Common cause variation comes from the combined effect of many small, ever-present sources: ambient temperature, raw material spread, gauge noise, small tool-to-tool differences. A process showing only common cause is in statistical control and predictable within its limits. Reducing it takes a change to the system itself, such as a better recipe, tighter incoming material or new equipment, not a reaction to individual points.
That's common cause, it's bounced like that for six months. If you want it tighter, it's a recipe project, not an OCAP.
Also heard as
- chance cause
- random variation
- noise
Do not mix it up with
- Special cause: Special cause variation comes from a specific, identifiable event and shows up as a chart signal; common cause is the baseline noise of the system.
Related terms
Special cause
Specific, identifiable source of variation that is not part of the process's normal behavior and shows up as a chart signal.
Tampering
Adjusting a stable process in reaction to common cause variation, which makes the output worse rather than better.
SPC
Statistical process control: using control charts and run rules to tell routine process noise from real changes.
Process capability
How well a stable process's natural spread fits inside the specification limits, usually expressed as an index.