A Critique on the Foundational Response Surface Methodology for Exploring Optimal Regions
John E. Usen *
Department of Statistics, Cross River University of Technology, P.O.Box 1123, Calabar, Nigeria.
Essien J. Okoi
Department of Statistics, Cross River University of Technology, P.O.Box 1123, Calabar, Nigeria.
Eric M. Egomo
Department of Statistics, Cross River University of Technology, P.O.Box 1123, Calabar, Nigeria.
Ekeng N. Henshaw
Department of Statistics, Cross River University of Technology, P.O.Box 1123, Calabar, Nigeria.
Edet B. Hogan
Department of Statistics, Cross River University of Technology, P.O.Box 1123, Calabar, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
The interest of most process engineers in industries is usually to optimize the yield of their processes. Not until 1951, imprecise methodologies were used in industries for this purpose. However, in 1951, G. E. P. Box and K. B. Wilson invented the technique of Response Surface Methodology (RSM) as one used for the optimization of the yield of processes. Being an initial idea, this paper has considered RSM as a foundational idea. In particular, it criticizes this foundational idea from the angle of its intuitive approach to searching for near-optimal settings of industrial processes, should such processes fail to run at optimal settings. RSM uses the tools of canonical transformation and analysis (a trial-and-error routine) for this search. Regardless, the foundational response surface methodology is acknowledged to be primarily efficient for determining the optimum response.
Keywords: Foundational response surface methodology, near-optimal settings, canonical transformation, canonical analysis.