ROOT CAUSE USING 5 WHYS

These materials are from Jeffrey K. Liker and the book, Developing Lean Leaders at All Levels.

Problem solve your way towards our Developing Lean Leaders’ Summit with Ritsuo Shingo, Paul Akers and 8 other speakers in Santorini, Greece – July 31 to August 4; https://lean2017.com

THE 5 WHY PROCESS

Driving to the root cause is probably the most important part of problem-solving, and driving to the root cause is often misunderstood.

Root cause sounds very, very scientific. One might mistakenly think there's one root cause, and you must use every statistical method possible and every means possible to find the precise root cause so that you're working on the right problem. This way you can come up with countermeasures.

The reality is that if you're doing problem-solving every place, all the time, every day; you could spend all of your time trying to find the root cause and never accomplish anything. It becomes obvious that you need to take shortcuts, and you need to realize that sometimes you're going to hit the bull’s-eye and sometimes you're not.

Taiichi Ohno originally taught root cause problem-solving through the five why method. He believed that if you observe the process, think and keep challenging yourself, do I really know the root cause; you’ll be able to answer the question, “Why did that happen?”

Generally, asking why? five times seems to be about the right number. Although you're looking at data, you are not using the most sophisticated multiple regression methods or design experiments. You'll still get close to the bull’s-eye. 

 COMMON MISTAKE: JUMPING TO SOLUTIONS

The more common problem is not that we don't get to the root cause; the more common problem is that we don’t even try. It is common to immediately think we know what the problem is. When this happens we immediately jump from the problem to the solution.

Consider a situation where a man is jumping into a pool with water in it. Imagine if you assume that pool has no water in it. Jumping can be characterized when you look at the problem, and you start brainstorming ideas and immediately implement those ideas. 

Sometimes you need to do that for little problems. For example, you might have an hour-by-hour board, and every hour you ask the workers to write down whether they met the target. If they didn't meet the target, they need to understand why. They need to write down why they did not meet the target, and immediately after that they would make a judgement on what the countermeasure should be and write that down in a  countermeasure column.

Now, they're jumping from problems to solutions. But they're doing it for small problems that are occurring hour by hour, and when you collect those problems and find the biggest ones, you should do a root cause analysis to get to the heart of the issue.

 ROOT CAUSE ANALYSIS

OHNO’S CIRCLE: WHYs

Taiichi Ohno advises observing the production floor without preconceptions and with a blank mind. 

Repeat why five times for every matter. He was famous for his Ohno circle. He would draw a circle on the floor using chalk, then he would direct the employee. “Stand in the circle,” he said, “and watch production and keep asking why and try to understand the problem and the root cause. Start now.” 

Two hours later he'd show up and ask what you saw. Two hour after that he would show up again and repeat the process. Normally he would make you stand in that circle all day. You might take breaks. Otherwise, you're standing in the circle, observing the same things over and over.

Each time he comes back, you have deeper analysis; you've asked another why. Notice that he didn't ask you to find the guilty parties; he didn't ask you to find the five who’s.

Usually the first why answer involves a person who made a mistake, then by asking why that person made a mistake you most likely drive to a system cause. 

 THE NARROWING AND FOCUSING PROCESS

The starting point is a large problem that may be very vague, for example; many quality problems. We want to solve these quality problems. So many things cause quality problems that we don't even know where to start.

Then you drill down and you find the most likely cause. This is before you even start asking why five times. Then you will find the point of cause.

In a famous Taiichi Ohno story, he tells one of his students to go and observe the assembly line, and the student sees a problem, and the student is ready to start solving it and then he asks them, “Where did that problem occur?”

The problem is a part; it doesn't fit right, and you're having problems fitting it, so we’re going to try to fix that problem. He asks them to go back to the manufacturing process that produced the part, and after that he might even ask them to go to the supplier. 

You first need to find where the problem occurs and at that point, known as the point of cause, and then you ask why it occurs, about five times, to get to the root cause. 

 I said you need to find the point of occurrence. There is one caveat. You need to focus on what you can control. 

The following example is a reasonable 5 why analysis:

Let’s imagine that the problem is that the defect rate is too high; it's not meeting our goal.

The reason is because we have too many defective parts.

Why? Because parts are not being assembled correctly in the assembly process (usually that's the last process).

 Why? The answer is because operators are making mistakes.

Why are operators making mistakes? Because the parts don't align properly.

Why don't the parts align properly? Because the design is poor, which means now we need to go to the engineers who are perhaps in a different location or even in a different country, and we need to tell them that they have to design the parts correctly. 

Once you start pointing the finger to another party, especially one that you have little influence over, you need to rethink the analysis. Keep in mind that you may not see that new design for months or even years if you followed this path. There are also probably very long lead times.

Then you have to ask yourself, “Is there a different answer to the five why questions that will lead us to something we can control?”

By doing this analysis, we are doing what is called effective root cause analysis. So we still have the operator error and we still know that the parts don't align properly, but now we ask ourselves, “Are we doing something to align the parts incorrectly?” This will focus our problem-solving exercise in an area which we have control over.

Why are parts incorrectly aligned? Because we are doing something that causes this misalignment.

Why are we able to misalign the parts?

The reason is we don't have an error proofing device that will notify us of an incorrectly aligned part, and prevent that part from passing to the next station. 

The error proofing device you create might be a fixture; it might involve training the operator; but we need some way to assemble parts that don't fit perfectly. The result is to satisfy the customer.

It doesn't mean that you're not going to communicate with engineering so that they can error proof the design at some point, but it means we're going to do something right now that will solve the problem. 

COUNTERMEASURES

Once we have the root cause, or what we think is the root cause?again we're not going to hit the bull’s-eye every time; then we need to come up with ideas to eliminate that cause or to eliminate that problem at the cause. We often call those solutions.

Toyota, however, calls them countermeasures and that's because Toyota does not believe that they know it is a solution. And, in fact, a great solution today may be replaced by an even better solution tomorrow. 

To illustrating that dynamic; the countermeasure is what we are doing today to counter that gap between where we want to be and where we are.

In fact, what we think is a good countermeasure needs to be proven scientifically by running experiments. We now associate the countermeasure with a hypothesis. If I do this, I will close the gap; that is our assumption which requires us to test it.

COMMON MISTAKE

Another common mistake is that we think we know more than we actually do. “I got it, I got it, I got it, I got the answer,” and then we push through that answer and maybe it works for a short time or maybe it helps a little bit, but is that the best answer? Perhaps somebody in the group might have a better answer, a better countermeasure. 

The disease of over confidence is one of the biggest barriers to problem-solving. In fact, if you already know how to solve the problem; if you think you're that brilliant, then you don't need a problem-solving process. 

Last chance to register for our Developing Lean Leaders Summit at https://leanleadership.guru/community/ . Register for Santorini, Greece, or Chicago, USA.

Finally, join our newest course for only $80 this month at https://leanleadership.guru/toyotakata/