Gauge R&R Studies for Destructive Measurements

Abstract

The standard method for assessing a measurement system's precision is a gauge R&R study. Such a study involves an experiment in which each of a number of objects is measured multiple times. From the results, the spread of multiple measurements on a single object (the measurement spread) can be estimated. A serious complication is encountered when objects are affected by the measurement or when the true value of objects is variable. Such measurements are called destructive. Applying the standard gauge R&R set-up to destructive measurements is either impossible or results in an overestimation of measurement spread. This article studies alternative set-ups, which can be applied to destructive measurements if certain conditions hold. Even if required conditions are not completely met, the proposed approaches will at least lead to a smaller overestimation of measurement spread.