An Investigation of the Optimal Discount Factors of the Double EWMA and DIA Controllers

Abstract

Motivated by the question of how to reduce the time for a complete search technique to find the optimal discount factors for commonly used run to run (R2R) controllers, we study properties concerning the discount factors of the double exponentially weighted moving average (dEWMA) and its variation, the initial intercept iteratively adjusted (IIIA), controllers for a linear process. We show that the search time can be reduced more than 99.9 percent compared to the benchmark complete search technique.

Key Words:

• Double EWMA controller
• initial intercept iteratively adjusted (IIIA) controller
• local optimal
• optimal discount factors
• run-to-run control

Additional information

Notes on contributors

Yu-Chi Chang

Yu-Chi Chang B.S. degree in mathematics, the M.S. degree in industrial engineering, and the Ph.D. degree in industrial engineering and engineering management from the Tsing-Hua University, Hsin-Chu, Taiwan, R.O.C. From 1989 to 1993, he served as Associate Researcher in the Mechanical Industry Research Laboratories (MIrL), Industrial Technology Research Institute (ITRI), Hsin-Chu, Taiwan. He is an Associate Professor at the Department of Industrial Engineering and Management, Chienkuo Technology University, Changhua, Taiwan, R.O.C. His research interests include run-to-run (R2R) process control, probability and statistics, and stochastic simulation.