Journal of Quality Technology vol. 40 issue 2 - April 2008
Abstract: [This abstract is based on the author’s abstract.] Standard measurement-system analysis criteria assume a measuring device measures a single product characteristic to determine functionality. In automotive-body manufacturing, however, data are taken for many quality characteristics and each response is evaluated independently. This study develops multivariate extensions of gauge-approval criteria precision to tolerance ratio, percent repeatability and reproducibility (R&R), and signal to noise ratio. These criteria use the volume of constant-density contours to characterize variability. A MANOVA method is used to estimate the variance-covariance matrices for one-factor, two-factor, and three-factor gauge studies. It is shown how to fit the MANOVA model and estimate the multivariate criteria using automotive body panel gauge-study data.
Keywords: Gage Repeatability and reproducibility studies (GR&R) - Analysis of variance (ANOVA) - Measurement system - Multivariate capability indices - Signal-to-noise ratio - Automotive
(0) Member Reviews