Culprit Emerges in Botched Take-Off: a Typo

By Brett Israel | December 21, 2009 4:48 pm

Thanks to the winter devastation wrought by this weekend’s storm, my weekend holiday travel plans were put on hold until, well, now.  So from 36,000 feet above the ground, courtesy of Delta’s free wifi (it’s the least they could do, seeing how they put me on hold all weekend with “Let it Snow” playing on a loop), I bring you a story of a flight canceled not by weather, but by a typo.

Back on March 20th at Melbourne Airport, a United Arab Emirates (UAE) aircraft’s tail made contact with the runway during take-off (known as a tail-strike), and the plane was having trouble taking off at all. In fact, the tail hit the ground three more times beyond the runway and the landing gear took out a strobe light and a localizer antennae. Through some slick piloting, the airplane’s captain was able to get off the ground, dump fuel, and return to the same airport.

The cause for the tail-strike? A number 3 where a number 2 was supposed to be, as reported by IEEE Spectrum:

The  Australian Transport Safety Bureau (ATSB) confirmed its preliminary findings that determined that “the pre-flight take-off performance calculations were based on an incorrect take-off weight that was inadvertently entered into the take-off performance software on a laptop computer used by the flight crew.”

The aircraft’s first officer typed in the aircraft’s weight as being 262.9 tonnes, while the actual weight was 362.9 tonnes. So yeah, he was a little off, which caused the bumpy take-off. The flight crew members found responsible resigned shortly after the event.

Apparently my flight crew at LGA was paying closer attention.

Related Content:
DISCOVER: The Physics of… Airplanes
Discoblog: Airlines, Desperate to Lower Fuel Costs, Turn to GPS
Discoblog: So Is It a Disease? United Airlines to Double-Charge Obese Fliers

CATEGORIZED UNDER: Technology Attacks!
MORE ABOUT: airlines, travel
  • Emily Lakdawalla

    Wow. Now that’s some badly written software.

  • Imipak

    RISKS Digest is full of hair-raising (and/or expensive, amusing, highly inconvenient etc) consequences of HCI issues like this (OK, mostly a lot less potentially lethal) — still, highly recommended for scuttlebutt about all sorts of computer/human fail.

  • Wilson

    I am really surprised (and, honestly, a bit scared) that there are no checks for this. I mean, I would expect the plane itself to have some idea of how heavy it is (based on how it it accelerates when power is applied, if nothing else, but the landing gear could very well report the weight it’s holding) and to react accordingly if improper take off procedures are taken…

  • John Hebbe

    If the error described here (transposing a 3 and a 2) occurred, it due to crew error; negligence. Any test of reasonability on the part of the Captain –the final authority- should have sent up a red flag. This statement is based upon my 51 years of flying, thirty four of these on commercial planes.
    The other suggestions having to do with weighing devices built into the landing gear or acceleration rates are not viable for many reasons.

  • Wayne

    If the crew used a Flight Management System (avionics equipment built-in to the aircraft) such a mistake could never happen. As it was, they used a laptop computer, since that’s cheaper than an FMS… but laptop software could get you killed. Avionics equipment are verified to a very high level. Laptop software isn’t.

  • James


    If you read the IEEE Spectrum link, it looks like the laptop was running software that was provided by Airbus.

    This was a reduced-thrust takeoff, apparently. My guess is for noise abatement.

  • Kaleberg

    Most big airlines have the weight and balance info entered into the gross weights and flight planning systems automatically. They also have dispatchers who are expected to give the loadings, settings and routings a glance before they are forwarded automatically to the flight computers. Small airlines often use laptop based flight planning systems, and they are pretty good, but the pilots are on their own, and all sorts of errors can creep in. Wayne is right though. A full featured FMS has a lot more scar tissue than the basic vendor supplied software which is built from the plane’s engineering specs. It’s fun reading the code sometimes, and talking to an old timer: “It adds 5,000 pounds.” “Huh? Oh, that’s that crash in Detroit in ’86. Wrong flap setting.”

    There are a lot of reasons for using reduced thrust. It is quieter, it uses less fuel, but more importantly, it puts less wear and tear on the engines. An airline can go from a 5,000 hour rebuild to a 7,000 hour rebuild on some models. Of course, most pilot’s unions require that the pilot be allowed to use full thrust at discretion. There are airlines which would rather risk the plane or lose a good pilot than do more frequent maintenance. For safe practice, some runways, like the old Hong Kong airport runways, required full thrust to keep the airplanes out of people’s apartments. That shows up in the code too. As I said, fun reading.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


Quirky, funny, and surprising science news from the edge of the known universe.

See More

Collapse bottom bar