In MSA, is it acceptable/valid to replace “operator” with “machine” for measurement systems that don't involve an operator? An example is where I want to check the repeatability and reproducibility amongst three separate but identical automated measurement systems.
Don - A lot depends on what the test criteria is for - machine, fixtures used on the machine, set up/qualification of the machine prior to measurement. MSA manual ( from AIAG ) while addresses basic criteria, falls short of many aspects that are brought forth in the world of automation. Softwares do come to our rescue, but than the user ( inputs ) are critical to process the information given!
Absolutely that is acceptable. Just make sure to document that in the report. Be sure to randomize the part order before each trial. If you are doing 3 trials, consider doing trial 2 on day 2 after a power- down/power-up cycle and maybe trial 3 is done in some other normal & acceptable condition (e.g. day 3 late evening). That will give you a strong look at Reproducibility.
@Don MacArthur
The terminology helps in discovering the idiosyncrasies of what we do - pointing to likely suspects. I once found reproducibility for a CMM MSA. Impossible - was the first thought… on further investigation I found that they were required to do an initial optical alignment setup… The rest is is history and the reassignment of how the reproducibility was accounted for…
@James Crescenzo
My preference is to do a calibration in such cases to understand the measurement uncertainty.
