A case study: Eliminating nuisance within-part variation in assessing a measurement system

Abstract

In this article, we consider a gauge R & R study in which a part measured in production is randomly placed in the measuring device. In assessing a measurement system, one does not want a possible within-part variation included in the estimated gauge variation and we propose a way to eliminate it. We consider a pellet measurement system and demonstrate the benefits of eliminating within-part variation in its assessment.

Keywords:

• 4:1 rule
• gauge R & R study
• measurement system assessment
• repeatability
• reproducibility
• TUR (test uncertainty ratio)

Acknowledgement

We thank C. C. Essix for her encouragement and support. We also thank two anonymous referees whose insightful comments on an earlier version helped to improve the exposition of this article.

Additional information

Notes on contributors

M. S. Hamada

Michael S. Hamada is a Scientist and holds a PhD in Statistics from the University of Wisconsin-Madison. He is a Fellow of the American Statistical Association, the American Society for Quality, and Los Alamos National Laboratory. His research interests include design and analysis of experiments, measurement system assessment, quality control, and reliability.

B. W. O’Brien

Brendan W. O’Brien is an R & D Engineer with a BS in Mechanical Engineering from New Mexico Institute of Mining and Technology. His research is focused on automation of non-touch dimensional measurement and glue dispensing processes.