한국시뮬레이션학회논문지 (Journal of the Korea Society for Simulation)

한국시뮬레이션학회 (The Korea Society for Simulation)
권호 표지
DOI QR코드

DOI QR Code

공간채움 조건을 만족하는 컴퓨터 실험 시나리오의 효율적 생성

Efficient Generation of Space Filling Scenarios for Computer Experiments

  • 임동순 (한남대학교 산업경영공학과) ;
  • 김정훈 (국방과학연구소 제6기술연구본부) ;
  • 최봉완 (한남대학교 국방M&S연구센터)
  • 투고 : 2012.10.23
  • 심사 : 2013.05.06
  • 발행 : 2013.09.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

공학설계분야에서 시뮬레이션 모델을 이용한 실험은 매우 중요한 역할을 한다. 그러나 최적의 설계 파라미터를 구하기 위한 시뮬레이션 실험은 많은 실행시간과 자원의 소요를 필요로 한다. 이를 극복하는 방안으로 입력 변수와 성능척도 간의 관계를 표현한 메타모형이 효과적으로 이용된다. 메타모형을 수립하기 위해서는 샘플 시나리오들을 입력으로 하는 시뮬레이션 실행이 요구된다. 이때 샘플 시나리의 수와 질이 메타 모형을 수립하는데 걸리는 시간과 메타모형의 정확성을 결정한다. 공간채움 특성은 샘플 시나리오들의 질을 결정하는 중요한 조건이 된다. 이 논문은 maxmin, Audze-Eglais, centered L2-discrepancy의 3가지 공간채움 척도에 기초한 샘플 시나리오 생성 방법을 제안하고, 실험을 통해 이들 생성방법에 대한 성능을 분석한 결과를 논의한다.

In general, simulation models are effectively used in the field of engineering design. The experiment with simulation models to obtain optimal design parameters, however, is a time-consuming task and requires a lot of resources. Hence, meta-models representing the relationships between input variables and performance measures are exploited to efficiently determine the value of design parameters. To construct a meta-model, a number of simulation executions with sample scenarios are required. The number and quality of sample scenarios determine not only the level of efficiency in constructing the meta-model but also accuracy of the model. Space-filling condition is regarded to be an important condition for the quality of scenarios. This paper proposes sample scenario generation methods based on space-filling measures such as maxmin, Audze-Eglais, and centered L2-discrepancy. The performance of these scenario generation methods are evaluated through experiments.

키워드

참고문헌

  1. Audze, P. and Eglais, V. (1977), "New approach to planning out of experiments", Problems of Dynamics and Strength, Vol. 35, pp. 104-107.
  2. Bates, S. J., Sienz, J, Langley, D. S. (2003), "Formulation of the Audze-Eglais uniform Latin hypercube design of experiments", Advances in Engineering Software, Vol. 34, pp. 493-506. https://doi.org/10.1016/S0965-9978(03)00042-5
  3. Fang, K., Lin, D., Winker P., Zhang, Y. (2000), "Uniform Design: Theory and Application", Technometrics, Vol. 42, No. 3, pp. 237-248. https://doi.org/10.1080/00401706.2000.10486045
  4. Fang, K and Lin D. K. (2003), "Uniform experimental design and their applications in industry", Handbook of Statistics, Vol. 22.
  5. Hickernell, F. J. (1998), "A generalized discrepancy and quadrature error bound", Mathematics of Computation, Vol. 67, pp. 299-322. https://doi.org/10.1090/S0025-5718-98-00894-1
  6. Johnson, M. E., Moore, L. M., and Ylvisaker, D. (1990), "Minimax and maxmin distance design", Journal of Statistical Planning and Inference, Vol. 26, pp. 131-148. https://doi.org/10.1016/0378-3758(90)90122-B
  7. Johnson, R. T., Jones, B., and Fowler, J. W. (2008), "Comparing Designs for Computer Simulation Experiments", Proceedings of the 2008 Winter Simulation Conference, pp. 463-470.
  8. Kleijen, J. P. (2005), "An Overview of the design and analysis of simulation experiments for sensitivity analysis", European Journal of Operational Research, Vol. 164, No. 2, pp. 287-300. https://doi.org/10.1016/j.ejor.2004.02.005
  9. Liefvendahl, M. and Stocki, R. (2006), "A study on algorithms for optimization of Latin hypercubes", Journal of Statistical and Inference, Vol. 136, pp. 3231-3247. https://doi.org/10.1016/j.jspi.2005.01.007
  10. Morris, M. D. and Mitchell, T. J. (1995), "Exploratory designs for computer experiments", Journal of Statistical Planning and Inference, Vol. 43, pp. 381-402. https://doi.org/10.1016/0378-3758(94)00035-T
  11. Park, J. S. (1994), "Optimal Latin-hypercube designs for computer experiments", Journal of Statistical Planning and Inference, Vol. 39, pp. 95-111. https://doi.org/10.1016/0378-3758(94)90115-5
  12. Pronzato, L. and Muller, W. G. (2012), "Design of Computer Experiments: Space Filling and Beyond", Statistics and Computing, Vol 22, pp. 681-701. https://doi.org/10.1007/s11222-011-9242-3
  13. Wim, C. M. (2005), "Kriging Metamodeling in Discrete-Event Simulation: An Overview", Proceedings of the 2005 Winter Simulation Conference, pp. 202-208.
  14. Ye, K. Q., Li, W., Sudjianto, A. (2000), "Algorithmic construction of optimal symmetric Latin hypercubes", Journal of Statistical Planning and Inference, Vol. 90, pp. 145-159. https://doi.org/10.1016/S0378-3758(00)00105-1