Special vs. Common Causes of Variation
A colleague was telling me the other day that he has a client with a problem. His manufacturing defect rates are 5%, and he wants to know how I would approach fixing it. The first thing that came to mind was I wonder if the process is stable and if all of the special causes of variation had been removed. Although my colleague told me he felt it had been because the process was so old, I was not sure. After all, how many people really understand the difference between common and special causes of variation enough to be willing to spend money to eliminate them? This month, the topic will be understanding the differences between special and common causes of variability.
Special Cause
There are infinite types of variability. Thankfully, only some affect the process. Special causes of variability are those that happen external to the process. They are not intrinsically not part of the process, and when they occur, they lead to a statistically significant deviation from the norm.
Examples of special causes of variability are:
The first labels on a roll of self-adhesive labels are damaged, marred, or otherwise unusable.
The cartons near the door of a warehouse are exposed to rain and degrade.
The first piece off a molding machine are malformed, short shots, or have excessive flash. - A power spike in the line causes an instrument to read a larger signal.
It should be noted that special causes of variation may be quite "common", but that does not change the fact that they do not have to be there.
Common Cause
Common causes of variation are more insidious. These are in integral part of the process, and typically require more skill in reducing.
Examples of common cause of variability are:
- Variation in the weight of an extruded textile or plastic tubing
- Variation in moisture content of a resin
- Particle size distribution in a powder
- Wax thickness in a roll of heat transfer labels
Common cause variation is the natural variation in a process, and it defines the statistical variability of a process.
Suggestions for Improvement
While common cause is often attacked by an Engineering department, it is fruitless if special causes are not eliminated. Special causes can so affect a process they will drive up variability substantially. They are typically not pursued as they are often procedural in nature and it is more appealing to install a new meter than it is to change human behaviors.