Statistical process control on autocorrelated process

Abstract

Statistical process control techniques have found widespread application in industry for process improvement and for estimating process parameters or determining capability. Unfortunately, the assumption of uncorrelated or independent observations is not even approximately satisfied in some manufacturing processes. All manufacturing processes are driven by inertial elements, and the frequency of sampling becomes short relative to the process time constant the sequence of process observations will be autocorrelated. There are two major approaches in dealing with autocorrelated process data in process control, that is, residual-based approaches and methods that modify control limits to adjust for autocorrelation. This paper investigates control charts for detecting special causes in an ARIMA(0,1,1) process that is being adjusted automatically after each observation using a minimum mean-squared error adjustment policy. It is assumed that these special causes can change the process mean, process variance, the moving average parameter, or the effect of the adjustment mechanism. The objective is to find control charts or combinations of control charts that will be effective for detecting special causes that results in any of these types of parameter changes in any or all of the parameters.