cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
Pops
Level I

cpK/ppk

Is it correct to calculate cpk when I have no subgroup in the process data (grouping=1), meaning the sampling is continous, like tablet weight taken one after the other in one lot? How is cpk calculated in this case (the withingroup variation)?

1 ACCEPTED SOLUTION

Accepted Solutions
Peter_Bartell
Level VIII

Re: cpK/ppk

Correct? That's a different question. The 'correctness' of an approach is dependent on many aspects to the process capability analysis problem. For example: 1. Does your process data exhibit any evidence of assignable cause variation? 2. Does your data appear to be normally distributed? The answer to these questions influences the approach that is optimal for calculating process capability indices over and above the sheer mathematics associated with the reference in my first reply. 

View solution in original post

3 REPLIES 3
Peter_Bartell
Level VIII

Re: cpK/ppk

Here's a reference for you in the JMP Online Documentation.

 

http://www.jmp.com/support/help/13/Capability_Analysis_2.shtml#1161749

 

As of January 2018, use this link:

http://www.jmp.com/support/help/13-2/Capability_Analysis_2.shtml#1161749

 

 

Peter_Bartell
Level VIII

Re: cpK/ppk

Correct? That's a different question. The 'correctness' of an approach is dependent on many aspects to the process capability analysis problem. For example: 1. Does your process data exhibit any evidence of assignable cause variation? 2. Does your data appear to be normally distributed? The answer to these questions influences the approach that is optimal for calculating process capability indices over and above the sheer mathematics associated with the reference in my first reply. 

Pops
Level I

Re: cpK/ppk

Thank You

Data are normal distributed with no assignable cause variation. I assume the answer I was looking for (found in Process Capability in JMP 2012) is that if I do not group data JMP will automatically use moving range subgroup size 2 to calculate cpk.

KR