Quality teams use SPC because it stops two kinds of mistake: reacting to random noise as if it meant something, and missing a real shift until it becomes scrap. It is a standard tool in the control phase of DMAIC.
Where it comes from
Walter Shewhart developed the control chart at Bell Telephone Laboratories in the 1920s. W. Edwards Deming later taught SPC widely, including to Japanese industry after the Second World War.
At a glance
Noise or signal?
Swipe sideways to see the whole diagram.
How to apply Statistical process control
- 1
Choose what to measure
Pick a characteristic that matters to the customer or predicts quality downstream, such as a critical dimension, fill weight or cycle time.
- 2
Check the measurement system
Make sure the gauge is accurate and repeatable before charting. A noisy measurement hides the process.
- 3
Pick the chart
Use X-bar and R or individuals charts for measured data, and p, np, c or u charts for counts of defects or defectives.
- 4
Set control limits
Calculate the center line and control limits from a stable period of data, usually three standard deviations either side of the mean. Control limits come from the process, not from the specification.
- 5
Watch for signals
Look for points outside the limits and for patterns such as runs and trends, using a set of rules such as the Western Electric rules.
- 6
Respond and record
When a signal appears, stop and find the special cause, fix it and note what happened on the chart. Leave the process alone when there is no signal.
Worked example
A filling line charts the weight of five bottles every hour. Most points sit within the limits, so operators leave the filler alone. One morning seven points in a row fall above the center line. The team checks the filler and finds a worn valve seal, replaces it before any bottle goes out of specification and notes the cause on the chart.
Common mistakes
- Using specification limits as control limits, which hides whether the process itself is stable.
- Adjusting the process after every point, which adds variation instead of removing it.
- Charting data nobody reviews, so signals appear but nobody acts.
When Statistical process control is not the right tool
SPC needs a stable, repeating process and enough data points. For one-off jobs or very low volumes, a checklist or first-article inspection may be more practical.
SPC and process capability
A control chart shows whether a process is stable. Process capability shows whether that stable process fits inside the customer's specification, using indices such as Cp and Cpk.
Check stability first. Capability figures from an unstable process do not predict what it will do next. The sigma level calculator turns defect counts into a sigma level.