What is the difference between Cp, Pp, Cpk and Ppk?
In the CNC machining world there is a fair amount of confusion as to how Cp, Pp, Cpk and Ppk are calculated. The best way to think about this is that all of these measures represent how controlled your production process is, and for Cpk and Ppk, as “weighted” by how close your process is to meeting your customer’s expectations for parts with the right dimensions. Sounds confusing, right? Let’s take a look at the following 3 targets for a visual explanation.
Shots vary widely – Cp, Cpk, Pp, and Ppk will be low (lots of variation, not on bullseye)
Shots grouped closely, but not on the bullseye – Cp and Pp will be high, but Cpk and Ppk will be low (little variation, but not on the bullseye).
Shots tightly grouped and in the bullseye – Cp, Cpk, Pp, and Ppk will all be high. This is what you want!
The difference between the two “k” indices is in how the variation in your production run is calculated. In the case of Cpk, the measure of variation is considered separately for each subgroup. (For example, if you selected 10 parts at random to check each day, each day could be considered a subgroup). This is in contrast with the Ppk calculation, which is based upon the variation across the entire production run.
Hint: Think of Cpk and Ppk as measures of how well you are actually performing, as opposed to Cp and Pp which tell you theoretically how well you could do (if you could hit the bullseye).
The advantage to Cpk is that it is more sensitive to variations during the production run, meaning that you can identify and correct sources of variation as they happen.
If Cpk and Ppk are similar and greater that one, your production run is likely in statistical control. However, if there are large differences between the two measures, time for your QA team to figure out why!
For further reading I suggest the following: