Abstract
In this article, we consider experimental designs for gauge R & R studies used to assess measurement systems. We present standard full factorial designs with long forgotten block designs and compare them via a simulation study for various run sizes. The simulation study shows that the block designs which accommodate more parts and especially more operators provide better inference even though their run sizes are the same as the full factorial designs.
Acknowledgements
We would like to thank two anonymous reviewers for their insightful comments on an earlier version that helped to improve the exposition of this article.
Disclosure statement
No potential conflict of interest was reported by the author.
Supplemental material
The Supplement presents the simulation study results for all the designs listed in Section 3.
Additional information
Notes on contributors
M. S. Hamada
Michael Hamada is a retired Scientist from Los Alamos National Laboratory (LANL). He earned a Ph.D. in Statistics from the University of Wisconsin-Madison. He is a Fellow of LANL, ASQ, and ASA. His research interests include design of experiments, reliability, quality control, and measurement system assessment.