Abstract
Combining an EWMA chart with a Shewhart chart is traditionally recommended as a means of providing good protection against both small and large shifts in the process mean. Capizzi and Masarotto have proposed an EWMA chart with a variable smoothing constant (AEWMA) for the same purpose. In this paper, we optimize the designs of the AEWMA and of the combined EWMA-Shewhart schemes with regard to pairs of shifts in the process mean and compare their performances. When the schemes are optimized for the same pair of shifts, their ARL profiles practically coincide. The choice between the AEWMA and the EWMA-Shewhart scheme then becomes a matter of personal preference. We also explore using an AEWMA together with a Shewhart chart, but we find no performance improvement. An additional contribution of this paper is the tables of optimal designs of each scheme for several pairs of shifts.
Additional information
Notes on contributors
Bruno F. T. Simoes
Bruno Francisco Teixeira Simoes received his B.S. degree in Statistics from the Rio de Janeiro State University and his M.S. degree in Production Engineering from the Pontifical Catholic University of Rio de Janeiro (PUC-Rio), Brazil. Nowadays, he is a Ph.D. student in the Department of Industrial Engineering at PUC-Rio. His research interests include statistical quality control and multivariate statistical analysis.
Eugenio K. Epprecht
Eugenio Kahn Epprecht is an Associate Professor in the Department of Industrial Engineering at the Pontifical Catholic University of Rio de Janeiro (PUC-Rio), Brazil. He received his B.S. and M.S. degrees in Electrical Engineering from PUC-Rio and his Ph.D. in Computer Science from Facultes Universitaires Notre-Dame de la Paix, Belgium. His current research interest is in statistical quality control. He has published papers in the Brazilian Journal of Operations & Production Management, IIE Transactions, International Journal of Production Economics, International Journal of Production Research and Quality Engineering. He is an active reviewer for several journals, and a member of ISBIS and ASQ. At the time this paper has been finished he was spending a sabbatical year in the Department of Applied Statistics and Operations Research and Quality of the Universidad Politecnica de Valencia, Spain.
Antonio F. B. Costa
Antonio Fernando Branco Costa is a Professor in the Department of Production at UNESP — Sao Paulo State University, Brazil. He was a postdoctoral fellow in the Center for Quality and Productivity Improvement at University of Wisconsin, Madison. His current interest is in statistical quality control. He has published papers in the Brazilian Journal of Operations & Production Management, Journal of Quality Technology, European Journal of Operational Research, IIE Transactions, Journal of Applied Statistics, International Journal of Production Economics, International Journal of Production Research, International Journal of Advanced Manufacturing Technology, Quality Technology and Quantitative Management, Quality and Reliability Engineering International and Communications in Statistics. He is an active reviewer for several journals and an ASQ Certified Quality Engineer. He was the recipient of an IIE Transactions best paper award.