Response surface methodology's steepest ascent and step size revisited: Correction


Kleijnen J., den Hertog D., Angun E.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, vol.170, no.2, pp.664-666, 2006 (Journal Indexed in SCI) identifier

  • Publication Type: Article / Article
  • Volume: 170 Issue: 2
  • Publication Date: 2006
  • Doi Number: 10.1016/j.ejor.2004.10.024
  • Title of Journal : EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
  • Page Numbers: pp.664-666

Abstract

This Short Communication corrects Table 1 and Figs. 2 and 3 that were published in a recent article by the same authors, in this journal. The article discussed response surface methodology (RSM), which searches for the input combination maximizing the output of a real or simulated system. RSM uses steepest ascent (SA), which is scale-dependent. The article derived scale-independent `adapted' SA (ASA). The two search directions were explored in Monte Carlo experiments. Unfortunately, the canonical and the non-canonical cases were mixed up. This Communication still shows that-in general-ASA gives a better search direction than SA. (c) 2004 Elsevier B.V. All rights reserved.