The Five Whys is a technique for discovering the root cause (or causes) of a problem by repeatedly asking the question, “Why?” Five is an arbitrary figure. You never know exactly how many times you’ll have to ask “why.” The Five Whys technique helps to identify the root cause(s) of a problem and see how different causes of a problem might be related
To begin, the problem should be described in very specific terms. Then, ask “why” it happens. If the answer doesn’t identify a root cause, then ask “why” again. The root cause has been identified when asking “why” doesn’t provide any more useful information. Continue asking “why” until the root causes have been identified. This can take more or less than “five whys.”
Remember, always focus on the process aspects of a problem rather than on the persons involved. Finding scapegoats doesn’t solve problems!
Here’s an example. A program office wanted to find out why they missed their Initial Operating Capability (IOC) date.
We missed our IOC!
Why? Our contract delivery schedule slipped.
Why? There were a lot of engineering changes.
Why? The contractor didn’t understand our initial requirements.
Why? Our technical data package wasn’t prepared very well.
Why? We only took one week to prepare it.
At this point, the group recognized poor requirements planning as a root cause of their problem. As a result, they decided to allow more time up front in the planning process for requirements analysis.
Suppose you hear the following statement: “We used to have a respectable defect rate; it ran about 100 defects per million opportunities (dpmo). Then, a few months ago, it doubled to about 2000 dpmo.”
Using the Five Whys technique, the subsequent discussion might go something like this:
Q: Do you have any ideas why the big increase in the defect rate?
A: I’m not sure, but it could be a couple of things. For example, it might be the new materials that our new supplier is sending us, or it could be the higher conveyor speeds that we have to run on the line in order to meet the new production goals.
Q: What do you think it is?
A: I think it might be a combination of the new materials and the higher conveyor speeds?
Q: Why do you think that’s the reason for the higher defect rates?
A: Well, we used to run higher speeds on a job we had a few years ago, and from time to time the defect rate would jump up and then come back. We were never absolutely sure why.
Q: How about the new materials? Why do you think that they may be contributing to the increase of the defect rate?
A: It’s nothing official yet, but we’ve heard that the new supplier has been cutting corners and the quality of the new materials we’ve been receiving from them is questionable.
Q: That all sounds pretty interesting, but why do you think that it might be a combination of the material quality and the higher conveyor speeds?
A: It’s the timing. The defect rate didn’t really jump up when the conveyor speeds increased, but it did when the new materials entered the line.
Q: What can we do to find out for sure?
A: We can use a Design of Experiments (DOE) to figure out if the interaction between the new materials and the conveyor speeds are really why the defect rates have increased.
Ask “why” five times to determine the root cause.
1. Why is this machine not running?
Because its drive belt is broken
2. Why is its drive belt broken?
Because the drive gear was not turning fast enough
3. Why was the drive gear not turning fast enough?
Because the drive shaft’s lubrication reservoir ran empty
4. Why did the drive shaft’s lubrication reservoir run empty?
Because the PM for this machine is overdue by almost 2 weeks
5. Why is the PM for this machine overdue by almost 2 weeks?
Because the lubrication maintenance person is on a 2 week vacation
6. Why didn’t someone else cover for the lubrication maintenance person during his vacation?
Because we do not have a vacation coverage plan for the maintenance department, and operators are not trained and empowered to do lubrication.
Sometimes you have to ask “why” more than five times to get to the root cause.