Welcome to Journal of University of Chinese Academy of Sciences,Today is

›› 2009, Vol. 26 ›› Issue (1): 9-17.DOI: 10.7523/j.issn.2095-6134.2009.1.002

• 论文 • Previous Articles     Next Articles

The optimization of varying orders in one-factor-at-a-time experimentation basing on prior knowledge

Guo Rui, Zhang Jian-Fang   

  1. 1School of Mathematical Sciences, Graduate University of Chinese Academy of Sciences, Beijing, 100049;
    2School of Management, Graduate University of Chinese Academy of Sciences, Beijing, 100080
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-01-15

Abstract:

The One-factor-at-a-time (OFAT)plan was not recommended in practice in design of experiment, but following the recent study of OFAT, the effect and role of OFAT are better understood and realized, which has some advantages in optimization . Basing on our latest two papers, we first analyzed the prior knowledge, then studied optimization strategies of varying orders choices in OFAT experimentation based on prior knowledge. Some benefit and useful results for OFAT were gotten by simulation on many examples. Meanwhile, we did the comparison of OFAT based on prior knowledge with the orthogonal array under the equal number of experimental runs; the result is OFAT wins in average. Finally, we gave the conclusions and described the future work.

Key words: one-factor-at-a-time experimentation, design of experiment, interaction, prior knowledge, orthogonal array