Commentary: Is Fatigue Fatal?

THE RULES THAT SPECIFY HOW long an airline pilot can be on duty before resting and how long that rest must be fill little more than a page in the Federal Aviation Regulations.The Air Line Pilots Association document “Guide to Flight Time Limitations and Rest Requirements,” detailing how those rules work, is over 19 single-spaced pages long and bears the warning “This booklet cannot answer all questions concerning the application of the rules….”

Until terrorist attacks were made against the United States on September 11, the confused and inadequate state of airline duty-time regulations was thought by some to be the single greatest threat to U.S. airline safety. In fact, just one day earlier, on September 10, the major airline pilots union issued a press release headlined “ALPA Irate Over Court Stay of Pilot Fatigue Rule.” Several pilot unions had charged that the rules were poorly written and unclear, were being fiddled with by the airlines to make pilots work more hours, and were out of date. But the Federal Aviation Administration had convinced a U.S. Court of Appeals that the rules, FAR Part 121.147, were adequate, and although they needed some fine-tuning, there was no need for haste.

A bit of haste might have averted a June 1999 American Airlines accident in Little Rock, Arkansas, when a flight crew, obviously fatigued at the end of an overlong but legal duty day, did a lousy job of landing their MD-82 during a gusty thunderstorm. The captain and 10 passengers were killed when the airplane went off the end of the runway. Yet the FAA continues to argue that “the vast majority of pilots are receiving the amount of rest required by the FAA’s rule.” (Hard to imagine the Feds defending other regulations this way, saying that “the vast majority” of turbine engines are well built or that “the vast majority” of pilots have valid medical certificates.)

Current FAA rules on airline pilot hours limit flight time to eight hours (this can be exceeded “due to circumstances beyond the control of the carrier,” such as inclement weather) and duty time to 16 hours. Thus, if a pilot takes off on an eight-hour flight from Point A to Point B at 0900 and arrives at the airport at 0700 to check the weather and go through his preflight routine, his total duty time is 10 hours. If, however, the pilot shows up at 0700 for the same flight but then waits an additional three hours for a mechanical problem to be resolved before takeoff, the flying time to Point B is the same but his duty time has now ballooned to a long but still legal 13 hours. The Air Line Pilots Association would like to limit duty time to 12 hours, while the Air Transport Association, which represents airline management, would like duty time to be extended past 16 hours in the event of problems with weather, aircraft maintenance, and air traffic control.

For most of the country’s workers, the clock starts when they arrive at work and stops when they leave. Should pilots operate any differently? As one former accident investigator put it, “Awake is awake. Why should there be any difference between time spent briefing, checking weather, preflighting, awaiting equipment, and sitting out delays, and time spent flying?”

Airline pilots may indeed be receiving the rest required by the rules, but the FAA hasn’t the faintest idea whether pilots are in fact rested. There is virtually no hard data on the effectiveness of pilot rest periods. Which suggests that rather than saying “Airliners aren’t falling out of the sky with pilots asleep at the switch, so everything must be okay,” the FAA should liberalize the regulations to allow flight crews more sleep and then put researchers to work studying rest patterns. If it’s demonstrated that pilots are getting more rest than they need, then and only then should the regs be ratcheted back up.

A pilot’s rest might come at an odd, hard-to-sleep time—two in the afternoon until 11 p.m., say—and the nine hours of “rest” might well include two hours of round-trip travel, check-in/out between airport and hotel, and time to eat a meal. Pilots also sometimes contribute to their own fatigue. A surprising number commute long distances to work, stretching duty days to extreme lengths. Former NTSB accident investigator Rudy Kapustin recalls working a Pan Am crash in San Jose, Costa Rica: “The pilot originated the flight in Florida, but his home was somewhere in the Northwest,” he recalls. “So he had commuted a six-hour flight, and gotten up two hours before that, before he’d even started his working day. He got all screwed up on the approach and hit short of the runway. He was completely out of it.”

Some pilots also bid for a month’s worth of trips in grueling blocks, so they can minimize trips to their base and get all their flying done in closely scheduled chunks. That routine doesn’t help their sleep cycles. And senior captains often bid for the long international routes. “I spent six months doing nothing but flying from the West Coast to the Far East,” says one middle-aged captain. “The average legs were 12 or 13 hours long, and it’s so boring that your alertness is slowly eroded away. By the time you actually shoot the approach, you’re not very sharp despite the fact that you’ve had crew rest. By the time you reach Japan, you’re wide awake yet you can feel yourself fatigued. You try to go to bed, you can’t sleep, and by the time you’re flying back eastbound, your body is trying to tip over.”

Fatigue is an insidious thing. If it weren’t, people wouldn’t fall asleep while driving. “Those guys who crashed at Little Rock should have gotten off in Dallas [at the end of the previous leg] and walked away from the airplane,” opines a 767 captain. “But they chose to press on, probably feeling they were okay. That’s the thing about being tired—you really don’t feel it yet, and your decision-making capability is diminished.”

The NTSB listed crew fatigue as a major factor in the Little Rock crash, but how many other accidents have resulted at least in part from fatigue? Flaps mis-set, navaids mistaken, procedures confused, emergencies mishandled, approaches botched, ATC clearances misheard…. Were these all simply the result of “pilot error,” or did pilot fatigue play a part? We’ll never know, for sleep-deprivation accidents rarely leave clues, and if pilots survive, how many are likely to admit: “I just destroyed a $40 million airplane, killed dozens of people, ended my career, and opened myself up to enormous liability because I violated the FARs and I flew when I was too tired”?

If a pilot flies when he is fatigued, he is in violation of FAR 91.13, which states: “No person may operate an aircraft in a careless or reckless manner so as to endanger the life or property of another.” The FAA considers that this one rule will cover everything from sleepy pilots to those who have sprained an ankle playing softball, overdosed on their meds, drunk too much, suffered an emotional upheaval, or gone through anything else that would make them unsafe in the cockpit.

Some airlines say that their policy permits pilots to abandon a flight due to fatigue with a guarantee of no recriminations. Some pilots say, “Yeah, right. Try it and see what happens.” Yet others maintain that the option is there at least in part so that those who apportion blame after an accident can say, “It was her fault; she should have taken herself out of the cockpit if she was that tired.”

“On my last month at American, I ran into one of those creeping-delay situations,” recalls retired American Airlines captain Robert Besco, who holds a Ph.D. in industrial psychology and has written extensively about aviation safety. “They wanted us to take one more leg and I said no. The chief pilot got all over my case about it, the copilot was still on probation and wouldn’t stick his neck out and agree that we were dead tired, so we went ahead and flew from Oakland back to Dallas. We didn’t crash, so it must have been safe.” Unfortunately, the burden of adhering to the duty-time regulations falls largely upon pilots, who are asked to make go/no-go decisions just at the time when their judgment is likely to be compromised.

The Airline Transport Association avers that because pilot fatigue has not been proven to be a problem, more restrictive duty-time regulations would add unnecessary costs to airline operations. In an era of $1.98 airport security, $300 Blue Light Special round trips to Europe, cattle-car coach travel, and a near-total abandonment of customer service in what is, after all, simply a service industry, perhaps it’s time to pay some costs once thought unnecessary.