Abstract
Statistical process control (SPC) techniques have been greatly implemented in industries for improving product quality and saving production costs. As a primary tool among these techniques, control charts are widely used to detect the occurrence of assignable causes. Recently proposed synthetic control charts perform well for monitoring process mean, fraction nonconforming and mean of interarrival times. For monitoring other types of process parameters conveniently, this study proposes a general model of synthetic control chart with regard to non-specified process parameter monitored. The chart is a combination of the Shewhart chart and conforming run length (CRL) chart. The variable sampling intervals (VSI) schemes of the synthetic chart are constructed for improving chart performance further. The Shewhart chart, VSI Shewhart chart, and synthetic chart can be regarded as special cases of the VSI synthetic chart. The general model of synthetic control chart and its VSI schemes are then, respectively, applied to three kinds of parameters of a normally distributed process: process variability, process mean and variability, and process mean vector. These synthetic charts are composed of Shewhart sample range (R) chart or sample standard deviation (S) chart, Shewhart Max chart, and Shewhart chi-squared ( ) chart or T-squared ( ) chart, and CRL chart, respectively. The operation, design, performance comparisons with other types of control charts such as Shewhart chart with runs rules, Shewhart chart with warning line, cumulative sum (CUSUM) chart and exponentially weighted moving average (EWMA) chart, and numerical example are described. Tables of optimally designed synthetic charts can assist the users to design the chart, and the presented VSI schemes can be implemented to further improve the chart performance if the current one is unsatisfactory.