A Practical Guide to SPC Control Charts for the ASQ CQE

Statistical process control (SPC) is where quality engineering earns its name, and the control chart is its central tool. If you can read a control chart correctly and choose the right one for the data in front of you, you have mastered one of the most heavily weighted areas of the CQE Body of Knowledge.

I am John Lee of Alpha Training and Consulting. This guide covers what control charts do, how to pick the right chart, and the rules for spotting an out-of-control process—the exact reasoning the CQE exam rewards.

What a control chart actually does

Every process varies. The genius of the control chart, developed by Walter Shewhart, is that it distinguishes two kinds of variation. Common-cause variation is the natural, random noise inherent in a stable process. Special-cause (assignable) variation is a signal that something changed—a new material lot, a tool wearing out, an operator error. The control chart tells you which one you are looking at so you do not overreact to noise or ignore a real signal.

Choosing the right chart

The first decision is whether your data is variables (measured) or attributes (counted).

Control limits vs. specification limits

This distinction trips up many candidates. Control limits are calculated from the process data—typically the mean plus or minus three standard deviations of the plotted statistic. They describe the voice of the process. Specification limits come from the customer or design engineer—they describe the voice of the customer. Never place specification limits on a control chart; they answer a different question, which capability analysis addresses.

Reading out-of-control signals

A point outside the control limits is the most obvious signal, but it is not the only one. The Western Electric and Nelson rules define non-random patterns that also indicate special causes:

• A single point beyond three sigma.
• Two of three consecutive points beyond two sigma on the same side.
• Four of five consecutive points beyond one sigma on the same side.
• Eight or more consecutive points on one side of the center line (a run).
• Six points in a row steadily increasing or decreasing (a trend).

When a chart signals a special cause, the response is to investigate and eliminate the assignable cause—not to adjust the process blindly. Once only common-cause variation remains, the process is in statistical control and you can meaningfully assess whether it is capable of meeting specifications. That next step, process capability, is covered in its own guide.