The NTSB cited three accidents and an incident involving regional airlines as the basis for a pair of recommendations to the FAA related to pilot fatigue last month. The Board called on the FAA to develop guidance for operators to establish “fatigue management systems” and methodology to assess their effectiveness, including their ability to improve sleep and alertness, mitigate performance errors and prevent incidents and accidents.
“The Safety Board is extremely concerned about the risk and the unnecessary danger that is caused by fatigue in aviation,” said NTSB chairman Mark Rosenker. “We have seen too many accidents and incidents where human fatigue is a cause or contributing factor.”
The most recent case cited by the NTSB involved a Mesa Airlines Bombardier CRJ200 that flew past its destination airport, General Lyman Field in Hilo, Hawaii, after its pilots fell asleep at the controls during a February 18 revenue flight. ATC repeatedly attempted to contact the crew for more than 18 minutes, as it flew over Maui, crossed the big island of Hawaii and headed southeast over the Pacific Ocean. The airplane traveled 26 nm beyond its intended destination airport before the flight crew responded.
Testimony during a June 10 NTSB hearing in Washington revealed that the crew
of the Mesa flight–operating as Go!, the Phoenix-based airline’s Hawaiian subsidiary– were working the third day of a trip schedule that involved repeated early start times and demanding sequences of short flight segments. The NTSB learned that after the incident, a doctor diagnosed the captain with severe obstructive sleep apnea, which left untreated results in reduced sleep quality, chronic fatigue and, in severe cases, cognitive impairment. Mesa fired both pilots.
Fatigue Training Needed
The Board also cited the Oct. 19, 2004, accident in Kirksville, Mo., in which a Corporate Airlines Jetstream 31 struck several trees on its final approach and crashed short of the airport, killing both pilots and 11 passengers. Two seriously injured passengers survived that crash. The accident took place at the end of a 14.5-hour duty day for which the pilots had reported early and during which they had conducted five previous landings in poor visibility. Following that accident, the Board recommended that the FAA require operators to incorporate information about fatigue in their training programs and modify and simplify flight-crew hours-of-service regulations.
Another case cited during the hearing involved the runway overrun of a Shuttle America Embraer E170 on Feb. 18, 2007, at Cleveland Hopkins International Airport. Investigators found that the captain of that flight suffered from insomnia, and that he hadn’t slept for 31 of the 32 hours leading up to the accident. The captain told investigators that he didn’t remove himself from duty or advise the company of his fatigue because he feared getting fired. As a result of that incident, the Safety Board recommended an industry effort to formulate and implement an attendance policy that would allow flight crewmembers to decline assignments or remove themselves from duty if they felt impaired by a lack of sleep.
The hearing also addressed the runway overrun involving a Pinnacle Airlines CRJ200 at Cherry Capital Airport in Traverse City, Mich., on April 12 last year. The report cites the pilots’ decision to land without performing a landing distance assessment, which company policy required because of runway contamination first reported by airport ground personnel and continuing reports of deteriorating weather and runway conditions throughout the approach. The poor decision-making likely reflected the effects of fatigue produced by a long, demanding duty day and, for the captain, the duties associated with check airman functions, according to the NTSB.
“Our recommendations are designed to reduce injuries and deaths and prevent accidents like this from occurring,” said Rosenker. “Piloting an aircraft should not be guesswork. There are rules and guidelines that need to be followed at all times and it is imperative that the Federal Aviation Administration enforce these recommendations.”
Testifying at the hearing, Dr. Jana Price of the Office of Research and Engineering recommended a “comprehensive approach” to mitigate the hazards of fatigue. Such “fatigue management systems” use so-called countermeasures or strategies generally instituted by the operator and involving shared responsibility by management and employees. They might include scheduling practices and policies that consider normal circadian rhythm variations, attendance policies that allow crews to call in fatigued without fear of reprisal, medical screening and treatment for sleep disorders, changes to the working or sleeping environment and policies related to napping and education for managers, schedulers and crews about the hazards of fatigue.
Previous efforts in other countries to institute fatigue management practices have yielded mixed results. In 1995 the Civil Aviation Authority of New Zealand revised its regulations to allow air carriers to comply with either standard flight and duty regulations or an alternative, CAA-approved fatigue management system. The revised rule requires that an operator establish a plan that addresses a long list of factors related to fatigue, including rest periods before flight, time zones, night operations and circadian variability.
A 2006 report suggested that the practices used by the operators engaged in alternate fatigue management programs proved no better or worse than those of the operators following the standard regulations, however. Ideally, a comprehensive system should have performed better than one that merely specifies maximum flight and duty hours, insisted Price, and the report’s author suggested that the regulator improve its guidance to operators on how to implement fatigue management systems.
More recently, the UK’s CAA granted EasyJet a temporary variance to flight and duty regulations under the provision that it would demonstrate that its schedule changes would not lead to increased crew fatigue. A subsequent study tracked performance and errors and found a “significant” improvement under the modified system, which it has since permanently adopted and expanded.