Group for Research in Decision Analysis


A Formal Test for Binary R&R Measurement Systems

, , and

Measurement systems studies are an integral part of most quality improvement processes. A desirable measurement system must produce repeatable and reproducible values, such measurement systems are called R&R. Statistical test of R&R, despite its importance, is less studied and no formal statistical test is proposed for widely applied attribute measurement systems. We propose a statistical test of R&R for an attribute gauge allowing implementation of a significance test. The efficiency of the methodology is demonstrated on an example.

, 10 pages