A study that came over the transom recently speaks to improving safety on airline flight decks. Unfortunately another study, or at least the beginnings of one, that could add substantially to what is known about aviation safety has been disowned by its government sponsor, diminishing its impact.
The first study, an analysis of 558 air carrier mishaps from 1983 to 2002, revealed a 40-percent decline in mishaps involving pilot error, from 14.2 per 10 million flights to 8.5. Conducted by researchers at Johns Hopkins Bloomberg School of Public Health, the analysis drew from National Transportation Safety Board investigation reports, reviewed and coded by a commercial pilot. The term "mishaps" is used to include injury-producing events that do not involve crashes, resulting from turbulence or events on the ground while the aircraft is not moving.
Pilot-error mishaps as a proportion of the total declined from 42 percent in 1983-1987 to 25 percent in 1998-2002. Nevertheless, the overall rate of mishaps from all causes, including bad weather, clear air turbulence and mechanical failure, remained stable over the 20-year period, averaging 33 per 10 million flights. The progress among pilots was offset by an increase in events, not attributed to pilots, that occurred while aircraft were stationary or during push-back (likely involving ground personnel and air-traffic control), a trend deserving special attention, the researchers advise.
But it is the airborne side that concerns us. Pilot error was broken out into categories of carelessness, poor decisions, mishandling aircraft kinetics, mishandling wind or runway conditions, and poor crew interaction. Error related to flawed decision-making declined from 6.2 to 1.8 per 10 million flights, a 71-percent reduction. The most common type of flawed decisions, those related to weather, declined 76 percent, from 4.6 to 1.1 per 10 million flights.
The findings of the study point to the contribution of navigation, display and weather reporting and sensing systems to improved situational awareness. Researchers mentioned displays in particular.
"The decreasing trend in pilot-error mishaps probably reflects, in part, advances in technology such as cockpit displays that permit pilots to avoid threatening weather and to determine their exact location in relation to airports, runways, etc.," the study states.
"Emphasis in recent years on crew resource management has likely been another important factor in the 68 percent reduction in mishaps involving poor crew coordination."
All good information. But additional, comprehensive insight into what pilots themselves experience has been jettisoned by NASA, otherwise a trusted font of aeronautical research. You’ll recall the space agency was in the news last October for withholding results of its National Aviation Operational Monitoring Service (NAOMS), a compilation of some 29,000 telephone surveys of airline and GA pilots on safety-related issues. Concern over pilot confidentiality and upholding public confidence in the airline industry were cited as reasons. The multi-year project, conducted for NASA by Battelle Memorial Institute, cost taxpayers a reported $11.3 million.
Responding to a Freedom of Information Act request and congressional pressure, NASA finally did release information on the last day of 2007 — some 16,000 pages of raw, redacted data with no analysis. Other than fragmented references to fatigue, difficult communications and other issues, it’s hard to decipher.
In what reads, according to the transcript, like a combative teleconference with reporters explaining the release, Administrator Michael Griffin and Bryan D. O’Connor, NASA chief of Safety and Mission Assurance, defended the data dump, saying the research was never properly peer-reviewed, nor the findings validated. Besides, they argued, the whole point of the exercise was to develop a methodology for collecting safety data, not to crunch the actual data.
O’Connor said the information would be referred to the National Academy of Sciences to assess the methodology, though any hoped-for trend analysis was "not in the cards."
Put me in the camp of the reporter who said the following: "If there was never an intent to analyze the results, then I don’t really understand why the study was undertaken."