RCA and 5 Whys
Following up on our last blog post about Ishikawa diagrams and root cause analysis (RCA), one of the other very popular problem-solving techniques we are asked about is the 5 Whys. The theory of 5 Whys is that asking the question “Why?” each time a reason is given will yield the cause of a problem and, therefore, the fix for the problem. Sakichi Toyoda, founder of Toyota Industries, is credited with creating the technique, which has added to its allure.
Here is a classic, short example that is often used to demonstrate the power of the 5 Whys in action:
The vehicle will not start. (the problem)
1.Why? - The battery is dead. (first why)
2.Why? - The alternator is not functioning. (second why)
3.Why? - The alternator belt has broken. (third why)
4.Why? - The alternator belt was well beyond its useful service life and not replaced. (fourth why)
5.Why? - The vehicle was not maintained according to the recommended service schedule. (fifth why, a root cause)
Why? - Replacement parts are not available because of the extreme age of the vehicle. (sixth why and additional whys, optional)
One shortcoming of the 5 Whys is the need to test the answer to each why as you proceed. In the example above, to answer the first “why” you need to make sure there are no defects with the starter, spark plug cables, battery cables and other parts involved in the starting process. Let’s assume the problem is the battery, there are a number of causes for a dead battery, including the age of the battery, a defective battery, the headlights were left on or a faulty alternator. We can go on, but we think you get the picture that 5 Whys requires a great deal of work to generate and eliminate a list of possible answers to the question “why?” at each step and to rule out each one. Without this, you might replace the alternator only to find out the battery was defective, and you wasted time and money on the wrong fix.
By contrast, RCA requires capturing as much information as possible about the problem, such as when the problem occurred, what happened before and after the problem was noticed, what type of variation was created, etc., before digging in on possible causes. Generating specific information about the dimensions and boundaries of the problem allows you to focus more quickly on a few possible causes and to eliminate unlikely causes as you go along.
The 5 Whys has also been criticized because it tends to focus on what the problem solver already knows, rather than being expansive about possible causes. Using the battery example, mechanics know the three most common reasons why a vehicle won’t start and how to quickly test each reason before making a fix. Do they need the 5 Whys? Probably not, but they may use the technique to document the steps they used to rule out other causes. But if a production machine develops a problem never seen before, will the 5 Whys be helpful?
Finally, because of the need to test multiple answers to the question “why?” at every step, problem solvers using the 5 Whys often focus on the symptoms and don’t get to the root cause of the problem and take corrective action.
Often we find that 5 Whys leaves a “cause unknown” statement that can become the focus of a rigorous RCA process. Using factual information around dimensions and boundaries in our RCA process will help in the elimination of possible causes to get to the true cause.
As with Ishikawa diagrams, we are not saying 5 Whys is a bad approach to problem solving. Rather, it is important to understand when 5 Whys can be used effectively, which is for problems that occur often and where the causes have been narrowed down to a few that can be tested against the specific issue at hand. However, we have seen many companies using 5 Whys spin their wheels on more complex problems where the possible causes are unknown. The result is a lot of time spent without getting to and fixing the real cause of the problem.