“I don’t like to use bowtie for my incidents, it feels too simplistic. I much prefer 5-why or Root Cause Analysis” is something we hear once in a blue moon when we are at conferences. This irks us as it is similar to someone saying “I don’t like cheese, I much prefer chalk”. Some, often new, safety people seem to misunderstand the goal of the bowtie. As a result, they often use it in the wrong situations or for the wrong goals and end up disliking it. In this short blog, we will go into what bowties are for, and what they aren’t for. As bowtie is sometimes mistaken for incident analysis techniques. We will also go into what incident analysis techniques are for, and what they aren’t for.
The bowtie method is a qualitative risk assessment method that can be used to analyze and communicate how high-risk scenarios develop. The goal of the exercise is to proactively find all the ways in which an organization can prevent these high-risk scenarios from developing through control measures, also known as barriers.
If gaps are found, it is advised that the organization takes up measures to fill those gaps, e.g. by implementing more barriers or strengthening existing barriers.
As such, bowtie methodology is primarily an abstract, conceptual, paper-based exercise in which we try to predict all the ways things can go wrong.
Incident analysis techniques such as 5-why, STAMP, Tripod Beta, BSCAT, BFA etc., are methods to investigate how one of the high-risk scenarios (such as identified in a bowtie) have come to be. The goal of this exercise is to reactively find all the ways in which a specific high-risk scenario developed. Quite often the scenario developed because some barriers did not perform their function, and during the proactive bowtie exercise we were unable to predict the incidents’ exact combination of events. During the incident analysis step, more time can be invested, to really dig in, and find the root cause of events. As a result, you can go into more details with incident analysis techniques. With bowtie methodology, however, you do not have this luxury. As with bowtie methodology, when the gaps are identified, it is advised that the organization takes up measures to fill those gaps, e.g. by strengthening or adding barriers.
As such, incident analysis techniques are primarily about learning from real-world feedback in which we try to correct and adjust for the combination of things we failed to foresee during the bowtie workshops.
|Bowtie methodology||Incident analysis techniques|
|Paper world||Real world|
Figure 1: Bowtie methodology side-by-side comparison with Incident Analysis techniques
Now that the differences are clear, we can look at the similarities. Both bowtie methodology and (most) incident techniques focus on barrier management. We can imagine that a potential scenario that was identified in a bowtie, occurs in the real-world due to barriers failing. When we go to investigate this scenario using an incident analysis technique, we can learn why the barrier failed. This is valuable information that can be mapped back onto the bowtie (see Figure 2).
Figure 2: Incident analysis can be used to support bowtie methodology
Ironically, people had it wrong the other way around – bowtie methodology is not meant to be used for incident analysis, incident analysis can be used to support bowtie analyses.
Has this article sparked your interest or do you have any questions? Contact us.