Ford Gage R&R - Required Number of Appraisers per MSA

[Angelo]

Gage R&R

I need clarification as to the required # of appraisers per MSA. I'm getting hassled by some customers (particulary STA's at Ford) for submiting R&R studies with only 2 appraisers, 2 trials. Is there any documentation saying that 3 appraisers and 3 trials are required for a study?

 

[JaySturgeon]

I need clarification as to the required # of appraisers per MSA. I'm getting hassled by some customers (particulary STA's at Ford) for submiting R&R studies with only 2 appraisers, 2 trials. Is there any documentation saying that 3 appraisers and 3 trials are required for a study?

Angelo,

Refer to pages 99, 100 and 101 of the MSA Manual, Third Edition

 

[Atul Khandekar]

Generally, the more data you have, the closer your results will be to reality and there will be better confidence in the results. As a rule of thumb, you should generate at least 15 operator 'ranges' in the GRR Study for a minimum level of statistical confidence in the results. From the Range-average calculation sheet, you will observe that each operator generates one range for each part for his/her trials. So with 5 parts, 2 operators and 3 trials, 10 ranges are generated. With 10x3x3 scheme, 30 ranges are generated.

To quote from the manual:
As with any statistical technique, the larger the sample size, the less sampling variation and less resultant risk will be present.

Hope this helps..

 

[TommyT - 2003]

Re: Gage R&R

They are citing the Ford Customer-Specific requirements released July, 2003, sect. 4.35 says variable gage studies SHOULD utilize 10 parts, 3 operators, 3 trials. Since it says should, there's some lattitude. If you can show good reason why your way should be acceptable, ie, there's only two guys making parts, you SHOULD be allowed to do what's the best business method to ensure the quality. If not, welcome to the "Raised the Bar" club & do it their way.

 

[howste]

Re: Re: Gage R&R

TommyT is right. Here are the exact words in the Ford CSR:

3.16 Should
Indicates a mandatory requirement with some flexibility allowed in compliance methodology. Organizations choosing other approaches to satisfy a "should" must be able to show that their approach meets the intent of ISO/TS 16949:2002

4.35 Measurement systems analysis (ISO/TS 16949:2002 cl. 7.6.1)
Variable gauge studies should utilize 10 parts, 3 operators and 3 trials.
Attribute gauge studies should utilize 50 parts, 3 operators, 3 trials.
Effective attribute gauge study samples include parts within specification and parts outside specification for each criterion being measured and within the expected range of manufacturing variability.