Title: A combined array approach to minimise expected prediction errors in experimentation involving mixture and process variables

Authors: Navara Chantarat, Theodore T. Allen, Nilgun Ferhatosmanoglu, Mikhail Bernshteyn

Addresses: Prince of Songkla University, P.O. Box 38, Khohong, Hat Yai, Songkhla 90112, Thailand. ' The Ohio State University, 1971 Neil Avenue, 210 Baker Systems, Columbus, OH 43210-1271, USA. ' The Ohio State University, 1971 Neil Avenue, 210 Baker Systems, Columbus, OH 43210-1271, USA. ' 1580 Cardiff Road, Columbus, OH 43221, USA

Abstract: This paper proposes experimental planning methods for experiments involving mixture variables. These methods generally foster relatively accurate predictions using fewer experimental runs. Methods are based on a criterion that is the expected squared value of these errors, taking into account model bias and the application of alternative candidate sets. We compare the proposed methods with alternatives, using case studies from the literature. Further, we show that computer generated combined arrays based on new, combined models can be expected to yield lower prediction errors compared with relevant alternatives including design methods combining central composites and simplex centroids and D-optimal block designs.

Keywords: mixture experiment; mixture design; process variables; expected integrated mean squared error; EIMSE; bias; combined arrays; experimental planning; prediction errors.

DOI: 10.1504/IJISE.2006.009053

International Journal of Industrial and Systems Engineering, 2006 Vol.1 No.1/2, pp.129 - 147

Published online: 21 Feb 2006 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article