their companies should provide better UPRT.
If one credits respondent pilot opinions, the inescapable conclusion is that air-transport UPRT should be improved.
But are improvements likely to be forthcoming?
On the “yes” side, adverse media exposure to LOC-I accidents in the 1990s was partly responsible for the advent of air-transport UPRT. Should a proliferation of LOC-I accidents occur in the future, industry might respond with improved training or regulators might require it.
A second factor favoring improvement is the work of ICATEE, the International Committee for Aviation Training in Extended Envelopes. ICATEE is sanctioned by the Royal Aeronautical Society, London, UK. Committee deliberations are near a conclusion, and currently ICATEE leadership is working closely with ICAO to improve UPRT worldwide.
It is anticipated that ICATEE will recommend improvements to UPRT including an expanded, career-long UPRT training syllabus, enhancements to full flight simulators to support improved stall training, and on-aircraft training for UPRT training tasks that cannot be accomplished in existing or enhanced full flight simulators.
Other factors, however, weight against UPRT improvement.
First, air-transport profit margins are low, and line-pilot training is very expensive. Companies are disinclined to spend training dollars unless regulators compel them to do so.
In addition, cost-benefit arguments for improved UPRT may be less than convincing. In some views, automation is doing an acceptable job in preventing upsets. Flying today is very safe in first-world countries. There hasn’t been a fatal air-transport accident in the U.S. in over four years. Why train more for something that almost never happens? Is it cheaper to pay the cost of the rare LOC-I accident than train to prevent it?
A final deterrent to better UPRT, I think, is the willingness of the air-transport regulators and the flying public itself to tolerate LOC-I accidents without demanding improved training.
In Peter Breughel’s Landscape with the Fall of Icarus, the Greek mythic aviator Icarus, obscured in the lower right-hand portion of the painting, is crashing headfirst into water. Only his frail white legs are visible. A plowman and a shepherd proceed routinely in their tasks while a ship sails obliviously away from the youth’s death.
The speaker in W. H. Auden’s poem “Musee Des Beaux Arts” is in a museum looking at Breughel’s painting. What he says explains why the death of Icarus goes unnoticed. Does it also explain why pilots believe they won’t crash and why most of us turn the blind eye to the LOC-I threat?
About suffering they were never wrong,
The old Masters: how well they understood
Its human position: how it takes place
While someone else is eating or opening a window or just walking dully along…
In Breughel's Icarus, for instance: how everything turns away
Quite leisurely from the disaster; the ploughman may
Have heard the splash, the forsaken cry,
But for him it was not an important failure; the sun shone
As it had to on the white legs disappearing into the green
Water, and the expensive delicate ship that must have seen
Something amazing, a boy falling out of the sky,
Had somewhere to get to and sailed calmly on.
1