SPC Control Charts

All control charts have three common elements:

Plot points: Plot points usually represent individual measurements, averages, standard deviations, or ranges.
Centerline: The centerline is usually (but not always) the average of the points plotted on the chart.
Control limits: Control limits represent the amount of variability in the process.

The Four Foundations of Shewhart’s Control Charts

There are four foundational guidelines to Shewhart statistical process control charts:

Shewhart statistical process control charts always use control limits that are set to 3 sigma units on either side of the central line. These 3-sigma limits are always based on the chart’s data and define when action should be taken on the process. Control limits are never based on any calculation using the specification limits. Specification limits are the customer requirements and define how to treat the product, not the process.
Always use an average dispersion statistic or a median dispersion statistic when computing 3-sigma control limits. Using the average or median of several dispersion statistics increases the robustness of the chart.
The conceptual foundation of Shewhart’s charts is the notion of rational sampling and subgrouping.
Control charts are effective only to the extent that the organization can effectively use the knowledge gained to take action.