January 2, 2018
Statistical Process Control: Three Types of Control Charts
If you have already made the decision to embrace a statistical process control (SPC) method—such as a control chart, which can visually track processes and abnormalities—you are already well on your way to bringing manufacturing quality control to your operations.
So, what’s the next step?
Determining which type of control chart to utilize.
“That decision is based on the style of the data stream that you would like to be presented,” Steve Wise, vice president of Statistical Methods for InfinityQS, says. “Trained statisticians will know when they walk up to a data stream what core control chart they should start with, but the first step is to ask yourself what decisions you want to make from this chart. That will help you dictate the chart type and sampling strategies.”
Control charts typically fall under three types. Let’s take a look at each.
Xbar and Range Chart
The most common type of chart for those operators searching for statistical process control, the “Xbar and Range Chart” is used to monitor a variable’s data when samples are collected at regular intervals. The chart is particularly advantageous when your sample size is relatively small and constant.
Individual-X Moving Range Chart
When it doesn’t make sense to take multiple readings, the “Individual-X Moving Range Chart” is the ideal option. This particular chart is used to monitor variables data when it is impractical to use rational subgroups. “When data is very expensive or there is a whole lot of time between samples, the concept of Xbar and Range makes no sense,” Wise explains. “It is better to go with Individual-X Moving Range.”
Xbar and Standard Deviation Chart
This chart is primarily used to show how much variation or "dispersion" exists from the average or expected value. The Xbar and Standard Deviation Chart is touted for helping manufacturers, engineers and operators understand variation better.
“The more you peel back as you start exploring the data stream, the more you need to refine the chart type you select,” says Wise.