Root cause analysis is a collection of tools and processes we can use to determine the most important causes for an issue we are trying to resolve. This is an important function as one of the top 5 reasons for project failures is poor root causation / no root cause identified. If we can identify the root cause to an issue, we have a good chance of solving it.
“Take away the cause, and the effect ceases.” – Miguel De Cervantes in Don Quixote.
Types of Root Cause Analysis Tools:
- Fishbone Diagram
- 5 Whys
- Most effective analytical techniques for root cause determination – Pareto and data analysis.
Getting Started with Root Cause Analysis
Before diving in with the aforementioned tools, it’s good to start with Causal Theory. In other words, Why does a certain problem exists?
To do this we start with a basic equation; Y = f(x) ;
Y is the Output of a process.
X are the Process elements that influence Y
Another way to state this is
Y is the crime ; X’s are the suspects that explain the crime of Y
Data door & Process door
Some of my instructors refer to the data door and the process door. Both can be used to get to root causation but certain tools are better in certain circumstances.
Effectiveness project – use data door.
Efficiency project – use process door.
Root Cause Analysis: Open-Narrow-Close
To validate root causes identified in the Open-Narrow-Close efforts, the team should employ three tools/techniques:
In the open phase you want to gather as many ideas as possible. To do that you might use a cause and effect diagram.
Generate as many suspects (people of interest) , Brainstorming (Cause and Effect Diagram, Fish Bone Diagram)
Narrow (SME of the team) – eliminate duplicates, narrow thru multi voting to get to a narrower list
– 5 Ys to break biggers Xs to smaller managable X’s
Close – validate using Hypothesis and convict them or set them free
Validated X’s will be worked on to IMPROVE
Potential Cause List
Use this to summarize your results at the end of the Analyze phase.
Put the diagram of the current process steps, defined outputs & the factors influencing those outputs, and a table listing the results of process analysis. Include all data sources used, how data was collected and the tools used in analyzing the data (hypothesis testing).
Prioritization of Potential Causes
- Do NOT Consider
- If a potential solution exists for the potential cause.
Root Cause Analysis Videos
Good video, awful sound. I muted it and just clicked through at a good pace. It gets the point across.
Other Helpful Notes
Also see Linear Regression for Y = f(x)
Also see Causal Theory=f(x)
Also see Hypothesis Testing