What is Gauge Capability? Dial Thickness Gage and Digital Vernier Calipers

[igsram233]

Recently we had an process audit audit by our potential customer.

While at the Quality Stage he wanted to show him how we measure the dimensions of the component. We showed him that we use Dial Thickness Gauges and Digital Vernier Calipers. He then asked us about something called Gauge Capability ( )and whether we had done that on the instruments. we did not know that.

We know a little bit about Process Capability (Cp & CpK).

I would like some one to help me in that.

Thanx in advance.

[Miner]

Your customer is probably referring to MSA, Measurement System Analysis. The evaluation of the Xbar chart (part of the MSA) for >50% of the points out of control is sometimes refferred to as Measurement Capability, but your customer is probably refferring to the total MSA.

Try searching within this forum for additional information. I also recommend obtaining the MSA manual from AIAG (Automotive Industry Action Group) at www.aiag.org.

[Atul Khandekar]

Also see:

MSA vs. Measurement Capability Analysis - What are the differences?

[Hershal]

I recommend you have the instruments calibrated by your local NABL-accredited calibration lab, and make sure you get the uncertainty reported.....then you will know just about everything about the instrument and its reported measurements.....then all you need is you appliction of the measurements.

Hershal

[ralphsulser]

Originally Posted by Atul Khandekar

Also see:

MSA vs. Measurement Capability Analysis - What are the differences?

Atul,
You have been very helpful about this subject. However, I must admit that GR&R has not been my forte. My question is what are the targets we should be looking for in the results of a GR&R using the ANOVA method? Such as what are good numerical scores for the different reported values? How do we determine what is acceptable and what is not, and by how much?
Someone else has been doing all these, but as previously mentioned we now have to report results a ANOVA. Appreciate your continued assistance.

[ralphsulser]

Originally Posted by ralphsulser

Atul,
You have been very helpful about this subject. However, I must admit that GR&R has not been my forte. My question is what are the targets we should be looking for in the results of a GR&R using the ANOVA method? Such as what are good numerical scores for the different reported values? How do we determine what is acceptable and what is not, and by how much?
Someone else has been doing all these, but as previously mentioned we now have to report results a ANOVA. Appreciate your continued assistance.

Anyone else care to chime in on this one? We have Ford coming in to scrutinize our results.

[Miner]

I may not understand the question completely, but if I do, the targets are the same for the ANOVA method as they are for the Range method.

That is, </= 10% is ideal, between 10% and 30% is acceptable depending on the criticality of the application and the economics of improving, and > 30% is not acceptable.

The results for each method should be very similar. The Range method is easier to use when performing calculations by hand, while the ANOVA method is more accurate and can separate the Operator x Part interaction. The procedure for collecting the data is the same, so if you have a computer and the software available, there is no reason not to use the ANOVA method.

The interpretation of the results is basically the same for each method.

[ralphsulser]

Originally Posted by Miner

I may not understand the question completely, but if I do, the targets are the same for the ANOVA method as they are for the Range method.

That is, </= 10% is ideal, between 10% and 30% is acceptable depending on the criticality of the application and the economics of improving, and > 30% is not acceptable.

The results for each method should be very similar. The Range method is easier to use when performing calculations by hand, while the ANOVA method is more accurate and can separate the Operator x Part interaction. The procedure for collecting the data is the same, so if you have a computer and the software available, there is no reason not to use the ANOVA method.

The interpretation of the results is basically the same for each method.

Thanks Miner, can you tell me what do the rest of these numbers tell me:
How do I interpret as OK, or NG and how much. 3 appraisers, 3 trials, 10 parts.
Two-Way ANOVA Table With Interaction

Source DF SS MS F P
Part 9 0.0000152 0.0000017 2.2443 0.069
Operator 2 0.0000422 0.0000211 28.0328 0.000
Part * Operator 18 0.0000136 0.0000008 0.9823 0.491
Repeatability 60 0.0000460 0.0000008
Total 89 0.0001170


Alpha to remove interaction term = 0.25


Two-Way ANOVA Table Without Interaction

Source DF SS MS F P
Part 9 0.0000152 0.0000017 2.2136 0.030
Operator 2 0.0000422 0.0000211 27.6493 0.000
Repeatability 78 0.0000596 0.0000008
Total 89 0.0001170


Gage R&R

%Contribution
Source VarComp (of VarComp)
Total Gage R&R 0.0000014 93.34
Repeatability 0.0000008 49.43
Reproducibility 0.0000007 43.91
Operator 0.0000007 43.91
Part-To-Part 0.0000001 6.66
Total Variation 0.0000015 100.00


Study Var %Study Var
Source StdDev (SD) (6 * SD) (%SV)
Total Gage R&R 0.0012007 0.0072045 96.61
Repeatability 0.0008738 0.0052428 70.30
Reproducibility 0.0008236 0.0049414 66.26
Operator 0.0008236 0.0049414 66.26
Part-To-Part 0.0003209 0.0019252 25.82
Total Variation 0.0012429 0.0074573 100.00


Number of Distinct Categories = 1
Gage R&R for Response

Appraiser 1 Trial 1 0.652 0.65 0.65 0.65 0.651 0.65 0.651 0.65 0.651 0.65
> Trial2 0.65 0.65 0.65 0.65 0.651 0.65 0.65 0.651 0.65 0.65
Olden Trial3 0.65 0.651 0.65 0.651 0.65 0.652 0.65 0.651 0.651 0.65
Trial4
Trial 5
Total 1.952 1.951 1.95 1.951 1.952 1.952 1.951 1.952 1.952 1.95
Average-Appraiser 1 0.65067 0.6503 0.65 0.6503 0.6507 0.6507 0.6503 0.6507 0.6507 0.65
Range1 0.002 0.001 0 0.001 0.001 0.002 0.001 0.001 0.001 0
Appraiser 2 Trial 1 0.653 0.651 0.65 0.649 0.652 0.651 0.65 0.652 0.65 0.651
here-> Trial2 0.65 0.651 0.652 0.651 0.65 0.651 0.652 0.652 0.65 0.651
Jamace Trial3 0.651 0.651 0.65 0.651 0.65 0.651 0.65 0.654 0.653 0.653
Trial4
Trial 5
Total 1.954 1.953 1.952 1.951 1.952 1.953 1.952 1.958 1.953 1.955
Average-Appraiser 2 0.65133 0.651 0.6507 0.6503 0.6507 0.651 0.6507 0.6527 0.651 0.6517
Range2 0.003 0 0.002 0.002 0.002 0 0.002 0.002 0.003 0.002
Appraiser 3 Trial 1 0.653 0.652 0.652 0.651 0.653 0.653 0.652 0.653 0.652 0.652
Trial2 0.65 0.652 0.652 0.651 0.653 0.653 0.652 0.653 0.652 0.651
Wanda Trial3 0.652 0.653 0.652 0.65 0.654 0.651 0.652 0.653 0.652 0.652
Trial4
Trial 5