If your control chart has plotted points that are not within the limit or show a non-random pattern in variation, this is considered assignable cause variation. You should be able to assign a cause or cause to it, as special cause variation can occur unexpectedly and is caused by something other than randomness.
A control chart can identify one of two types of variation: assignable cause (also known as a special cause) and common cause. Let’s look at what assignable cause variation looks like and compare it to common cause variation. This article will explain how to determine if your control signals an assignable cause, and how to respond if it does.
A control diagram shows two types of variation. Common cause variation is a random variable that results from process components or 6Ms. special cause variation can be assigned.
Your process variable is considered unstable or out of control when your control chart signals assignable causes variation. The Western Electric rules can help you identify signals of assignable cause variation. They include:
- One point beyond the upper limit or below the limit
- A trend that has 6 or 7 points consecutively increasing or decreasing
- A repeating or cycle
- A series of 8 or more points consecutively on either side of the average or center line.
Assignable cause variation may be due to a defect or fault, mistake, delay in processing, accident, or shortage. It could also be due to a unique combination of factors that work together to improve the process. Your process can be unpredictable if there are no assignable causes. It is important to identify and search for the exact assignable cause. Your process may have been improved by it. If so, you should incorporate it into your process to ensure that improvement is maintained and retained. It can harm your process, so you should seek to get rid of it.
What is the importance of an assignable cause?
Provides direction for action
You need to be able to identify the causes and understand what they mean. You shouldn’t ignore assignable or special causes.
Every unusual point does not have an assignable cause
You may also throw two dice at the craps tables at your casino. Are there any determinable reasons for throwing an 11 or 10? Or is it just a random chance? You would not expect the process to roll a pair of fair dice to reveal 10s or 11s. But what about a 13. It would be an unexpected result and most likely the result of something strange happening with the dice. This is also true for your process. If your control chart does not indicate it, don’t assume that an assignable special cause is being assumed.
This is useful for determining if your improvements were successful.
Your control chart should transmit signals of special cause variation when you are trying to improve the process. You can connect that signal to the specific cause of your improvement and you will know it worked.