Root Cause Analysis
Safety Moment #18: Incident Investigation: Words, Words, Words show how words mean different things to different people. Hence there can never be total agreement as to the what is a true “root cause”.
Not only are there different lines of thought to do with finding root causes, each set of events and causes extends backward indefinitely. It is said that the philosopher William James was once talking to a lady who believed that the earth rested on the back of a turtle. When he asked what the turtle itself rested on she reputedly said, “It's no use, Mr. James – it's turtles all the way down.” (Philosophers refer to this as the infinite regress problem.)
The concept of a never-ending chain of events which have causes, which are in turn events that have causes, and so on can be seen in the Why Tree approach to root cause analysis.
- Why did the pump seal fail?
Because the wrong type of seal was installed.
- Why was the wrong type of seal installed?
Because there was a mix-up in the ordering process.
- Why was there confusion with the ordering process?
Because the new enterprise management software was not configured properly.
- Why was the software not configured properly?
Because our equipment specialists were not consulted by the IT department.
- Why were they not consulted?
Because it was assumed that the old ordering system was up to date.
- Why was an unpublished change made to the ordering system?
Because . . . . . .
And so on and so on and so on.
This question/answer process can go on indefinitely, in principle all the way to the formation of the earth. Indeed, if the causes of an incident are pursued for long enough the team will eventually be discussing philosophical, moral, and theological issues to do with human nature. This is obviously absurd; a sensible stopping point is required. So, a rule of thumb for Why Trees is to stop at level five — at that point you are starting to come up with some form of root cause.
Once more, it must be stressed that the inability to find a true “root cause” is not a sign of failure. The analysis to do with the pump seal, for example, provides useful insights. But there are likely to be many other equally useful insights that can be developed from other trains of thought.