107378_en-US

liliconcom

Process Performance Graph and accommodating non normal data distributions

I see from Zora Mlejnkova excellent presentation (https://community.jmp.com/t5/Discovery-Summit-2017/Process-Capability-Analysis-for-Non-Normal-Processes-With-JSL/ta-p/44065) that the Process Performance Graph assumes that the data is normally distributed (see below "NOTE: Currently works with normal distribution only !!".  The presentation was in 2017, can other distributions be "automatically" used to determine the Cp/Cpk and Pp/Ppks within this platform?

 

 

 

 

Re: Process Performance Graph and accommodating non normal data distributions

As you have already discovered, the Process Performance Graph is for normal distributions only.  The Process Screening platform has no method of fitting other distributions.  You can use the Process Capability platform to fit non-normal distributions and obtain the appropiate capability indices.