An assessment of the robustness of gauge repeatability and reproducibility analysis in automotive components

Abstract

Measurement system analysis is a vital component of many continuous improvement initiatives. In the manufacturing industry, variation in a measurement system is a key characteristic to be assessed and improved upon. Measurement systems are commonly evaluated by performing a gauge repeatability and reproducibility (gauge R&R) study, which analyses any variations between measurements by a gauge (repeatability) and between results obtained by different appraisers (reproducibility). To determine the preferred method for gauge R&R analysis; average and range, and two-way analysis of variance were applied to data taken from four real case studies from the automotive industry (a stabilizer clamp bracket, a coated steel sheet, a steel sheet body panel, and a brake disc). Possible differences in gauge R&R characteristics were compared, and residual diagnosis was used to discern violations in two-way ANOVA. To develop a robust road map for gauge R&R analysis, interactions between gauge R&R parameters, residual diagnosis, and number of distinct categories were analysed.