Implement methods of Round Robin test

[zlcareer214]

Hello Everyone,

I don't know if my question is proper to state here, but this is the most relevant place I could find...

Our organization is going to use Round Robin test to detect the difference or degree among our several testing labs located in different places, is it possible that somebody can share some guidance on how to do Round Robin test or some example to show this activity to me for study? (I tried to search this topic within this Forum, and got some reference, but I couldn't find a systematical introduction to this tool...)

Thanks very much.

 

[zlcareer214]

From my understanding to some examples I could find, it seems just like to test several samples for several times in several labs... and make some matrix from different aspects... then calculate the statistical parameters to figure out some labs fitting better to the calculated ideal parameters...
But I guess a typical Round Robin test should not be so simple... Wish somebody could kindly help me to better comprehend it. Thanks very much!

 

[Miner]

This is probably the best location in which to post. You have a few options:

• There are several ISO standards that address this:
  • ISO 17025 Interlaboratory Comparison
  • ISO 17043 Proficiency Testing.
• There is also a Bland-Altman approach for analyzing differences between labs.
• And if you are familiar with the AIAG R&R methodology, you can add laboratories as an additional factor in an expanded R&R study. If you go this route, use ANOVA to analyze and focus on the percent contribution statistic to evaluate the results.

 

[zlcareer214]

This is probably the best location in which to post. You have a few options:

• There are several ISO standards that address this:
  • ISO 17025 Interlaboratory Comparison
  • ISO 17043 Proficiency Testing.
• There is also a Bland-Altman approach for analyzing differences between labs.
• And if you are familiar with the AIAG R&R methodology, you can add laboratories as an additional factor in an expanded R&R study. If you go this route, use ANOVA to analyze and focus on the percent contribution statistic to evaluate the results.

Thanks very much for your replying, Miner, I believe that you may offer some support to me, and thanks for your reminding of several approaches to my problem, these should definitely be my options as back up.
I have to say, one more difficult on my side, the big boss in my company (a German chemical enterprise) are fetishistic in Round Robin test and appointed that we should use this tool (An old colleague used to implement this tool for several cases in the past but retired without successor last year), so I still have to put Round Robin test as my first choice...
Thanks!

 

[Miner]

Round robin testing does not refer to a specific methodology. It is an old and generic name used to refer to the general concept of having multiple labs test the same parts. All of the approaches that I mentioned would be considered Round Robin test methods.

 

[zlcareer214]

Round robin testing does not refer to a specific methodology. It is an old and generic name used to refer to the general concept of having multiple labs test the same parts. All of the approaches that I mentioned would be considered Round Robin test methods.

Thanks very much Miner, and now I'm clear. Really thanks for your patient explanation very much!