Variation is a given for every process. Distinguishing between the variation caused by the process inputs and that of special causes is important to understand so you can manage the variation.

Overview: What is common cause?

Common cause, in the context of variation, is a term coined by famous statistician Dr. W. Edwards Deming to describe the natural fluctuation in your process caused by the inputs or elements of the process.

The typical inputs to a process are people, methods, materials, environment and equipment. The combined variation from these individual sources will produce predictable and random variation in the output of the process. This steady state variation is the best you can expect given the variation of the input factors.

Since common cause variation is inherent in your process, any attempt to reduce or eliminate it will require action be taken on the source(s) of the individual input or factor variation. This will require statistical and root cause analyses to determine the source as well as how to reduce or eliminate it. Tools such as the control chart, multi-vari chart, fishbone diagram and hypothesis testing are useful for identifying the possible source of the common cause variation.

An industry example of common cause

It is not uncommon for organizations to misuse a control chart when trying to react to common cause variation. The control chart below shows an in-control process exhibiting common cause variation. All points are within the control limits and appear to be random.

Unfortunately, the supervisor of the department did not understand how to properly interpret and use a control chart. He praised the team for performing well at point 6 (higher is better) and angrily wanted to know what happened at point 10. The fact is, nothing happened. This would be analogous to asking why a pair of dice threw a 3 or an 11. You would expect 3s and 11s from a fair pair of dice.

Since the variation is common cause, there is no simple answer. Asking what happened will only send people on a fruitless and frustrating mission to find an excuse rather than a reason. The supervisor must identify the specific source of the variation and then make a fundamental change in the process to alter the variation.

How do you determine whether your process is exhibiting common cause variation?

The best tool to use is the control chart. Depending on the type of data you have, you will select the appropriate type of control chart to use. If the process is exhibiting common cause variation, the pattern on the control chart will have all the data points within the upper control limit and lower control limit and show a random pattern over time.

How can you reduce common cause variation?

Common cause variation is a result of the combined variation of the people, methods, materials, equipment and environment of your process. You must drill down and identify which of the potential sources of this inherent variation is the primary cause of the variation. By changing that factor, you will reduce the common cause variation.

Is all common cause variation good?

There is no value judgment of “good” or “bad” associated with common cause variation. You might have common cause variation in a process that consistently produces defective products. Saying it is common cause variation only means it is steady state and predictable, not good or bad.