New capability indices for evaluating the performance of multivariate manufacturing processes

Abstract

Generally, an industrial product has more than one quality characteristic. In order to establish performance measures for evaluating the capability of a multivariate manufacturing process, several multivariate process capability indices have been developed in the past few years. Among them, Taam's MC p and MC pm indices have the drawback of overestimation and Hubele's three-component capability vector lacks simplicity in practice. In this article, taking the correlation among multiple quality characteristics into account, we develop two novel indices; NMC p and NMC pm . Using two numerical examples we demonstrate that the true performance of multivariate processes are accurately reflected in our NMC p and NMC pm indices and in their associated interval estimates. Finally, simulation results show that our indices outperform both those of Taam and Hubele.

Determining the optimal allocation of parameters for multivariate measurement system analysis

“…To take the correlation among multiple quality characteristics into account for MGRR, we adopt the idea proposed by Pan and Lee (2010) and revise the denominator of the P/T ratio proposed by Majeske (2008). Accordingly, the revised P/T ratio is defined as…”

“…To take the correlation among multiple quality characteristics into account, Pan and Lee (2010) proposed a novel multiple process capability index NMC p . They revised Taam's et al (1993) modified engineering tolerance region based on the assumption that the correlation of multiple quality characteristics is consistent with the correlation among specifications.…”

“…Traditional MSA only considers a single quality characteristic. With the advent of modern technology, industrial products have become very sophisticated with more than one quality characteristic and there is high correlation among them, which can be found in solder paste stencil printing process (Pan & Lee, 2010). In performing the multivariate MSA for correlated quality characteristics, quality practitioners are also expected to determine the optimal allocation of sample size of parts (p), number of operators (o) and repeated measurements (r) for economic reasons.…”

A review of univariate and multivariate process capability indices

“…Pan and Lee [56] also suggested considering the existing correlation between the measures of the product characteristics in order to obtain a M C pm index. Here, and Spiring [52].…”

“…and in Braun [54] a modied SR, which is also inclined but it is now centered in the target value (which cannot be the center of the SR) is proposed. The approach in Pan and Lee [56] is extended to the non-normal case in Pan, Li and Shih [57] by using a weighted standard deviation method to approximate the original probability density function with segments from 2 v multivariate normal distributions. Jalili, Bashiri and Amiri [58] introduced another multivariate PCI, which is also valid for unilateral specication processes.…”