How to find out which formula was used if there are different formulas available based on different conditions?

[#1]

Hi all,

I'm wondering if there's a way to find out which exact formula JMP used to calculated a certain value if there are several formulas available depending on different conditions.
For example, calculation of Cpk and Ppk values: Depending on whether the data is normally distributed or not, different formulas are used for their calculation (see links of JMP documentation below). And again for non-normal distributions, it depends on a certain condition which of the two formulas is finally used for Cpk and Ppk calculation. So is there a way to quickly trace back what JMP did to calculate Cpk and Ppk - without calculating it on my own?

Thanks for your help!

Cpk / Ppk for normal distributions: https://www.jmp.com/support/help/en/17.2/index.shtml#page/jmp/statistical-details-for-capability-ind...
Cpk / Ppk for non-normal distributions: https://www.jmp.com/support/help/en/17.2/index.shtml#page/jmp/statistical-details-for-capability-ind...

Note: I using JMP 18

[#2]

Where do you have your Cpk/Ppk values? For example in Process Capability (from distribution platform) it seems to be visible here

[#3]

Thanks for your reply. I use Cpk/Ppk calculation via the distribution platform (like you showed above) and via the process screening platform. What you showed in your example above is exactly what I was looking for. Unfortunately, I don't have this information if I run it on my side. I guess I have to change my preferences for the distribution platform

[#4]

Have you told JMP to use something else than normal distribution?

[#5]

I just resetted the preferences of the distribution platform to default (I expect that JMP uses then the normal distribution) and, sadly, I still don't get the information which methods JMP uses to calculate the capability indices. But thanks for the suggestion!

[#6 · Accepted solution]

I would guess if you told JMP to use non-normal distribution it will tell what method was used. Otherwise it will assume normal distribution.

Using normal distribution (or no changes to distribution)

Told JMP to use nonnormal distribution

Normal is still visible on the left

From Process Screening -> Launch Process Capability -> Compare Distributions -> Selected as Mixture of 2 Normals (visible on top)