% of Contribution or % of Study Variation? Gage R&R using Minitab

Hi all

By Using Minitab, when GR&R study is conducted (crossed method), what are the conditions when % of contribution or % of study variation is a more appropriate metric to use?
Thanks in advance

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

% of study variation is %GRR. This is the appropriate metric to use to determine whether a gage is suitable for use in process control, or in continual improvement (hypothesis test, DOE's, etc.).

% contribution is a tool, not a metric. Its only purpose is to point you to the parameter (i.e., repeatability, reproducibility, operator x part interaction, etc.) that contributes the most to the measurement variation. That allows you to focus on the correct parameter to reduce measurement variation.

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

In that case why Minitab displays a Total GR&R % in % Contribution?
It seems just to be the addition of EV and AV.. not the square root of (EVsqure + AVsquare). am I right?

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

Minitab bases the % Contibution on the variances (VarComp), which are additive. The variance (VarComp) of the repeatability is EVsquared. The variance of the reproducibility is AVsquared.

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

Miner said:

% of study variation is %GRR. This is the appropriate metric to use to determine whether a gage is suitable for use in process control, or in continual improvement (hypothesis test, DOE's, etc.).

% contribution is a tool, not a metric. Its only purpose is to point you to the parameter (i.e., repeatability, reproducibility, operator x part interaction, etc.) that contributes the most to the measurement variation. That allows you to focus on the correct parameter to reduce measurement variation.

Click to expand...

But why? Acceptance criteria is typically set at <30% of tolerance and/or <30% of study variation. But I'm struggling with why you use the %SV (for minitab) which is calculated using 5.15*Stdev rather than using the % calculated from the actual variation components (%contribution). Wouldn't the %contribution give you the actual % of variation for R&R compared to total variation in your system. Common sense is telling me that I should use my %contribution to see if it is less than 30%. Why am I wrong according to minitab?

I'm missing something as to why you use the % calculated from the stdev rather than the % calculated from the actual variation numbers.

Thanks for the help.

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

%Study variation is directly comparable to the units of your data. Being based off of standard deviations.

%Contribution is calculated from variances and is comparing the variation as a units squared term.
(poor explanation by myself I know)

They show the same thing, but contribution should use <1% ideal, 1%-9% as marginal >9% unacceptable to give the same results as the %study variable guidelines of <10%, 10%-30% and >30%.

5.15 * Std dev was used way back in release 12 and you should find in later versions that Minitab is using 6* Std dev (by default) to stay up to date with current guidelines from the AIAG.

Still %Study variation only reflects how much of the variation in your study is accounted for by the Gage R&R. Which is heavily dependant on obtaining a random sample that represents the process you are measuring.

%tolerance and %Process can also be useful measures here in assessing measurement system variation.

Edit:
Actually that 5.15 term doesn't matter in a %Study Variation calculation as it is included both top and bottom of the divide. Its more important on %tolerance.

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

If I have a GR&R study that shows

Total Gage R&R %Study Variation %Tolerance
(%SV) (SV/Toler)
79.77 6.10
did it fail?

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

Laura O said:

If I have a GR&R study that shows

Total Gage R&R %Study Variation %Tolerance
(%SV) (SV/Toler)
79.77 6.10
did it fail?

Click to expand...

Welcome to the Cove.

I think maybe your data didn't get aligned properly when it was posted. Could you perhaps post the whole report, including the data? It's hard to tell what might be going on without seeing the whole thing.

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

I don't have the data. I am trying to get it. The document was sent as a picture in Powerpoint. Is there a way to upload it?

What I guess i am asking is, when you look at the report, do you use the (%SV) number or the (SV/Toler)? (I know just enough to be dangerous).

Re: % of Contribution or % of Study Variation? Using Minitab - Gage R&R

Laura O said:

I don't have the data. I am trying to get it. The document was sent as a picture in Powerpoint. Is there a way to upload it?

What I guess i am asking is, when you look at the report, do you use the (%SV) number or the (SV/Toler)? (I know just enough to be dangerous).

Click to expand...

The number to be used depends on what it's being used for. You can attach a file (such as a Powerpoint slide) to a post here. When you open the window for a new post, scroll down to where you'll see "Manage Attachments." Hit that button and a window for uploading will open.