Crossed Gage R&R on a Extruded Tube

Q

QE

Hello Fellow Coves;

I am working on a GRR with one of the vendors. It turns out that the GRR on the Outer Diameter failed as the P/T ratio was greater than 30%. In this case it was 53%. The review of the material I had indicated that I should calculate a probability of misclassification but I am not seeing much hopes. Can someone take a look and advise how to proceed please.

Probabilities of Misclassification

Joint Probability
Part is bad and is accepted 0.000
Part is good and is rejected 0.000

Conditional Probability
False Accept *
False Reject 0.000
* WARNING * Probability of part within spec limits = 1.

Attachment:
• Raw Data in Excel Sheet
• Minitab File with the Analysis

Thanks
QE
 

Attachments

  • Analysis.xlsx
    10.2 KB · Views: 192
  • Minitab.MPJ
    200 KB · Views: 91
Q

QE

Hello Fellow Coves;

I am working on a GRR with one of the vendors. It turns out that the GRR on the Outer Diameter failed as the P/T ratio was greater than 30%. In this case it was 53%. The review of the material I had indicated that I should calculate a probability of misclassification but I am not seeing much hopes. Can someone take a look and advise how to proceed please.

Probabilities of Misclassification

Joint Probability
Part is bad and is accepted 0.000
Part is good and is rejected 0.000

Conditional Probability
False Accept *
False Reject 0.000
* WARNING * Probability of part within spec limits = 1.

Attachment:
• Raw Data in Excel Sheet
• Minitab File with the Analysis

Thanks
QE
A Quick check, can any one please review this and provide some guidance ?
 

Miner

Forum Moderator
Leader
Admin
The probability of misclassification is dependent on several things. The %R&R, the capability and the location of the distribution within the specs.

In your scenario, while you have a very poor R&R, you have respectable capability and appear to be fairly center within the specs (if I interpreted your data correctly). This means that as long as you stay centered you have low risk of misclassifying the product. However, this could change. If your process drifts closer to the tolerances, the probability of misclassification will begin to increase dramatically.
 
Top Bottom