Watch that chart! (Special Cause and Common Cause “issues” SPC)

So, SPC, or statistical process control, is a very large and lengthy subject. In fact, there are many books authored on the subject alone. I never realized the impacts of actually studying statistics until I had a class with Dr. Davis Boothe. If you want to get a lot more detail about the subject, beyond what I will continue discussing in future blogs, you can order material here.

Today, I want to focus on the charts. Since most of use graph data I wanted to show you an example or two of how the charts actually tell a story. A chart has two main types of variation: special cause and common cause. Variation is the distant from point to point in the process or the range and comparison to the goal. When most of us create a run chart, we are looking at these things. Most people insert a trend line, and look for trends towards goals. The data can tell us so much more.

spcialcausevariationSpecial cause variation, is probably the most identifiable. There are points in the data which “spike” or cause the process to go out of control for intervals of time (or data points). What is happening that is unique while that data point was being collected – or timing in the process? If you correct the spikes, by correcting the process, the process will actually come back into control. For example, if ever truck was late for delivery on Wednesday’s during first shift. Through a problem solving type event you discover that the shift starts two hours late that day. So, adjust the times of the trucks or crew on Wednesday’s to fix the problem.

Common cause variation is when the process has normal variability based on the parameters (controls) of the process. Most commoncausevariaionof the time the process needs to be redefined so that you hit goal. So if the median is 5, and you are running 10 with common variation, the process needs problem solved to reduce the variation to the desired goal (target). Maybe the machine needs adjusted and that will solve the problem – by calibrating the machine to hit a true zero – you now have the same variation of the process but hitting the desired outcome.

Most of the time, we are presented with a combination. If you have a combination of both types, one has to be solved before the other can be. After you identify the unique “spikes” focus on shifting the process. Charts can also indicate different things… if the data runs high for 8 points, then low for 8 points and the trend continues – consider looking at shifts or operators. ¬†Either way, it is important to study the chart and look for patterns. The data can give you indications as to where to focus.



Leave a comment

Filed under Business: General, Lean

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s