My chemist collected MSA data using mostly his standard data collection process. The correct and logical model using Fit Model would be:
Product
Date[Product] & Random
Repeat[Product,Date] &Random
My chemist would really like to use one of the regular MSA platforms. But the nested models for them do not allow Date to be nested within Product.
Product should be a fixed factor. It is not a Part in the classical sense.
If I give into my chemist's demands and give him a regular MSA JMP platform result, the calculated metrics are wrong.
I would like the model choices available for the MSA platform to be more flexible. Allow for a fixed term that is not the classical "Part." Allow for more combinations of fixed and random terms in the model beyond just 3. This is a very simple example, but most real examples in my industry have (or should have!) more than 3 factors.
If the correct model was more often available in regular JMP MSA platforms, I could move my chemists and engineers away from Excel templates using the range method. Please, anything to get them away from that!