Economic design of an integrated process control procedure with repeated adjustments and EWMA monitoring

Abstract

Statistical process control (SPC) reduces process variability by detecting and eliminating special causes of process variation, while engineering process control (EPC) reduces variability by adjusting the process to keep the quality variable close to target. This article considers an integrated process control (IPC) procedure that simultaneously applies SPC and EPC techniques to reduce the variation of a process. The process model considered is ARIMA(0, 1, 1) when the process is in control. A repeated adjustment EPC scheme is applied by compensating for the predicted deviation from target. It is assumed that a special cause can change the process level, the process variance, the moving average parameter, or the system gain. Two exponentially weighted moving average (EWMA) control charts are applied to the observed deviations to detect special causes. One EWMA chart is for detecting mean shifts, and the other is for detecting variance increases. The effectiveness of the IPC scheme is evaluated using the expected cost per unit time. One major objective of this article is to investigate the effects of the process and cost parameters on the expected cost per unit time. Another major objective is to give practical guidelines for the efficient selection of the chart parameters of the two EWMA control charts.