I suspect you can run it as a regular attribute study, where you have parts that are Good, Rework or Bad, and you report Pass if the operator attributes the correct characteristic to the specimen, Fail if they misinterpret it.
As I just posted in response to teh thread concerning call center Gage R&R, start with the article "When Quality Is a Matter of Taste, Use Reliability Indexes" by David Futrell. If you google it you will find a free copy. It handles multiple judgement outcomes. I have also attached a spreadsheet that will do the math for you.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to the use of cookies.