Common-Cause and Special-Cause Variation
SPC separates common-cause variation from special-cause variation. This distinction decides whether to improve the system or investigate a specific change.[1]
Prerequisites
Prerequisites: control chart basics.
Definition
Common-cause variation is the routine variation produced by the current stable system. Special-cause variation is evidence that the current point or pattern likely came from a different source than the stable system used to set the limits.
Interpretation
| Variation type | Meaning | Proper response |
|---|---|---|
| Common cause | Predictable process noise | Improve the system: method, equipment, material, training, measurement |
| Special cause | Assignable change or unusual event | Investigate, contain, correct, and prevent recurrence |
Worked Example
One filling-process bottle volume sits slightly above the center line but inside limits.
That is not enough evidence for adjustment.
Later, one point exceeds the UCL after nozzle replacement.
That special-cause signal should trigger the out-of-control action plan.
Common Mistakes
- Adjusting the process after every common-cause fluctuation.
- Ignoring a point outside limits because the product still meets specification.
- Calling all variation "random" and skipping investigation.
- Treating stable but off-target output like acceptable output.
Connections
| Related note | Use |
|---|---|
| Control charts | Signal detection |
| Control Limits and Specification Limits | Limit distinction |
| Quality tools | Cause investigation |
| Process capability | Stable-process capability |
References
NIST/SEMATECH, e-Handbook of Statistical Methods, "What to do if the process is Out of Control?", https://www.itl.nist.gov/div898/handbook/pmc/section1/pmc14.htm ↩︎