Causal Investigations: Coincidence or Not?

Over the years, as a forensic engineer, I have often been asked if a seemingly unrelated event just before an electrical or mechanical failure occurred was likely a coincidence. Those who have asked me this or who have worked with me will know my answer: “I don’t believe in coincidences”. Why not? Let me explain.

When a complex system, meticulously designed and maintained, fails, it's rarely a matter of chance. Instead, a detailed investigation frequently reveals a causal chain, often rooted in a seemingly innocuous event.

Consider the scenario of a gas turbine failing a few days after undergoing a major overhaul following 8 years of normal operation. While it might seem like an unfortunate coincidence, a seasoned engineer would likely suspect a deeper connection. Such a rapid failure post-maintenance is a strong indicator of a systemic issue, perhaps a faulty component, an error in the maintenance procedure, or even a design flaw that was exacerbated by the maintenance activity.

In the intricate world of engineering, true coincidences are rare. When a system fails, there's usually a logical explanation, often rooted in a chain of events that can be traced back to a specific cause. By employing rigorous investigative techniques, leveraging data-driven information, and addressing human factors, the root cause of a failure can usually be determined. This information can then be acted upon to significantly reduce the likelihood of future failures and ensure the safety and reliability of critical infrastructure.

As Albert Einstein is quoted as saying, "Coincidence is God's way of remaining anonymous." In the realm of engineering, however, the truth behind seemingly random events is often uncovered through diligent investigation and critical thinking.

While at it, let me also discuss apparent cause versus root cause. It is crucial to distinguish between the apparent cause and the root cause when conducting a causation investigation. Apparent causes are the immediate reasons for an event occurring; for example, “the fire was caused by a Li-ion battery going into thermal runaway”. However, why did the battery go into thermal runaway? Root causes are the underlying factors that led to the apparent causes. In other words, in my Li-ion battery fire example, what was the specific cause of the battery going into thermal runaway? The difference between the apparent cause and the root cause is sometimes not appreciated, even by seasoned investigators.