IE 11 is not supported. For an optimal experience visit our site on another browser.

Pilot error rate ‘significantly declined’

Pilot error is probably much less of a factor in mishaps involving U.S. airliners now than it was in the early 1980s, a new study suggests.
/ Source:

Pilot error is probably much less of a factor in mishaps involving U.S. airliners now than it was in the early 1980s, a new study suggests.

The study, by researchers at the Johns Hopkins Bloomberg School of Public Health, discovered that the number of U.S. airline accidents due to pilot error "significantly declined" between 1983 and 2002.

Although the overall rate of U.S. airline accidents remained stable throughout the period, the proportion of mishaps involving pilot error decreased 40 percent, they found.

"A 40 percent decline in pilot error-related mishaps is very impressive. Pilot error has long been considered the most prominent contributor to aviation crashes," said Susan Baker, the lead author of the study, published in the January 2008 edition of “Aviation, Space and Environmental Medicine.”

The researchers defined a mishap as being any U.S. airline safety event that the NTSB officially recorded as an accident, because it involved serious injury to one or more persons or significant damage to an airliner. The airlines included were any that were defined as scheduled or unscheduled U.S. air carriers under Part 121 of the Federal Aviation Regulations.

Although the long-term study didn't take into account any NTSB accident data after 2002, it's likely the reduction in pilot error that the researchers found has continued to the present day — even though the National Transportation Safety Board (NTSB) found pilot error was primarily to blame for last year's Comair crash at Lexington, Ky., in which 49 people died.

Cockpit resource management a factor
Baker attributes the significant decline in pilot-error-related accidents to improvements in pilot training, flight deck technology and the development of cockpit resource management (CRM) techniques. CRM is a crew-coordination discipline that came into being in the 1980s and the FAA, the NTSB and all U.S. and major international airlines emphasize it strongly today.

"We saw a reduction in pilot error crashes where crew interaction was a factor. That, and weather-related decisions," said Baker. "Trends indicate that great progress has been made to improve the decision-making of pilots and coordination between the aircraft's crew members."

Significant improvements in flight deck technology between 1983 and 2002 particularly contributed to the improvement in weather-related decision-making, she added.

The researchers analyzed 558 separate accidents. "A lot were not airplane crashes," said Baker. "Maybe a quarter involved turbulence." Many others involved accidents while aircraft were sitting at the gate or being pushed back for departure.

Baker and her colleagues studied the circumstances of pilot error, which they characterized as carelessness on the part of the pilot and other flight crew; flawed decision-making; mishandling of the aircraft; or poor crew interaction.

The study's key findings
Their key findings were that:

  • Mishaps related to bad weather, the most common decision-making error, fell by 76 percent.
  • Accidents caused by mishandling wind or runway conditions dropped 78 percent. "I would think training would have a lot to do with that," said Baker. "But it's not just training. I think a lot of it may be technology."
  • Mishaps caused by poor crew interaction declined 68 percent, from 2.8 to 0.9 per 10 million departures. "I would certainly attribute that to cockpit resource management," she said.
  • Pilot error occurred most often during taxiing, take-off, final approach and landing — which are the phases of any flight that most require flight-crew concentration.
  • However, the pilot-error-related accident rate during take-off fell by 70 percent during the 20-year period of the study.
  • The mishap rate increased most when aircraft were being pushed back from the gate or standing still — but pilot error happened least often in these accidents.

Overall mishap rate didn't decline
While the rate of pilot error-related mishaps declined 40 percent from 1983 to 2002, the overall mishap rate didn’t decline at all during the 20 years, said Baker.

Some accidents involved errors made by air traffic control or ground crews. The Johns Hopkins study found that mishaps while aircraft were motionless on the ground or being pushed back from the gate more than doubled, from a rate of 2.5 to 6 mishaps in every 10 million departures.

"More likely these have been caused by errors while driving ground vehicles or by ground crews during push-back," said Baker. "The increase in mishaps while aircraft are not moving may require special attention. We need to be taking a very careful look" at these kinds of accidents.

Greater ramp congestion
The researchers think the rate of ground mishaps may have increased because airport ramps have become more congested.

In analyzing the NTSB data for the first five years of the study, from 1983 through 1987, the researchers "saw nothing involving push-back," said Baker.

The lack of data probably wasn't due to the NTSB not recording ground-crew accidents at the time: A Johns Hopkins researcher performing a parallel study on ground-crew injuries from 1983 onwards didn't find a significant increase in the injury rate after 1987.

In recent years, the Airports Council International and the International Air Transport Association have pinpointed inadequate training of drivers of airport ramp vehicles as a particular problem.

However, U.S. airline safety trends are encouraging for travelers. NTSB data show that the overall rate of fatal U.S. airline crashes fell dramatically in the 20 years from 1987 to 2006, Baker noted.

In the decade from 1987 to 1996, the fatal accident rate was six accidents per 10 million departures. However, in the decade from 1997 to 2006, the rate fell to two fatal accidents per 10 million departures.

The new Johns Hopkins pilot-error study formed part of a larger project that — funded in part by the National Institute on Aging — focused on age-associated changes in commercial pilots. A study that formed an earlier part of the project found pilot error did not increase with pilot age — an interesting finding, given the recent passing of legislation by the U.S. Congress to increase the retirement age of commercial pilots from 60 to 65.