b.cognosco

[...] An illustrative example comes from the book Inviting disasters Inviting Disaster: Lessons from the edge of technology. It tells the story of a floating dormitory for oil workers in the North Sea that rolled over during the night killing over 100 people. The engineering experts quickly constructed different theories and complex explanations that focused on operational errors and management decisions.

All of these theories were wrong. It was eventually discovered through careful analysis that weeks earlier a crack in a support structure had been painted over, instead of being reported and repaired. This stupid, simple and small mistake caused the superstructure to fail, sinking the dormitory. Without careful analysis the wrong conclusion would have been reached (e.g. smacking the Atari) and the wrong lesson would have been learned.

Until you work backwards for moments, hours or days before the actual mistake event, you probably won’t see all of the contributing factors and can’t learn all of the possible lessons. The more complex the mistake, the further back you’ll need to go and the more careful and open-minded you need to be in your own investigation. You may even need to bring in an objective outsider to help sort things out. You’d never have a suspect in a crime lead the investigation, right? Then how can you completely trust yourself to investigate your own mistakes? [...]