The fields of incident investigation and root cause analysis are over-abundantly supplied with acronyms, like E&CF, ETBA, MORT, MES, etc. After much investigation, I've determined that to become really famous in this business, you've got to have at least one acronym attributed to you. Therefore, I hereby unleash the BOGUS test upon the world at large, as defined by these five factors:
- Beyond Control
Obviously, BOGUS is an acronym. What makes BOGUS better than most acronyms, however, is that it is easily pronounceable, is spelled the same as a real English word, and the meaning of that word is applicable to the concept. In other words, it is the perfect acronym, and it is all mine! Well, okay... you can use it too, but you should first read the explanatory text below.
- Beyond control: Some conditions are beyond our control, like stupidity, gravity, or the weather. We can't make them go away, nor can we change their fundamental natures. The problem is that by identifying such a condition as a cause, we run the risk of deciding to ignore it because its "beyond our control." The attribution of cause should instead be made to a lack of protection against a hazard.
- Obvious: At times, the cause of a problem seems completely obvious -- so obvious that we can't resist naming it. Items that fall in this category often involve actions by people, including "operator error" and "lack of procedure compliance." Stopping at this point is akin to finger-pointing, though. People do what they do for a reason, good or bad... dig deeper and find out why.
- Grandiose: Sometimes you hear cause statements that make you wish you knew what the investigator was smoking. "We did not leverage our core competencies to instill a culture of knowledge discovery and effect a paradigm shift to agile performance..." is an example of a grandiose cause statement. It would be better to say something like "... we don’t learn from our past mistakes, and that is hurting us." There is virtue in simplicity -- try to distill cause statements down to their pure essence.
- Unrelated: We often have more than one problem to deal with, and it can be tempting to tie one problem to another in order to save time and effort. However, in doing so we must ensure that we do not attempt to "force-fit" an unrelated cause onto a different problem. In trying to kill two birds with one stone, we might later find that both birds are alive and well, and happily making new baby birds that can't wait to grow up and come peck your eyes out.
- Simplistic: Earlier I said that there is virtue in simplicity. However, there is danger in being overly simplistic. We must recognize that some problems are more complex than others, and may result from the interaction of several different causes. If we don't identify all the relevant interactions, we may miss something truly important.
The best defenses against BOGUS cause determinations are rigorous application of Necessary and Sufficient logic during an investigation, and requiring corroborating evidence for every Causal Factor. Then when you're done investigating, use the BOGUS test as a final check of root cause statements, prior to developing corrective actions. Think of it as a quality control check of your root cause analysis.
Alternatively, you might want to use the BOGUS test if you're responsible for giving final approval for implementation of a corrective action plan. Please do me a favour, though... if you do decide to reject a report because of the BOGUS test, don't tell the report's author about me. I don't need that kind of attention!
by Bill Wilson