In this lesson, we go over a brief history and overview of the Process Behavior and Control Chart in QuikSigma. Have any questions? Leave a comment below and we’ll get back to you!
Control charts, or as we often call them process behavior charts, were invented by a fellow by the name of Walter Shewhart. So sometimes you’ll hear them called Shewhart charts. He worked at Bell Laboratories and he had many many processes that he was responsible for and he needed a simple tool that would signal to him which processes needed his attention and which could likely be left alone. Now, what he actually invented was the X bar and R chart, which is a very powerful form of process behavior charts. The I-MR or X-MR chart that we’re going to be looking at here in a minute, came along a little bit later and the I-MR chart is absolutely the Swiss Army knife of Six Sigma. It’s so simple and it does so much for you it’s just a tremendous tool and one that it’s well worth your time to master.
So to get to that tool in QuikSigma, I can go to baseline and select it from the process capability / behavior charts menu and those two tools should be joined at the hip and in QuikSigma they are. The other route in is through the toolbox which will also give you access to control charts and here we can select the flavor of chart that we want to do and the one we’re going to work on at the moment is the I-MR chart. So we’ve added data here and these two graphs are individuals and are moving range charts. The first thing that an I-MR chart does for you is it gives the data in a time series, which gives you a context. The second thing that it does is it will automatically alert us if our process has changed in a way that is not easily accounted for by normal random variation. This is the individuals chart, this is the moving range chart, these are individual data. Here’s 93.123 and here’s 105.753. If I want to see values, I can just lasso and right-click and turn on the probe options and show labels and the ones that I’ve lassoed I’ll get numbers by and that will also show me exactly what my limits are.
Now, on the other hand, here’s an example of a process that is very definitely not stable and predictable. Now the interpretation of this, this being our individual points and this being the difference between successive points, the first thing we look at is we’ve got a bunch of out of limit points down here and we’ve got some more up here. Those violate rule number one, which is a point outside the three sigma limits. You notice over here, this point has a little 3 by it. So that says that we violated rule 3, which is four of five points outside a 1 sigma limit on the same side. Other things that we can glean from this, see I’ve got a pretty big shift in my process right here. Well, the lower chart is based on these differences. So this point right here is saying that this shift in the data is too large to be accounted for by normal random variation. So this process needs some work because before it can be a Six Sigma process, it has to be stable and predictable. It has to be capable of meeting requirements and it has to be controlled and you do it in that order. This one needs the stability fixed before we do anything else.