Interpreting operator-part interaction in gage repeatability and reproducibility studies

Abstract

Measurement has been an important activity for centuries. The advent of Six Sigma initiatives has increased the assessment of measurement repeatability and reproducibility using Gage R&R studies. It is common practice that, when the operator-part interaction in Gage R&R studies is statistically significant, the associated variance component is used in estimating the test method reproducibility. In this article it is shown that the Operator-Part interaction may be due to systematic behavior of one or more operators thereby unnecessarily increasing the test method reproducibility variance. Several examples are included to demonstrate how the analysis is performed and how the results can be interpreted.

Keywords:

• Measurement variability
• gage R&R
• operator-part interaction
• repeatability
• reproducibility
• six sigma

Acknowledgment

The author is pleased to thank the Editor and Referee for their helpful comments and suggestions which improved presentation of this article.

Additional information

Notes on contributors

Ronald D. Snee

Ronald D. Snee is Founder of Snee Associates, LLC, a firm dedicated to the successful implementation of process and organizational improvement initiatives, using Quality by Design, Continued Process Verification and other databased improvement approaches that produce bottom line results. Prior to entering the consulting field Ron worked at DuPont for 24 years in a variety of assignments including eight years in pharmaceutical development. He also serves as an Adjunct Professor in the Pharmaceutical programs at Temple and Rutgers Universities. Ron is an Honorary Member and Fellow of American Society for Quality. He is a Fellow of the American Statistical Association and has received ASA’s Deming Lecture, Dixon Consulting Excellence and Hahn Quality and Productivity Excellence Awards. Ron is also a Fellow of the American Association for the Advancement of Science and Academician in the International Academy for Quality. He has published eight books and more than 330 articles in the quality, improvement and statistics literature. Ron received his BA from Washington and Jefferson College and MS and PhD degrees from Rutgers University.