Monday, February 08, 2010

Lessons from Flight 3407: Pay Attention

Last Tuesday, Feb. 2, the U. S. National Transportation Safety Board (NTSB) released a synopsis of its final report on the crash of Flight 3407 outside Buffalo, New York a year ago (Feb. 12, 2009, to be exact). All aboard the Colgan Air commuter flight from Newark to Buffalo were killed, as well as one person on the ground, raising the total number of fatalities to fifty. In my initial blog on this crash, I reported that there might have been a problem with the deicing machinery on the plane. But an NTSB hearing last May revealed that evidence had emerged of significant pilot error. At the time, there was much consternation in the press about the fact that the co-pilot, 24-year-old Rebecca Shaw, was earning only about $17,000 annually and was living with her parents to save expenses. The NTSB found that both Shaw and pilot Marvin Renslow acted unprofessionally during the flight, tolerating distractions from "personal portable electronic devices," chatting about irrelevant matters during the final approach, and reacting incorrectly to emergency warnings. There were three separate occasions at which Renslow could have corrected the deteriorating situation above Buffalo that night, but the NTSB found that his reactions were characterized by "startle and confusion," to quote from the official synopsis. He had a spotty training record, and this probably contributed to the fact that when the "stick shaker" (a device designed to warn pilots of an impending stall) was activated, he did not automatically push the stick forward as his training should have prompted him to, but instead pulled back, compounding the situation and ultimately causing him to lose control of the aircraft.

Anyone who has studied safety and engineering-related accidents knows that beneath every headline-grabbing tragedy, there are usually a lot more minor incidents, near misses, and recoverable errors that for one reason or another did not go all the way to sudden death for lots of people. The airline industry as a whole has been under financial pressure for many years, and so you have the situation where young, inexperienced, poorly paid pilots are making up an increasing fraction of the airline-pilot staff. For every Flight 3407, there are probably a dozen or more similar incidents that were caught at the last minute.

In its collective wisdom, the NTSB implicitly recognized this more systemic problem in its list of twenty-five recommendations to the Federal Aviation Administration. (One can question the bureaucratic wisdom of having one agency investigate accidents and recommending corrective action to another, but that is a discussion for another time.) The NTSB calls for better training, more careful scrutiny of pilots whose test scores show problems (or who flunk tests altogether), and more of what they call "sterile cockpit discipline."

The ability to direct one's attention is one of the supreme gifts of humanity, but if a person lacks the training or inclination to do so, all the skills in the world are useless. Both Renslow and Shaw were fatigued before they embarked on Flight 3407 (Renslow had slept in the airline's pilot lounge in violation of regulations), but fatigue was no excuse for inattention. Everyone who drives a car knows how important it is to avoid being distracted by cell-phone conversations, GPS readouts, or any of the other amusements and information sources we tend to surround ourselves with. This goes double with the much more complex task of flying a plane, for which sterile cockpit discipline, meaning doing your job and nothing else that will distract you, is a necessary prerequisite. Unfortunately, Shaw and Renslow did not learn that lesson well enough and paid for it with their lives.

The British comic John Cleese, one of the original Monty Pythons, once satirized overly chatty airline pilots in a hilarious sketch. Clearly bored out of his mind by the monotony of his flying job, he amused himself by getting on the intercom and making announcements such as, "Hello, this is the Captain speaking. There is absolutely no cause for concern," then remarking to his copilot, "There, that'll get them thinking." The sketch degenerates to the point where the passengers all jump out of the plane, at which point Cleese looks out the window and comments reflectively, "You know, I wouldn't be surprised if there was some trouble about this."

Analyzing humor is not a funny business, but neither is investigating air crashes. The sketch gained much of its energy from the fact that in the public mind, airline pilots are largely a stereotype of the responsible professional. The picture of a pilot intentionally twitting the passengers and goofing off is funny (in a TV sketch, anyway) because it is so incongruous with the profession's standard public image.

But professionalism has to be achieved one pilot at a time. The NTSB, in its investigation of Flight 3407, has spotted what it fears to be a dangerous trend of unprofessionalism in the cockpit. While its recommendations do not carry the force of law, I hope that the FAA, whose dictates are law, will take them seriously. Professionalism, smooth flights, and expert handling of dangerous situations are what the flying public expects from pilots. Even if we all have to pay a little more, it's worth it to know that the people into whose hands we place our lives every time we step on an airplane are worthy of that trust.

Sources: The synopsis of the NTSB final report on the crash of Flight 3407 can be found at Cleese's "Airline Pilots Sketch" is currently online at My previous blogs on this incident can be found in the entries for Feb. 12 and May 18, 2009.

1 comment:

  1. I think the NTSB is correct in some ways but has missed actual causes by a mile.

    Sterile cockpit discipline is essential, but as 757/767 pilot Kent Wien points out, the Colgan crew did not make any unnecessary chatter for more than 2 minutes before the Q400's airspeed dropped to a stall.

    So they were distracted, all right, but not by chatter. Kent wishes the NTSB had taken greater care to factor in complications arising from fatigue, an opinion which I think has some merit.

    One of Kent's commenters also notes that the aircrew used automation (i.e. autopilot) even though company policy was to hand fly in severe icing conditions.

    If the aircrew had been hand flying the aircraft prior to the accident, they would have noticed the speed falloff earlier--because it would be harder to maintain attitude with that power setting.

    So what would cause an aircrew to over-utilise automation when company recommends otherwise? Fatigue? Sure. Inexperience or lack of confidence in their ability to manually fly the aircraft in those weather conditions? Could be that, too.

    The NTSB's conclusions have some merit, but they are, I think, far off the mark.