A new multivariate extended homogeneously weighted moving average monitoring scheme incorporated with a support vector machine

Abstract

Statistical process monitoring (SPM) provides a way, by means of monitoring schemes, to be alerted when a process significantly changes, and to initiate investigations for potential causes of variation. However, these schemes can only determine if the process is stable or unstable and they cannot provide any additional information. Several researchers have recommended the integration of machine learning (ML) approaches into SPM as a solution for the shortfalls of the techniques used to construct traditional monitoring schemes. This paper introduces a new multivariate extended homogeneously weighted moving average (MEHWMA) monitoring scheme and investigates its performance in terms of the run‐length distribution using simulation. In addition, the proposed MEHWMA scheme is integrated with a support vector machine to allow for the classification of the out‐of‐control events which facilitates the identification of the root causes of variation in the process. A numerical illustrative example is provided using real‐life data.

Homogeneously Weighted Moving Average Control Charts: Overview, Controversies, and New Directions

“…Recently, Pieters et al [80] proposed the multivariate extended HWMA (MEHWMA) chart by imitating the multivariate extended EWMA (MEEWMA) chart. This chart is designed to monitor small shifts in a multivariate process mean vector.…”