In this lesson we go over a couple Six Sigma Root Cause Analysis Examples and how you can use them in your daily, and business life. If you have any questions, please leave a comment below. We’d be happy to help!
Here are some thoughts for you about root cause analysis. First, some basic principles. Number one, for the most part things happen for a reason, and we call those reasons causes. The second idea is that not all causes are equally important. Some are much more important than others. And finally, if we can understand root causes and control them then we can get the results that we like out of our process.
One of the commonly used root cause analysis tools, is what I call the two-year-old approach. You know, if you’ve ever had the pleasure of having a two-year-old in your life, you know they ask, “why?” a lot. So the point of the exercises to ask why at least five times. And here’s an example that sometimes is used. Hey, there’s oil on the floor. Well, the immediate reaction is, well, let’s clean it up so we have a clean workplace. But, it’s better to ask, why is there oil on the floor? Well, there’s oil on the floor because the machine overhead is leaking. Oh, why is the machine overhead leaky? Well, the machine overhead is leaking because it’s got a bad gasket. Why does it have a bad gasket? Because we have a policy and we got such a good deal on a large batch of these low quality gaskets.
There’s another tool, that similar in spirit, to the 5 whys approach, and that sometimes is called variation breakdown or it’s sometimes called the thought map. And all it does is, like five whys, keeps asking what influences this or why does this happen? But it does not make the assumption that it’s always just one cause. So one question is, what might affect the amount of time that it takes me to get to work in the morning? Well, off hand I can think of some different things. I might choose different routes. For example, I might have the scenic route, or the freeway route, or I might prefer the route past the coffee shop.
Certainly the level of traffic on the road is going to make a difference. How early I leave is going to make a difference. And I might choose different modes of transportation. That’ll affect how long it takes me to to get to work. Then, for some of these, one level is all you need to go to. For example, I may only have three routes to choose from and so this tree ends right here but for the amount of traffic on the road, well, that might be caused by the number of accidents, by the road conditions, by the amount of road work that’s going on, or by the time of day. If I can avoid rush hour, so much the better.
Well then you ask, what affects road conditions? And the road conditions are mostly affected I think by the weather, and it can be snowy, or it can be rainier, it can be clear. So basically you go through and make this little chart that lets the causes flow down, and the lowest level items on the chart are then what you would consider your root causes. Now, in some cases you’re going to come up against one of these where you say I don’t know, and that’s a good place to start an investigation.
Another tool that’s very useful is the Pareto Chart. This nested Pareto Chart is our patented invention, and it’s very handy for finding where, and when, and by whom, defects were made. That’s a really good simple root cause analysis tool. For a very thorough root cause analysis, we use a trio of tools in
concert. And that starts with the process map. And of course in the process map one of the important things that you’re doing is looking at your knob variables. The input variables that make things what they are and cataloging all of those.
Then we next move to the cause-and-effect matrix because remember we said not all variables were important. Not all causes are equally important, and in the cause-and-effect matrix, we prioritize and bring to the surface those variables that are likely to be most influential. And finally we go to
the FMEA and drill down to take a good look at the variables that the cause-and-effect matrix indicated were likely to be most important, and that’s a
very nice root cause analysis. For a root cause analysis tool with a little bit more mathematical rigor, we can turn to a process behavior chart or as some people call them control charts. And all that does is put your data in the order that they happened. It provides a center line, which is the average or sometimes the median, and it provides some limits here that define what constitutes an unusual event.
Well, if I get something like this, where I’ve got a point out of limit or if I’ve got shifts in my process, the rules of the process behavior chart will detect those. Then I have a very good basis for going and asking, what was going on right here? What happened that caused this to happen? This other stuff isn’t very interesting because statistically it’s all the same. We really can’t tell one of these points from another. But this one is different. That makes it interesting.
Sort of the granddaddy of the root cause analysis tools is what we call data mining, or sometimes exploratory data analysis, and that’s just a way of finding the hidden relationships in your data. Now, this may look a little forbidding, but believe me, it’s not. It’s extremely simple to use. What we have down here is some data that we have taken. We’ve kept track of how many pounds of potatoes we got and as input variables or causes, the temperature, and the water, and the type of soil that we planted in, and whether or not we used to pesticide. What the computer then does is it builds us a model that we can play with and we can slide this back and forth. And here’s our average harvest weight right here, 238.6, and if I get warmer weather, what happens? Well, it goes up to 252.8. And if I plant in sandy soil, I get only 240 pounds.
So, this gives me a wealth of information about the relationship between the root causes and the outcomes that I get. So that should give you some good ideas about root cause analysis tools and how to use them. Thanks for watching!