Monitoring ZA Fertilizer Production using Multivariate Maximum Chart Based on Bootstrap Control Limit

Abstract

Control charts are extensively used to monitor the production process. When there is more than one variable process are considered, the multivariate control charts are typically employed to monitor the mean vector and the variability process separately. In recent years, control charts have been developed for monitoring mean process and variability process simultaneously in a chart. A Maximum multivariate control chart (Max-Mchart) is one of the simultaneous multivariate control charts relying on exact distribution control limit. The objective of this paper is to evolve Max-Mchart based on Bootstrap control limits. This paper also compares Max-Mchart over the Hotelling T2 and Generalized Variance (GV) control chart. The interpretive examples are implemented to demonstrate the applications of the ZA fertilizer production dataset in carbonation step.