Root Cause Analysis – Art or Science?
There are many commonly held beliefs about root cause analysis that bother me. Perhaps the single most irksome to me is the statement "it's an art, not a science." I don't have anything against art, but I don't believe that this statement does justice to the practice of root cause analysis. In fact, I believe it is one of the most damaging perceptions that can be held by an investigator or be communicated to others.
So, why do people believe this? One widely-held perception is that root cause analysis is not repeatable, i.e. the belief that different analysts performing independent investigations of the same issue will not arrive at identical results. Another commonly-stated reason is that it can be difficult to state the results of a root cause analysis with much precision, especially if issues of human or organizational performance are involved.
In addition, I believe that many people instinctively recognize that some aspects of root cause analysis are inherently subjective. By necessity, RCA requires that an analyst compare that which is to that which ought to be... and what ought to be is often a matter of opinion. Furthermore, the development of recommendations (the most obvious outcome of root cause analysis) is certainly subjective in nature, as there is rarely an absolute standard to determine which solution is best, even for purely technical issues.
However, I don't believe any of the above justify characterizing root cause analysis as an art, or as "more art than science." In general, art is the application of creativity for its own sake without any objective criteria for judging quality. In contrast, root cause analysis, while containing elements of creativity, is rarely (if ever) applied without a specific purpose, or without objective criteria for what constitutes a quality outcome.
I would argue that root cause analysis is a science, or is at least a process that must be performed scientifically. The following description of scientific method from Wikipedia (as of October 2006) provides a good summary of my viewpoint:
Scientific method is a body of techniques for investigating phenomena and acquiring new knowledge, as well as for correcting and integrating previous knowledge. It is based on observable, empirical, measurable evidence, and subject to laws of reasoning.
Note the emphasis on the use of evidence and reasoning for investigating and acquiring knowledge: this could very well serve as a working description of the root cause analysis process. Consider also that science can refer to both natural (or "hard") sciences like physics and chemistry, or social ("soft") sciences like economics and sociology. The following description of social science from Wikipedia (in October 2006) provides additional insight:
The social sciences are groups of academic disciplines that study the human aspects of the world. They diverge from the arts and humanities in that the social sciences emphasize the use of the scientific method and rigorous standards of evidence in the study of humanity, including quantitative and qualitative methods.
So, even root cause analysis efforts that delve into issues of human and organizational performance must be performed scientifically and be subject to rigorous standards of evidence. (Of course, this has little bearing on the parts of a root cause analysis that deal solely with physical/technical issues.)
In summary, the root cause analysis process contains many elements that are not consistent with the belief that it is an art. These elements (evidence, reasoning, objective standards), however, are fully consistent with the characterization of root cause analysis as a science, or at least as a process dominated by scientific thinking. While certain aspects of the process may be subjective in nature, even these must be performed within an objective, scientific framework for the process to have any validity. Thus, the assertion that RCA is "more art than science" is not justified, and should not be promoted.
by Bill Wilson