What data do you need?
Prepare numeric measurements, lower and upper specification limits, and meaningful subgroup information when available. Specification limits come from customer or engineering requirements; they are not control limits.
Worked example
Suppose LSL = 9, USL = 11, mean = 10.2 and within-subgroup standard deviation = 0.2. Cp = (11 − 9) / (6 × 0.2) = 1.67. Cpk = min((11 − 10.2) / (3 × 0.2), (10.2 − 9) / (3 × 0.2)) = 1.33. The lower Cpk reflects the mean being closer to the upper limit.
How to use this Six Sigma tool
- Inspect time-ordered measurements using control charts.
- Enter measurements and specification limits in process capability analysis.
- Compare Cp/Cpk with Pp/Ppk and review assumptions before reporting results.
How to interpret the results
Check process stability before interpreting capability. Normal-distribution estimates need appropriate distribution assumptions. An index threshold such as 1.33 is a common benchmark, not a universal acceptance rule; follow the requirement for your process.
What is the difference between Cp and Cpk?
Cp compares the specification width with six within-process standard deviations. Cpk uses the distance from the mean to the nearer specification limit. A high Cp with a lower Cpk suggests an off-center process.