MSA on calculated values

Hello everyone,

First post here, so be gentle on me. :)

I was wondering something and I couldn't find the answer anywhere on the internet.

This is the situation. In our plant we perform certain Control Plan measurements on our products.
For a certain imaginary dimension we measure samples on width. Let's say the tolerance for this dimension is 100±5 mm and we measure this dimension over the length of the product several times. For this measurement method we have done a MSA and the GageR&R result is just below 10% = OK.

For a secondary dimension, we calculate the variation of the width over the length of the product. This is being done by calculating the range (Xmax - Xmin) of the all the measurements of that first dimension. The tolerance here is 0±0,2 mm.

Is it correct / useful to calculate the MSA for this secondary dimension?
If so, should we then use the specification of the secondary dimension or not (since it is a calculation)?

With best regards,


TS Contributor
It would be correct to calculate the R&R on the secondary condition using its specification.. While it is calculated, it is still dependent on the measurements used in the calculation.

This forum tends toward statistics used in the social sciences. There are very few of us that are involved in industrial statistics and quality. I recommend the following forums that specialize in quality. These are the Elsmar Cove and Quality Forum Online.