Sophos News

Single-digit data entry glitch led to plane tailstrike

A co-pilot’s fat-fingered entry into an iPad caused a plane to scrape its tail on the ground – a potentially deadly incident – during takeoff in August 2014, the Australian Transport Safety Bureau (ATSB) has concluded.

Actually, make that two data entry errors, given that the pilot also forgot to carry the “1” when he was jotting down his calculations for the plane’s takeoff weight on a notepad.

The ATSB released the results of its investigation on Monday.

The investigation found that the tailstrike was caused by two, independent data entry mistakes that both led to an identical, erroneous takeoff weight being input.

The captain’s mistake was that he recorded the zero fuel weight and fuel load on a notepad in order to come to the takeoff weight.

But while he was doing his calculation, he failed to carry the 1.

He reported the takeoff weight of 66,400 kg into his on-board performance tool to figure out the take-off speeds and engine setting, but in actuality, the plane was 10,000 kg heavier than that.

What are the chances that his co-pilot would make a second error and come up with the identically wrong takeoff weight?

That’s exactly what happened: the first officer calculated correctly to arrive at a takeoff weight of 76,400 kg, but that’s not what he input into his iPad.

Instead, he made what the ATSB called a “transposition error.”

The ATSB defined that term as being “when an individual inadvertently swaps two adjacent numbers or letters while speaking or writing down a value or word.”

In other words, the co-pilot fat-fingered the incorrect total weight of 66,400, chopping 10,000 kg off the correct weight.

Given that both the pilot and the first officer input matching takeoff weights, the crew didn’t pick up on the error when they compared the two figures.

Because of that 10,000 kg difference between the reported vs. the actual weight of the plane, settings for take-off speed and engine thrust were miscalculated, with the result that they were too low.

Without enough thrust and speed, the plane, a Boeing 737-838 (VH-VZR) operated by Qantas, overpitched and clipped the runway when it was rotated.

The plane was taking off from Sydney airport when it happened.

The tailstrike went undetected by the pilots, but one of the cabin crew in the back of the plane heard what they called a “squeak” during rotation.

The crew, suspecting a tailstrike, conducted a tailstrike checklist, checked various sensors, and contacted mission control.

Seeing no indication of a strike, they decided to continue flying to Darwin and landed there without any problem.

After landing, the captain noticed that some paint had been scraped off a protective tailskid, showing that the tail had indeed slightly clipped the ground during takeoff.

Tailstrikes can be extremely dangerous. They can damage planes and endanger lives.

In 1985, 15 crew members and 505 of the 509 passengers on board Japan Airlines Flight 123 died when a Boeing 747 suffered explosive decompression due to a panel having been damaged years earlier in a tailstrike incident and not having been correctly repaired.

None of this can be blamed on an iPad, mind you.

This was pure human error, not an iOS glitch.

Both of the pilots were well-rested, and both had over 10,000 hours of flying experience, so neither grogginess nor inexperience was at the heart of the incident, the ATSB said.

These type of errors happen even with experienced pilots, it said:

Data input errors can occur irrespective of pilot experience, operator, aircraft type, location or take-off performance calculation method.

Effective management and systems can “significantly reduce the risk of data input errors,” though, the ATSB says.

To that end, as a result of the lessons learned from the tailstrike, Qantas has tweaked its pre-flight procedure so that the pilot and co-pilot check their calculations against the plane’s reference manual.

Image of plane tail courtesy of Shutterstock.com