What's Hot

Nassim Nicholas Taleb's blog, an inspiring read | Incerto

Tuesday, July 16, 2013

Ishikawa Diagrams Explained

Several risk analysts have asked me how difficult Ishikawa Diagrams are to build and in this short article we are going to quickly construct one of these schematics.

"In ISO 31010 Fishbone Diagrams and Root Cause Mapping are recommended techniques for structured analysis techniques, section B.12.4 of ISO 31010. Do you have an example of such a Fishbone Diagram and are there any software tools that can assist with these techniques?"

Ishikawa Diagrams
Ishikawa Diagrams are often referred to as Fishbone or Cause-Effect Diagrams but in principle, they are all similar analysis techniques. First developed by Kaoru Ishikawa who was a leading expert in quality management in the fifties and sixties, the diagram and analysis technique ended up taking the name of its author. These schematics are often used to investigate how different factors can combine together to cause a specific outcome and are common techniques for quality control or defect inquires. The diagrams are quite popular because they are easy to understand and they can also be appraised by technical or non-technical managers alike.

The concept that a risk event or a specific conditional outcome can be driven by many causes, leads analysts to list and classify these various different factors in tree like arrangements. Ishikawa diagrams generally flow left to right with the causes pointing into a line that will eventually connect to the outcome.

For example; We may believe a fire can be caused by the combination of different circumstances such as environmental conditions, the effectiveness of human policy to control fires, the mismanagement of a process and so on.  So how would we present these factors in a Ishikawa diagram?

Ishikawa Diagram in R-Project | Causal Capital   [Click image to enlarge] 

In the screen snapshot above, I have used the Six Sigma library in the R-Project statistical language to build up an example Ishikawa diagram of our fire hazard case study. The method was very easy to program and it only took me a matter of minutes to complete.

Ishikawa diagrams are certainly a tidy way to present causes that drive outcomes but they do have some drawbacks that risk managers need to keep in mind. Firstly, these diagrams don't show us which causes are most common or the most important to manage, nor do these charts quantitatively measure or represent how the causes may combine or correlate together to create an outcome that is more severe or frequent. For these types of risk analysis, other tools and statistical functions will need to be considered and we'll build a quantitative model for measuring these additional requirements in our second article on Ishikawa diagrams.

1 comment:

  1. Nice explanation. You can find more ishikawa templates and examples at creately diagram community that can be used by anyone.