The importance of analyzing failures and successes

What are the differences between the causes of a disaster or accident and the causes of a successful trip, event or project? Can success even have “causes?” Is there any reason to study successes to avoid future failures, or better yet, to increase the chances of success?

These, perhaps, sound like stupid questions. But they are not so stupid. The difference between a successful delivery of products and a disaster on the freeway might be very small. Not only is the difference small, but usually it is difficult to track the things that went right that actually had an impact on the larger system.

Cause Trees

In other words, if an accident happens because a tire blows, we can reconstruct a timeline and link it to a cause tree to find the root causes. We might look at the trailer tire that failed and see less than a 2/32” of tread along with a road hazard. If we speculate that the thickness of the tread was a cause of the accident, we create a tree from there of “why” questions. We might ask a follow up question like “Why is the tire so worn?”

Once that is answered, we would follow up again and again until we are satisfied. This technique is known as “Five Whys.”

If we can build a whole tree of causes and events, we can see which causes contributed to the accident. Each branch of the tree can be tested to see if it contributed to the accident.

The way to mentally test a branch of the tree is to ask: “If X didn’t happen, would the accident still have happened?” This ability to switch a branch on and off and see the effect on the accident is extremely useful.

The Key is Learning

A few months ago, the Dragon capsule was the first commercial spacecraft in history to visit the International Space Station. Everything went well, and it was a success. If we could distill the elements of that space flight, could we make subsequent flights safer?

The question is: If you have a successful event, trip, surgery, repair, etc., how can you learn from it? Some ideas might be:

  • Think through all the things that can go wrong. Build a system to account for those issues.
  • When something works, capture it in a procedure.
  • Create a checklist of the steps.
  • Learn from failures by doing a complete failure analysis.
  • Once a failure analysis is complete, look down the branches of the cause tree. Each of those branches is a potential latent failure. If you deal with them you will have a more robust system.

Success is great. Be careful that you have looked at your successful activities and that you understand which procedures, practices or products actually contribute to the outcome.

Remember, many doctors did not believe that washing your hands would limit infections.

Many societies thought that following proper rites would insure a good harvest.

It is important to establish the relationship of the event to its causes.

Good luck doing it.

Joel Levitt has trained more than 6,000 maintenance leaders from more than 3,000 organizations. Since 1980, he has been the president of Springfield Resources, a management consulting firm that services a variety of clients on a wide range of maintenance issues.