On July 7, 2017 at 23:56 (San Francisco/Pacific Time) Air Canada flight ACA759 almost landed on a taxiway where four aircraft where waiting to take off. ACA759 missed the first aircraft by only 100 vertical feet. This could have been a terrible accident. Have a look at this animation of the incident:
“Was this incident due to fatigue?” I have been asked this question more than a few times about this and other incidents. Frankly, I don’t know if fatigue played a role in this incident. Very little data about the San Francisco incident has been made public so far.
How do safety investigators look into fatigue?
What I do know is that the National Transportation Safety Board (NTSB) will be following a very strong methodology to investigate the possibility of fatigue’s impairing influence. Investigator’s first red flag would be that the incident happened at night, possibly after a really long day and perhaps at the beginning of the pilot’s circadian trough, a period of high fatigue and low performance for most people. Over my greater than nine years with the Transportation Safety Board of Canada (TSB), I got to know many of the human factors investigators at the NTSB and I am sure this red flag would have sparked an investigation into fatigue’s role. The NTSB uses a method to investigate fatigue that is very similar to the method I optimized for the TSB.
The first two things they will likely look at are how long the pilots were awake and whether their circadian rhythms were aligned such that they were landing during their circadian troughs. The first factor is easy enough to decipher. It’s simple math. Incident time minus the time the pilots woke up and you get an accurate calculation of time awake. If the pilots were awake for 17 hours or longer around the time of the incident, this would concern the NTSB investigators. I call this the Continuous Wakefulness Fatigue Risk Factor.
This second factor is a little tricky to investigate because the investigators will have to estimate what time the pilots’ brains and bodies thought it was at the time of the incident. Pilots will often have to cross many time zones and try to sleep at all sorts of strange times. This can easily throw their biological rhythms out of whack. If the pilots’ circadian rhythms were terribly desynchronized, it will be almost impossible to know what time the pilots’ biologies might have been set to at the time of an incident. One thing they will know, is that this circadian desynchronization would have increased the risk of fatigue.
If, however, the pilots in the July 7th occurrence called Toronto home base and they were lucky enough to have their circadian rhythms nicely anchored to Eastern Time, then it would be safe to say that their brains and bodies thought it was three hours later, or 02:56, right in the middle of their circadian troughs, a time of day effect that increases the risk of fatigue related-performance impairment. I refer to the circadian desynchronization and timing effects collectively as the Circadian Rhythm Effects Fatigue Risk Factor.
NTSB investigators won’t stop here. They will also consider the effects of other fatigue risk factors. My list includes a total of six:
- Acute Sleep Disruption
- Chronic Sleep Disruption/Sleep Debt
- Continuous Wakefulness
- Circadian Rhythm Effects
- Medical/Psychological Conditions, Illnesses and Ingestibles
- Sleep Disorders
Why is it so challenging to link fatigue as a contributing factor to an accident?
After looking at all the fatigue risk factors, the NTSB investigators will have a very good idea of whether the pilots were fatigued when they lined ACA759 up with the taxiway. But this won’t prove that fatigue played a role. Investigators will have to first figure out which of the pilots’ abilities may have been compromised, if any. Then they will have to look at whether these abilities could have been affected by fatigue or some other influence. This is the most difficult step and one of the reasons why it is so hard to identify fatigue as a contributing factor.
It is always a challenge to identify which abilities were involved in an incident. For example, could the pilots’ visual abilities have been affected by a poor airport layout? The taxiway did run right alongside the runway. Or could a cockpit display have affected the pilots’ abilities to understand (cognitive ability) the position of the aircraft? As aviation becomes more automated, flight management computers become more complex, and if something is not configured properly, displays can provide pilots with inaccurate information. In these examples, abilities were not compromised; instead, information was simply being used as presented. It would be hard to argue that fatigue played a role.
Visual and cognitive abilities are dependent upon non-fatigue related influences as I have suggested above and they can be affected by many other factors. If some abilities were compromised, the investigators’ next task will be to try to figure out if the identified abilities can be affected by fatigue and whether a totally different explanation may better fit the data. This normally requires research into relationships between abilities and fatigue and other influences. Abilities that are commonly affected by fatigue include:
- Problem Solving
- Decision Making
- Reaction Time
- Ability to Pay Attention (referred to in fatigue science as Vigilance)
Fatigue as a contributing factor?
The last step investigators will take is to bring all the information together to formulate a conclusion about fatigue’s role. Fatigue will only be identified as a contributing factor, if the following three conditions are met:
- there is data that demonstrates the pilots were likely fatigued, and
- if the research indicates that the impaired abilities in question could be affected by fatigue, and
- only after all other plausible explanations have been exhausted.
It will be interesting to follow this investigation and to see what the NTSB concludes about fatigue’s influence. The Transportation Safety Board of Canada will be following the investigation closely, through their occurrence # A17F0159.
If you would like to follow along, here is a link to the NTSB’s investigation: https://www.ntsb.gov/investigations/Pages/DCA17IA148.aspx