Abstract
The first step of many process improvement projects is to conduct a repeatability and reproducibility (R & R) study. The major objective of such a study is to determine whether a measurement procedure is adequate for monitoring a process. If the measurement error is small relative to the process variation, then the measurement procedure is deemed “adequate”. It is common practice to treat all experimental factors in an R & R study as random effects. However, in many applications, the nature of the measurement procedure and the protocol used to conduct the study make the assumption of fixed effects more appropriate for some factors. In this paper, we consider a two-factor experiment in which one factor is fixed and one factor is random. Confidence intervals for parameters of interest are derived for this model and evaluated using computer simulation.
Additional information
Notes on contributors
Kristynn K. Dolezal
Kristynn K. Dolezal is a Senior Materials Quality Engineer.
Richard K. Burdick
Dr. Richard K. Burdick is Professor of Statistics in the College of Business. He is a Member of ASQ.
Nancy J. Birch
Dr. Nancy J. Birch is Associate Professor of Decision Sciences in the College of Business and Public Administration.