Abstract
Process monitoring and process adjustment strategies are two important parts of the process improvement methods, and they are often considered separately but should be integrated together. Integrated moving average (IMA) model is the most common disturbance model, and step shift model is one type of more complicated disturbance model that often exists in many real applications. In this paper we investigate the case when IMA background disturbance is subject to random step shifts with a certain probability. We propose a process adjustment procedure with feedback control, together with a so-called quasi-feedforward control based on process monitoring of the output errors. This control strategy was proved to be very robust against parameter misspecifications in the disturbance model. We also further investigate the effects of type I and type II errors in process adjustment on this IMA plus random step shift disturbance model.
Additional information
Notes on contributors
Lihui Shi
Lihui Shi is currently a data scientist working on big data projects at eBay Inc at Bellevue, WA. He received his Ph.D. in Industrial and Systems Engineering from the University of Washington at Seattle in 2012. He also received a MS degree in Statistics from University of Washington at Seattle. His research interests include statistical process control, process adjustment, reliability theory, design of experiments, etc.
Kailash C. Kapur
Kailash C. Kapur is a Professor of Industrial and Systems Engineering, College of Engineering, University of Washington. He has served as the Director of Industrial Engineering from 1993–1999. He was a Professor and the Director of the School of Industrial Engineering, University of Oklahoma from 1989–1992. Dr. Kapur received the Ph.D. degree (1969) in Industrial Engineering from the University of California, Berkeley. He has co-authored the book Reliability in Engineering Design, John Wiley & Sons, 1977. He is a Fellow of ASQ and IIE.