MSA manual tells nothing on that.

can anybody help me whrn 30%>R&r >10%?

I read many web-pages but no idea.

MSA manual tells nothing on that.

can anybody help me whrn 30%>R&r >10%?

I read many web-pages but no idea.
AIAG provides guidelines for dealing with GR&R results (it's expected that results >10% and <31% will be reviewed for appropriateness), but what the results mean in any given situation must be determined locally, on a case-by-case basis. To give an example, if R&R is 20% of tolerance, but the tolerance is wide and the process is only using a small percentage of it, 20% might be perfectly reasonable. On the other hand, with a tight tolerance and most of the tolerance being consumed by the process, 10% could be too much. MSA provides data for decision-making, not necessarily clear-cut limits.

thanks ur reply.

does that mean the tolerance is the only reference? is there any other?

Can anybody share me your cretira under such situation?

I am annoyed with that.

thanks ur reply.

does that mean the tolerance is the only reference? is there any other?

Can anybody share me your cretira under such situation?

I am annoyed with that.
R&R may also be calculated as percent of variation; this may be within-study variation (where only the variation of the parts used in the study is taken into account) or process variation, where the total variation of the process, if known, is "plugged in" to the calculations. Again, its a matter of choosing what's most appropriate for the situation. You shouldn't do MSA because your customer expects it; you should do it because you want to know what effect measurement error is having on variation.

R&R may also be calculated as percent of variation; this may be within-study variation (where only the variation of the parts used in the study is taken into account) or process variation, where the total variation of the process, if known, is "plugged in" to the calculations. .


How can I statistically prove this calculation you are talking about here. I do want to make a decision based on the total variation and not on the basis of the tolerance .

I want to show that even though I have a not so good %GRR based on total variation calculation,,,how can I say that the gage is going to perform well statistically. Any example.

Thanks in advance Jim.

Page 77 of the msa manual, at the bottom, it pretty much says there is nothing from you and your customer agreeing to use a gage if it meets what they will accept. So talk to your customer and explain why you need to use it.

How can I statistically prove this calculation you are talking about here. I do want to make a decision based on the total variation and not on the basis of the tolerance .

I want to show that even though I have a not so good %GRR based on total variation calculation,,,how can I say that the gage is going to perform well statistically. Any example.

Thanks in advance Jim.

I'm not sure what calculation you're referring to. Whether or not GR&R results should be considered acceptable is a function of (A) customer requirements and (B) the application. You need to be able to evaluate the GR&R output in terms of risk. If you have some actual data--an example GR&R report--please post it and we'll be able to give more specific advice.