Wednesday, December 4, 2013

Accident and Incident Investigation – Part 3 (A Case Study)

A few weeks ago we went through some of the high level aspects of investigating an accident or incident (Part 1 and Part 2). Recent events in New York, where a passenger train derailed give us an opportunity to apply some of this knowledge, especially in light of the information that has come to light and the subsequent reaction (Click here for a review of the story).

If you take a minute to read the linked story above you’ll see that the train engineer was apparently nodding off right before going around a corner at over 80 miles per hour when he was only supposed to be going 30 miles per hour. As a union representative said, “most people are leaning towards human error.”

This is likely where most people stop their investigations. It was the engineer’s fault, they claim. Name, blame, shame, and re-train.

However, as good investigators we know that human error is not the place to stop an investigation – it’s where we start. As Sidney Dekker says “human error is a symptom of problems deeper in the system.” So let’s apply this knowledge and see what we come up with.

If you read the story, the engineer of the train reported that he had a full night of sleep the night before and was reported to not have any traces of drugs or alcohol in his system after the accident. The engineer had a good disciplinary record as well. So why did he start falling asleep?

His lawyer reported that he got a case of “highway hypnosis”. Basically, when you’re travelling for a long time, not doing anything, your mind starts to get bored. You lose focus and your mind starts to wander. This is something most motorists have experienced when driving for long periods of time, when their minds start to wander and then “wake up” 15 minutes later not remembering driving for the last 15 minutes. Basically your mind was on autopilot, which works great most of the time. But occasionally autopilot doesn’t work and the human needs to step in and correct the situation, hopefully before it’s too late.

So lets think about this for a second – a train engineer’s job is to pay attention to the tracks in front of him or her for long periods of time, without much mental stimulation. Are we surprised that occasionally these engineers “zone out”? No! It is human nature to get bored and zone out in those circumstances. What is surprising to us is that we don’t have more cases of accidents like this!

It’s still too early to know if this is exactly what happened in this case, but lets assume that the train engineer was right and he was in a highway hypnosis state. So a train engineer zones out in a situation where it is highly predictable that he would zone out and somehow that is his fault? That seems terribly unfair.

To illustrate our point, consider these questions:

Does the train industry not know that humans have a tendency to get bored? And if they do know, what have they done to keep those persons in safety critical jobs, such as train engineers, mentally stimulated to keep them focused when performing those jobs?

Why isn’t there some sort of alarm or warning system (more than just signage) to alert train engineers when there are significant curves or speed reductions upcoming, so that if those engineers do nod off for a bit people don’t have to die for a simple mistake?

Is there automation available that automatically slows down the train when heading into safety critical areas (if we can have automation that can fly planes, we should be able to have something that can slow down a train)? And if this technology exists, why hasn’t it been implemented to take the pressure off of the train engineers?

If we punish the train engineer for doing what he did, are we punishing someone for simply being human (i.e. imperfect)?


As we’ve discussed, ask not who’s to blame, ask what’s to blame. In this case, the train engineer clearly made a mistake. If we stop there though and replace the engineer with another person we should not be surprised if that person, at some point, makes the same or a similar mistake. If instead we take the focus off of blaming someone and start really looking deeper into the system to identify those weak points that are leading to these accidents we may actually be able to prevent future accidents. Remember, accident investigations are about improving the system to make it safer, not about finding blame.

No comments:

Post a Comment