Arriving at Actionable Human Factors Metrics

Oct 2, 2025
SM4 Aviation Safety Program

Written for Global Aerospace by Matthew van Wollen, Pulsar Informatics

Marketers use clever metrics to measure the size of their target audience or to gauge how well their advertising is performing. Want to know how many people are walking the streets of New York City? Look up subway ridership numbers.

Earlier this month, the governor’s office announced that NYC Transit counted 26.8 million riders in one week, a new post-pandemic record. Or take cellphone data: Location-analytics firm Placer.ai used it to find that foot traffic in New York City exceeded 2019 levels for the first time just this past July.

In aviation, we also collect data to quantify our understanding of performance. Though the metrics we track are perhaps more mundane than what a coffee chain would use to determine how many new locations should be opened on Fifth Avenue, they are nonetheless crucial to ensuring safe and efficient operations. You are probably familiar with some of them that are related to FOQA data: landing speed, runway excursions, hard landings and go-around rates. What about metrics related to human factors data?

Quantifying Human Factors

The effectiveness of the interface between human and machine is notoriously difficult to pin down (and becoming even more obscure in the age of AI). The industry has developed a good framework for tracking crew resource management, pilot workload and proficiency training. Hovering over all these factors is one that is trickiest to measure: fatigue risk.

Where does elevated fatigue come from, really? To help answer this question, commercial airlines have formal procedures in place for flight crew members to notify the operator that they experienced elevated fatigue during a flight duty period.

These fatigue reports (sometimes referred to as fatigue calls) enable the safety team to better understand the root cause of the fatigue incident and consider what can be done to prevent a recurrence. However, fatigue reports have even more value: Aggregated, they represent a structured dataset that yields useful quantitative safety metrics. Here are some examples:
• Rate of fatigue reports per 1,000 completed flight duty periods
• Rate of fatigue reports per month
• Percentage of extended flight duty periods that resulted in a fatigue report

Business and charter air operators, having fewer flights and duties than their commercial counterparts, may find that these quantity-based metrics have limited statistical power. However, the full value of fatigue report data is unlocked when paired with another data source: subjective and objective measures of fatigue risk.

Distilling Where Fatigue Comes From

When a flight crew member submits a fatigue report, their feeling of elevated fatigue may be related to operational factors: an extended duty period, night flying, changing time zones, etc. But it could also stem from individual traits.

These cover a wide range of characteristics, from personal susceptibility to fatigue stress to daily sleep needs not being met due to having a new baby at home. In fact, science has shown that most of the unexplained variance in measures of fatigue risk—the portion “left over” after considering the impact of work schedules—is attributable to these so-called non-operational factors.

Distinguishing whether a given fatigue report was triggered by operational factors or non-operational factors is critical to properly evaluating the fatigue report incidence rate. If a flight crew member faces a two-hour commute to the hangar, we would not be surprised to find that duty periods with an early morning start time for this crew member would be associated with a larger number of fatigue reports than for other duty periods or for other crew members who live closer to home base.

In addition, if the schedule-related fatigue level for this route was nominal (i.e., low risk zone) while subjective and objective measures of fatigue risk collected from the crew member were elevated (i.e., high risk zone), we would be able to draw the conclusion that the fatigue report was likely related to individual factors, not operational factors.

Leveraging Meaningful, Measurable Metrics

Analyses of human factors data, such as the above example, are made possible by collecting good data and establishing appropriate benchmarks. Pulsar Informatics has extensive expertise in helping air operators set up metrics that are meaningful and actionable for them.

We wouldn’t be able to offer advice on where to locate a café on Fifth Avenue, but we can certainly help create a fatigue risk management program that is tailored to your unique needs. Contact us to start the conversation today.