런규칙을 사용한 개량된 경계선 수정계획의 설계와 Markov 연쇄의 적용

The bounded adjustment is known to be more efficient than repeated adjustment when the cost is incurred for engineering process control. The procedure of the bounded adjustment is to adjust the process when the one-step predicted deviation exceeds the adjustment limit by the amount of the prediction. In this paper, two run rules are proposed and studied in order to improve the efficiency of the traditional bounded adjustment procedure. The efficiency is studied in terms of the standardized cost through Monte Carlo simulation when the procedure is operated with and without the run rules. The adjustment procedure operated with run rules turns out to be more robust for changes in the process and cost parameters. The Markov chain approach for calculating the properties of the run rules is also studied.