축차적 반응표면 분석을 통한 M&S 메타모형 구축에 관한 사례 연구 = A Case Study for Finding an Efficient M&S Meta Model through Sequential Response Surface Methodology

다국어 초록 (Multilingual Abstract)

In computer simulation models the output from the computer code is often deterministic, i.e., running the code twice with the same values for the input variables would give the same output. It is discussed why the response surface method with polynomial approximation for the true response function is a good approximation to the computer experiments model. A sequential strategy to find the proper reduced quadratic polynomial model is illustrated with a case study in the military war game computer simulation model.

참고문헌 (Reference)

1 임용빈, "현재의 공정조건을 향상시키기 위한 혼합물 반응표면 방법론" 한국품질경영학회 38 (38): 413-424, 2010

2 Batmaz, I., "Second-Order Experimental Designs for Simulation Metamodeling" 78 : 699-715, 2002

3 Myers, R. H., "Response surface methodology, 3rd ed" Wiley 2008

4 Welch, W. J., "Reply to"Comments on ‘Screening, predicting, and computer experiments’"[by J. M. Lucas(1996)]" 38 : 199-203, 1996

5 Kleijnen, J. P. C., "Kriging metamodeling in simulation : A review" 192 : 707-716, 2009

6 Sacks, J., "Designs for computer experiments" 34 : 15-25, 1989

7 Lim, Y. B., "Design and analysis of computer experiments when the output is highly correlated over the input space" 30 : 109-126, 2002

8 Stat-Ease, Inc., "Design Expert Version 8. 06. Minneapolis" MN 2011

9 Van Beers, W. C. M., "Customized Sequential Designs for Random Simulation Experiments : Kriging Metamodeling and Bootstraping" 186 : 11099-1113, 2008

10 Kleijnen, J. P. C., "A Methodology for Fitting and Validating Metamodels in Simulation" 120 : 14-29, 2000