藉由最佳化實驗設計提高及改善STRONG之計算效率

Improving the efficiency of STRONG with optimal designs in the presence of model misspecifications

楊世邦(Shih-Pang Yang)

指導教授 ： 張國浩

國立清華大學/工學院/工業工程與工程管理學系/碩士(2013年)

https://doi.org/10.6843/NTHU.2013.00190

摘要

隨機信賴區域反應曲面法 (STRONG)是一具有收斂及自動特性的反應曲面法。STRONG一個很重要的假設是，它的反應曲面與模型是具有同樣型式的；換句話說，是一個一階模型或二階模型。在許多的真實問題中，反應曲面往往是非常不具有線性特性以及難以預測的。在本篇論文中，我們調查最佳化實驗設計運用在一階模型以及二階模型的方法，以用來解釋模型誤解問題。我們也提出了漸進式的實驗設計架構使其與最佳化實驗設計結合，用來改進STRONG在計算上的效率。數值實驗驗證了我們所提出來的新的實驗設計架構確實能有效改善STRONG在計算上的效率。

關鍵字

STRONG ； 模擬最佳化 ； 最佳化實驗設計

並列摘要

Stochastic Trust-Region Response-Surface Method (STRONG) is an improved response surface method that possesses automation property and convergence guarantee. One important assumption underlying STRONG is that the response surface is of the same form as the metamodels, i.e., either a first- or second-order model. In many real problems, however, the response surface is very nonlinear and unpredictable. In this paper, we investigate the optimal first- and second-order designs in the presence of model misspecifications. We propose a sequential design framework that integrates these designs to improve the computational efficiency of STRONG. Numerical experiments verify the effectiveness of the proposed sequential design framework.

並列關鍵字

STRONG ； simulation optimization ； optimal designs

參考文獻

variance. R.G. Ingalls, M.D. Rossetti, J.S. Smith, B.A. Peters, eds., Proceedings

developments, and applications. M.E. Kuhl, N.M. Steiger, F.B. Armstrong, J.A.

F.B. Armstrong, J.A. Joines, eds., Proceeding of the 2005 Winter Simulation

Barton, R.R., J.S. Ivey. 1996. Nelder-Mead simplex modications for simulation

Box, G.E.P., N.R. Draper. 1959. A basis for the selection of a response surface