Pp and Ppk in Simple Terms
Pp asks: if the process were centred, would its total spread fit inside the tolerance? Ppk asks the same question using the nearer specification limit, so an off-centre process scores lower.
The difference from Cp and Cpk is the standard deviation used. Cp and Cpk use within-subgroup variation, which shows short-term potential. Pp and Ppk use overall variation across all data, which includes shifts between subgroups.
How They Are Calculated
- Pp = (USL minus LSL) divided by 6 times the overall standard deviation.
- Ppk = the smaller of (USL minus mean) and (mean minus LSL), divided by 3 times the overall standard deviation.
| Index | Variation used | Typical use |
|---|---|---|
| Cp, Cpk | Within subgroups | Ongoing capability of a stable process |
| Pp, Ppk | Overall, all data | Initial process studies and PPAP |
Example
A shaft diameter has a tolerance of 20.00 to 20.10 mm. A run of 125 parts gives a mean of 20.06 mm and an overall standard deviation of 0.008 mm. Pp is 0.10 divided by 0.048, about 2.08. Ppk uses the upper side: 0.04 divided by 0.024, about 1.67. The process is capable but running slightly high.
Why It Matters
The AIAG PPAP manual uses Ppk for initial process studies, with a common acceptance target of 1.67 or higher, and many customers set their own values in customer-specific requirements. A large gap between Cpk and Ppk is a sign of special causes, so check the control chart first. OperaHive SPC software calculates capability from the data collected on the line. Read more in our SPC and Cpk guide.