Ask the pilot

Is pilot fatigue a menace to airline safety?

Published April 17, 2009 10:19AM (EDT)

This is your captain sleeping ... er, speaking. Sorry, it's been a long day.

Passengers won't be happy to hear about it, but crew fatigue has long been a serious issue. In February 2008, both pilots of a Go Airlines regional jet fell asleep over Hawaii, missing nearly a dozen air traffic control radio calls and overshooting their destination by 15 miles. Last June, the same thing happened on board an Air India jet headed from Jaipur to Mumbai. The plane continued past Mumbai for more than 300 miles before the crew finally woke up. The National Transportation Safety Board has cited fatigue as a contributing cause in several accidents, including the 1999 crash of American Airlines Flight 1420, at Little Rock, Ark. The MD-80 slid off a runway during a thunderstorm, killing 11 people, including the captain, who'd been on duty for more than 13 hours. At least two crashes involving cargo jets -- one at Guantánamo Bay, Cuba, the other in Kansas City -- have been blamed more directly on pilot tiredness.

The airlines and the Federal Aviation Administration are very resistant to the tightening of flight and duty time regulations. Some regulatory loopholes have been closed in the past few years, but even these small changes faced vociferous opposition by carriers and their lobbyists. During federal hearings in August 1999, on the heels of the Little Rock crash, John Meenan, vice president of the Air Transport Association, said to Congress, "There has never been a scheduled commercial airline accident attributed to pilot fatigue -- not one, not ever." That had to be one of the most jaw-droppingly disingenuous comments I've ever heard from an airline industry lobbyist. That there aren't more fatigue-related incidents is testament to the good job that pilots do under tough conditions, not a justification for legally sanctioned somnambulism on the flight deck.

Ask yourself this: Whom would you prefer at the controls of your plane on a stormy night, a pilot who smoked a joint three days ago, or one who had six hours of sleep prior to a 13-hour workday in which he's performed half a dozen takeoffs and landings? The first pilot has indulged in a career-ending toke; the second is in full compliance with the rules. I have to assume that the FAA realizes the foolery of such enforcement policies, but it nonetheless chooses to put its resources into drug testing and other politically expedient issues. Meanwhile it procrastinates, performing study after study and poring over data from NASA circadian rhythm experiments in an attempt to answer one of the world's most perplexing questions: Is exhaustion a detriment to job performance?

And frustratingly, most of the agency's focus has been on long-haul flying. The circadian-scrambling effects of a 16-, 17-, even 18-hour nonstop are indeed of concern. But it's also true that long-haul fatigue is comparatively easy to manage. These flights carry augmented crews and have spacious on-board rest facilities. Layovers are a minimum of 48 hours, typically at luxurious five-star hotels. The more serious problem is at the other end of the spectrum: short-haul regional flying.

Regional pilots fly punishing schedules, operating multiple legs in and out of busy airports, often in the worst weather, with short layovers at dodgy airport motels. Cockpit time doesn't present the toughest challenges, however. The real menaces are the long stretches of duty time, and the fact that legal layovers can be painfully short.

On a given workday, a pilot might log only two hours on the flight deck. Sounds like an easy assignment, except when those two hours come at either end of a 14-hour duty stretch that began at 5 a.m., most of which was spent waiting out weather delays and killing time in the terminal. Or, a pilot may have packed eight full hours of flying, making numerous takeoffs and landings, into that same span. Up to 16-hour shifts are possible, containing as many as eight hours aloft.

Now imagine multiple days like that, back to back. In FAA-speak, the layover buffers between duty spells are known as "rest periods," and they are subject to adjustment based on the number of hours a pilot has flown or been on duty. If a crew signs off in Chicago at 9 p.m., and signs on again at 6 a.m., that's a nine-hour rest period. Legal minimum rest can be as short as eight hours. That doesn't sound terribly abusive, until you start subtracting the time spent doing paperwork, waiting for the hotel van, driving to and from the airport, scrounging for food and so on.

A pilot is considered off duty anywhere from 15 to 30 minutes after his final flight of the day shuts down at the gate. With paperwork and other duties to attend to, his rest clock often begins ticking while he is still at the airport -- sometimes still on the plane! And, the next morning, it ends not in the hotel lobby, but back at the airport at the moment of sign-in. Once you account for the time waiting for pickups, the drives to and from the hotel, having to hunt down food, etc., what exists on paper as a 10-hour layover might include only five or six hours of sleep.

The ultimate horror is something called a "continuous duty" or "stand-up" layover. A crew signs on at, say, 9 p.m. and flies a single short leg, arriving at 11 p.m. The next leg isn't scheduled until 7 o'clock in the morning. The crew "stands up" until then, technically on duty the entire night, either trying to catch a nap at the motel, or slouching around the airport crew room. My copilot and I once spent a stand-up overnight dozing in the aisle of our Dash-8 turboprop on the tarmac in Bangor, Maine.

As I've recommended in this space before, the most productive step that regulators can take is eliminating transit time from what it considers "rest." In fairness to a pilot and his passengers, the rest clock should not begin to tick until the minute he latches the door of his hotel room and should stop ticking no later than the minute he checks out. But I doubt this will ever happen. For one thing, it brings ambiguity into how long a layover will last, and that is difficult for airlines to work with.

I also advocate that pilots be allowed cockpit naps, as regulations permit in Canada and certain other countries. Such a proposal is seen as radioactive by the FAA, unfortunately, and is kept off the table.

Meanwhile, keep in mind that drowsiness and fatigue aren't necessarily the same thing. Ultimately, this isn't a problem of pilots sleeping when they ought to be awake, but rather one of pilots who are awake when they'd be better off sleeping. Boredom and a light workload make it comparatively easy to nod off during a long trip, regardless of how well-rested a pilot might be. From a passenger's point of view, you're much better off with a pilot who naps for a few minutes over the middle of the ocean than with an exhausted, overworked pilot facing a low-visibility approach on a stormy night. The aforementioned Go Airlines incident aside, it's difficult for a regional pilot to fall asleep on the job. When flights are short and task-saturated, there isn't enough time.

Take it from me, life is much better at the major airlines, where bargaining agreements and company rules typically go above and beyond the government's more skeletal regulations. I have flown regional routes, back-of-the-clock cargo, mainline domestic and, more recently, long-haul international. In terms of fatigue issues, that list is in descending order. The latter two are by far the easiest and most civilized. Sure it messes up your circadian cycles, but I'll take a 12-hour red-eye ocean crossing followed by 72 hours at the Marriott any day over having to wake up at 4 a.m., fly six legs in a turboprop, then suffer an eight-hour break at the Holiday Inn Express.

Still, the battle is never completely winnable. I am always at least somewhat tired at the end of any long-haul flight. (I'm often asked how I deal with the effects of jet lag. Actually, I no longer identify jet lag as such. There is, instead, a post-flight weariness that comes regardless of the number of time zones crossed, time of day or direction.) Of course, being tired after a flight is less important than being tired before one. The most helpful thing I can do is nothing more elaborate than getting as much rest as possible prior to departure. En route, I drink lots of water, enjoy the view, and get up frequently to stretch.

I also get a scheduled break. As many of you know, all commercial flights carry a minimum of two pilots -- a captain and a first officer. At my carrier, any flight over eight hours in planned duration brings along an extra first officer, designated as the "relief pilot." This allows for a rotating series of breaks, with each pilot spending roughly one-third of the flight relaxing or sleeping. (As for which of the two first officers gets the relief assignment, it usually comes down to recency -- that is, who is in greater need of a landing -- or else is determined by a coin flip.) Flights exceeding 12 hours carry both an extra first officer and an extra captain, allowing for even longer breaks.

Sometimes I sleep; sometimes I watch a movie or "Flight of the Conchords"; sometimes I work on columns like this one. As I apply the finishing touches to this story, I am at 34,000 feet over Brazil -- just approaching the Amazon city of Santarem, according to my video screen.

The luxuriousness of crew rest quarters varies among aircraft types. I fly a 767, which is generally too small for a dedicated bunk room or other self-contained quarters (a few 767s do have them; most do not), so instead we use a cordoned-off seat in business class, pictured here...

Wait, that's actually Singapore Airlines' new first class. Our premium seats are, let's just say, a little less posh.

On bigger planes, crews have the benefit of surprisingly comfortable quarters. They can be above deck, below deck or squirreled away somewhere in the main cabin, usually out of view. Here's the main bunk room on the new Boeing 777-LR, the world's longest-legged commercial aircraft. It's tucked into the forward ceiling, aft of the cockpit.

Singapore's 777s are outfitted like this, with rest facilities built into a removable lower-deck pod. The Airbus A340 uses this same concept, allowing for extra cargo space on shorter hauls. (Q: Where is the captain? A: He is sleeping in the cargo compartment.) Check out this Pullman-style configuration on an Etihad Airway 777-300.

Not all digs are so fancy. Here's a more Spartan, eight-bed unit in an older 747-400 operated by Northwest. If you're wondering why there are eight, it's because this particular area is used by the flight attendants. Ditto for this layout, seen on a KLM 747. F.A.s too are entitled to in-flight rest periods, though not always in beds like the ones shown; sometimes it's just a modified economy seat. Pilots and F.A.s usually have separate quarters; on the 747, the pilot bunks are in a small room on the upper deck.

The answer is no, so please don't ask.

And if now you'll please excuse me. That mysterious ding you hear midflight is often a wake-up call from the cockpit, to summon a slumbering (or typing) pilot back to work.

- - - - - - - - - - - - - - - - -

Do you have questions for Salon's aviation expert? Send them to AskThePilot and look for answers in a future column.


By Patrick Smith

Patrick Smith is an airline pilot.

MORE FROM Patrick Smith


Related Topics ------------------------------------------

Air Travel Ask The Pilot Business