Any suggestions on performing a gauge R&R study on a go / no go gauge. We do have MiniTab and it does this but how many parts, number of people, etc. should we use. :thanx:

Any suggestions on performing a gauge R&R study on a go / no go gauge. We do have MiniTab and it does this but how many parts, number of people, etc. should we use. :thanx:

Sean,

An advanced search on: "Attribute Gage Study" here at The Cove Forums, revealed this (http://www.google.com/custom?domains=Elsmar.com&q=Attribute+Gage+study&sa=Search&sitesearch=Elsmar.com&client=pub-1385417534940691&forid=1&channel=6124086287&ie=ISO-8859-1&oe=ISO-8859-1&cof=GALT%3A%23008000%3BGL%3A1%3BDIV%3A%23336699%3BVLC%3A663399%3BAH%3Acenter%3BBGC%3AFFFFFF%3BLBGC%3A000099%3BALC%3A000000%3BLC%3A000000%3BT%3A0000FF%3BGFNT%3A0000FF%3BGIMP%3A0000FF%3BLH%3A50%3BLW%3A350%3BL%3Ahttp%3A%2F%2Felsmar.com%2Fpng%2Fheader-G-search.png%3BS%3Ahttp%3A%2F%2FElsmar.com%2FForums%2F%3BFORID%3A1%3B&hl=en). Some assembly required...:)

Stijloor.

Any suggestions on performing a gauge R&R study ........ how many parts, number of people, etc. should we use. :thanx:
I just noticed this post, I apologize for not responding sooner. This is sort of a sample size problem. I suggest using all of the people that routinely use these gages, if possible. A gage R&R study is really a people Reliability study, its a matter of rearranging the data.

I also recommend using a staggered strategy. Start with a small number of parts. If the data are really great or really awful, stop the study. On the other hand, if the data shows something in between, the sample size would increase.

Of course, use the binomial distribution for the analysis.

Best wishes,

Tom