Title: Investigating estimation error reduction strategies in complex engineering systems

Authors: Paul Goethals; Byung Rae Cho

Addresses: Department of Mathematical Sciences, United States Military Academy, West Point, NY 10996, USA ' Department of Industrial Engineering, Clemson University, Clemson, SC 29634, USA

Abstract: When the manufacturing objective is process or product improvement, quality practitioners will frequently resort to one or more approaches within the broader class of response surface methodology. Several techniques, such as the dual response, robust parameter design, and desirability function approach, may be effective tools to solve the multi-response optimisation problem. All of these techniques are designed to identify the factor settings that lead to an optimal solution in terms of the mean or variance among characteristics. The skewness in the distribution of observations for one or more characteristics, however, is not considered. The techniques also traditionally rely on the fit of second-order response surface designs in estimating each response, which may be unreliable in some cases. In contrast, this paper offers an approach to solving complex multi-response optimisation problems that considers both the error associated with process skewness and the accuracy of a response surface.

Keywords: response surface methodology; RSM; robust design; estimators; variability; coefficient of variation; multi-response optimisation; process skewness; estimation error reduction; complex engineering systems.

DOI: 10.1504/IJDATS.2014.059014

International Journal of Data Analysis Techniques and Strategies, 2014 Vol.6 No.1, pp.43 - 72

Published online: 05 Jul 2014 *

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