Abstract
Abstract
Process control techniques from engineering and statistical process control overlap at the interface of the two process control methodologies. In recent years, process control practitioners have focused their attention on statistical process monitoring and feedback control adjustment. It is quite common to encounter problems that are related to feedback (closed-loop) stability, controller limitations and dead-time compensation in order to obtain minimum variance (mean square error) at the output. A stochastic feedback control algorithm for input adjustment can be derived by use of techniques from the two process control methodologies. This paper briefly reviews statistical feedback control adjustment and describes a method to model stochastic feedback control adjustment in order to control the quality of products at the output of a continuous process. The method takes into consideration situations under which the product quality mean is on target/set point and situations where it is not on target by assigning suitable probabilities for the respective situations. A test of significance is conducted for such situations when a feedback control adjustment is required to bring the product quality mean on target and a suitable probability model is developed for such a situation.