Statistical Process Control and Variation

Our processes produce a product or service that, hopefully, meet the needs of our customers. But the processes also produce something else: data. And statistical process control (SPC) is simply using that data (statistical) to make what we do (process) do what we want it do (control). At the heart of SPC are control charts. To effectively use SPC and control charts, you have to understand variation. In fact, all SPC training should start with the concept of variation.

Understanding Variation

Growing up, there were six children in my family. I remember we always had family dinner together. My father sat at one end of the table and my mother at the other. I sat next to my mother. As a young child, I always had to drink a glass of milk for dinner. Well, back in those days, we didn’t have the plastic cups with covers like they do now. No, my glass of milk was just a regularly sized glass.

When I was young, I would spill a glass of milk every now and then. Our table slanted toward where my mother sat. So, guess where the milk headed? Yes, right towards my mother, and she often had some choice words when this happened. After all, she had years of this happening since I was number five out of six children.

Of course, it was my fault. I just needed to be more careful. Or was that really true? This is where understanding the concept of variation is so important. Because if you understand variation, you will know that most of the problems that a company faces are not due to the people – it is due to the process – the way it was designed and managed on a day-to-day basis. In fact, as you will see below, the spilling of milk was, in all probability, not “my fault.”

Driving to Work – Common and Special Causes

Variation, when talking about SPC, comes from two sources, common and special causes. How long does it take you to drive to work? Maybe it takes you 25 minutes – some days a little more, some days a little less. As long as you are within this “normal” range, it is about what you expect. Suppose that “normal” range is 20 to 30 minutes. This “normal” variation is called common cause of variation. This type of variation is present in all processes – including your work processes. Common causes of variation are “consistent and predictable.” You don’t know how long it will take you to get to work tomorrow, but you know, as long as nothing unusual happens, it will take between 20 and 30 minutes.

Now for the unusual. Suppose you have flat tire on the way to work. How long will it take you to get to work? An hour? Two hours? This time is definitely out of the normal range of 20 to 30 minutes. This type of variation is called special cause of variation – it is not supposed to there – not part of the process. Special causes are sporadic – you don’t know when they will occur. Your processes at work have special causes of variation also – things that are not supposed to occur – that are not part of the way the process was designed or is managed on a day-to-day basis. Other possible special causes include snow and accidents.

With SPC, then, there are two types of variation – common and special. Why should you care? Because it tells you how to improve your process. If your process has a flat tire (special cause), you must find the cause of the problem and then try eliminate it from ever coming back, if possible. This is usually the responsibility of the person closest to the process. If only the natural variation (common cause) is present, you must fundamentally change the process. The key word is fundamentally — a major change in the process is required to reduce common causes of variation. And management is responsible for changing the process. So, to decrease the time it takes you to get to work, you would have to fundamentally change your process. This could include getting up earlier, moving closer to work, or changing the route you take to work.

What percent of your processes at work are due to common causes (natural variation) or to special causes (flat tires)? Dr. W. Edwards Deming estimated that 85% to 94% of the problems you face at work are to common causes. So, if you leadership blames all their problems on people – well, they are wrong at least 85% of the time!

Yes, Mom, I did spill milk. But according to the concept of variation, it was not necessarily my fault. I always wanted to move it closer to me so I could reach it, but you said to move it back because I might knock it on the floor. The glass was really big for my small hands. And then all the conversations with the 8 of us – too much to concentrate on my milk! Talking and reaching for the glass – the process was setup for spilled milk to occur!

So, when teaching SPC, always start with the concept of variation. Common and special causes of variation are the building blocks for using SPC effectively.