Fatigue is insidious. As a person experiences fatigue they are introduced to random instability in cognitive performance. In layman's terms it means that moment by moment the mind's performance can change drastically and instantaneously. This waking brain instability leads to random attention lapses, errors and eventual inadvertent sleep periods or micro-naps. One moment every cognitive process is working fine and the next moment could find you in the grip of sudden unconsciousness. If you have ever woken up in front of the television wondering how much of the movie you missed, you understand how micro-naps sneak up on you. In short, there is very little warning to a fatigue induced attention lapse. This reliable randomness is what makes fatigue so dangerous. It takes little imagination to decide what the outcome would be of a micro-nap during a critical phase of flight.
“It is rare for multiple underlying failures to coincide, and thus most lapses of attention tend to pass without consequence, giving a false sense that fatigue may not be a critical risk factor. However, because lapses of attention may occur at any moment, more frequently with greater fatigue yet unpredictably in time, an accident with fatigue as one of the underlying causes can happen any day regardless of an operation’s prior safety record” (Satterfield & Van Dongen, 2013, pg. 123).” The previous quote is a harrowing truth. More so because it is difficult to predict accidents and incidents notably because of the random nature of fatigue induced attention lapses. It is, however, possible to reduce the risk of a fatigue-induced event. The randomness and unpredictability make preventative fatigue measures all the more important. Yet there is resistance to putting in place some practical preventative measures.
As insidious as fatigue is, so are our biases. One of the most common biases in business is confirmation bias, where information contrary to one's position is ignored. Sometimes this confirmation bias leads to system justification, where an organization anchors to the status quo which hampers progressive thinking. People and organizations do not typically make decisions consciously to hinder progress. Biases in business are born from profit driven thinking. In many cases these biases and heuristics, or rules of thumb, serve positive functions. In the cases of confirmation bias and system justification, though, one inadvertent consequence is the sacrifice of progressive thinking.
The aviation industry's treatment of operational fatigue is one example. Airlines and flight departments depend on hours of service regulations and sometimes little else as a mitigation to fatigue. Hours of service limitations are not enough because they disregard the sleep/wake history of the flight crews and their circadian rhythmicity. As a result, these duty time regulations may be too restrictive and inflexible and at other times overly permissive and unsafe (Van Dongen, 2012). The differences in duty time regulations between scheduled service, on-demand carriers and Part 91 flight departments supports this assertion.
Following duty time regulations, some carriers turn to schedule-based mathematical models to minimize the fatigue impact on the overall operation. Most carriers and business aviation operations stop there, thinking they are exceeding the regulatory guidance. Other operators rely on subjective assessments in post-flight to report a fatigued crew or mitigate the following day's schedule. The problem with these risk assessments is that they rely on subjective determination by the pilots. Reliance on subjective self-determination of fatigue state and cognitive ability is misplaced as studies show that people exposed to cumulative sleep debt and/or acute sleep deprivation do not reliably assess their own level of fatigue (Van Dongen, 2012). In other words, type-A pilots are really bad at understanding their limitations when it comes to fatigue. Nobody wants to be the pilot who says, “No, I'm too tired.”
Even schedule-based mathematical models and subjective assessments put together do not complete the puzzle. Both are useful tools to avoid the most egregious fatigue-inducements and are necessary parts of a comprehensive fatigue risk management system. Completing the puzzle is as simple as adding an efficient predictive tool that accounts for the circadian disruption and daily sleep history of the flight crew that establishes a worst-case scenario for each flight duty period. This human factors data pool can then be analyzed against established thresholds, benchmarks and other interventions. Stemming from the addition of that one piece fulfills almost all of the requirements to satisfy ICAO Doc 9966 and 14 CFR 177.7. It is a bit of progressive thinking that a return on investment calculation would prove it to be a worthwhile endeavor. Yet the business biases seem to present a barrier to the acceptance of this idea.
It is commonly accepted that fatigue is a major causal factor in most aviation accidents and incidents. This wide acceptance seems to affect people and organizations to take fatigue for granted. Tired pilots fly worse than rested pilots. Everyone knows that therefore there is no need to measure it. The acceptance of this thought process leads to a collective confirmation bias that ignores evidence to the contrary thereby suppressing progressive thinking.
The industry gives a general perception of anchoring on a low accident rate as proof that the fatigue regulations and status quo is working. Whether or not this perception is based in truth is irrelevant. The perception is reality. This system justification provides the foundation of the staunch adherence to the status quo which hampers progress that is good for the individual and the collective interest of the industry. Without better data, better analysis and better information we as individual operators cannot provide our management teams with cost-saving alternatives. Without better data, better analysis and better information we as an industry cannot provide the law makers with better regulatory recommendations. Human factors data can be easily captured and done so with a positive return on investment. All is takes is a spreadsheet and an open mind to prove it.
Satterfield, B. & Van Dongen, H. Occupational Fatigue, Underlying Sleep and Circadian Mechanisms, and Approaches to Fatigue Risk Management, Fatigue: Biomedicine, Health & Behavior,2013, Vol1(3): 118-136.
Van Dongen, H. & Belenky, G. Model-based Fatigue Risk Management, The Handbook of Operator Fatigue, 2012: 489-508.