Metrology Monday! #101 A Discussion on Conformity Assessment, Decision Rules and Measurement Decision Risk – Suggestions for further reading

After 22 weeks of discussing Conformity Assessment, Decision rules and Measurement Decision Risk, I am ready to close this topic out and move on.  You have now been introduced to this topic, but there is always much more to learn about.  Today I will be sharing the references that I used for this topic and a few other good documents. I will also be sharing a few last thoughts.

The first set of documents I would strongly encourage you to read are from my predecessor, David Deaver from Fluke.  Some of the philosophy around documentary standards are a bit dated, but the mathematics behind the analyses are very solid.

“How to Maintain Your Confidence (In a world of declining Test Uncertainty Ratios)” 1993 NCSL International Conference

“Guardbanding with Confidence” 1994 NCSL International Conference

“Guardbanding and the world of ISO Guide 25, Is there more than one way?”  1998 NCSL International Conference

“A Study of and Recommendation for Applying the False Accept Risk Specification of Z540.3” from the 2010 Measurement Science Conference

I would also recommend reading the Handbook for the interpretation of ANSI/NCSL Z540.3-2006” available from NCSL International, as this standard drove a lot of discussion on the topic.  The guidance document describes several of the methods I have spoken about in detail.

Of course it is great to read the original papers from Jerry Hayes, “Technical Memorandum 24 October 1955 No. 63-106 Factors Affecting Measurement Reliability” and the papers he drew inspiration from, “A Method for Handling Errors in Testing and Measuring” Alan Eagle and “On Setting Test Limits Relative to Specification Limits” by Frank E. Grubbs and Helen J. Coon, both from Industrial Quality Control Magazine March 1954

Another one of the people I greatly admire in this field, Howard Castrup wrote “Risk Analysis Methods for Complying with Z540.3” which he presented at the NCSL International conference in 2007.  He also created the freeware program RiskGuard.

From NCSLI 2008, Michael Dobbert’s paper “A Guard-Band Strategy for Managing False Accept Risk” is absolutely brilliant. 

While I am still not entirely sold on it, I would encourage you to read “Using Reliability to Meet Z540.3’s 2% Rule” by Scott Mimbs from the 2011 NCSLI conference.

Colin Delker from Sandia National Laboratories has done some great contemporary work, and I highly recommend these papers.

“Decision Risk and Guardbanding of One-Sided Tolerances” NCSLI Conference

“Evaluation of Guardbanding Methods for Calibration and Product Acceptance” Sandia Report SAND2021-8069

“A Guardbanding method for managing false accept risk under process bias” NCSLI Conference

While it is a challenging read, one of the most important references is JCGM 106:2012 “Evaluation of measurement data – The role of measurement uncertainty in conformity assessment” from the BIPM.

That is most of the references that I have really learned from for this subject.  If you have a favorite reference on this topic, please share it in the comments (please don’t advertise books or references that you sell).

CONCLUDING THOUGHTS

There are a lot of different ways to solve this problem, and a lot of different rules that you can apply, and that is OK.  This is a relatively new science, and we are unifying many fields of metrology that may not have spoken to each other before, and each has solved the problem differently.

So long as you seek to understand the risk that your customer needs, and then apply decision rules and practices based on the amount of risk required, you will be doing the right thing. #MetrologyMonday #FlukeMetrology