NUMBER OF CYCLES IN EVOLUTIONARY OPERATION

  • Lim, Yong-B. (Department of Statistics, Ewha Womans University) ;
  • Park, Sung-H. (Department of Statistics, Seoul National University)
  • Published : 2007.06.30

Abstract

Evolutionary operation (EVOP) proposed by Box (1957) is a method for continuous monitoring and improvement of a full-scale manufacturing process with the objective of moving the operating conditions toward the better ones. EVOP consists of systematically making small changes in the levels of the two or three process variables under consideration. Data are collected on the response variable at each point of two level factorial design with the center point and a cycle is said to have been completed. The cycles are replicated sequentially until the decision is made on whether further cycle of experiments is needed to conclude the significance of any of main effects or interaction effects or the curvature. In this paper, an improved flow chart of EVOP is proposed and how to determine the number of cycles is studied based on the size of type II error. In order to reject the alternative hypothesis of interests with more confidence and conclude that we believe in the null hypothesis of no effects, we propose a counter measure $p^*-value$ corresponding to the p-value. The relationship of $p^*-value$ to the probability of type II error ${\beta}$ under the alternative hypothesis of interests is analogous to that of p-value to the probability of type I error ${\alpha}$. Also the implementation of EVOP with a mixture experiment is discussed.

Keywords

References

  1. Box, G. E. P. (1957). 'Evolutionary operation: a method for increasing industrial productivity', Applied Statistics, 6, 81-101 https://doi.org/10.2307/2985505
  2. Box, G. E. P., HUNTER, W. G. AND HUNTER, J. S. (1978). Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building, John Wiley & Sons, New Jersey
  3. CORNELL, J. A. (1990). Experiments with Mixtures: Designs, Models, and the Analysis of Mixture Data, 2nd ed., John Wiley & Sons, New York
  4. LORENZEN, T. J. AND ANDERSON, V. L. (1993). Design of Experiments: A No-name Approach, Marcel Dekker, New York
  5. MYERS, R. H. AND MONTGOMERY, D. C. (1995). Response Surface Methodology: Process and Product Optimization Using Designed Experiments, John Wiley & Sons, New York
  6. NORTON, V. (1983). 'A simple algorithm for computing the non-central F distribution', Applied Statistics, 32, 84-85 https://doi.org/10.2307/2348047