Integration of statistical selection with search mechanism for solving multi-objective simulation-optimization problems

• Published in: Proceedings of the 38th conference on Winter simulation pp. 294 - 303
• Main Authors: Lee, Loo Hay, Chew, Ek Peng, Teng, Suyan
• Format: Conference Proceeding
• Language: English
• Published: Winter Simulation Conference 03-12-2006
• Series: ACM Conferences
• Subjects: Computing methodologies; Computing methodologies > Artificial intelligence; Computing methodologies > Artificial intelligence > Search methodologies; Computing methodologies > Artificial intelligence > Search methodologies > Heuristic function construction; Computing methodologies > Modeling and simulation; Computing methodologies > Modeling and simulation > Model development and analysis; Computing methodologies > Modeling and simulation > Model development and analysis > Modeling methodologies; Information systems; Information systems > Data management systems; Information systems > Data management systems > Database administration; Information systems > Data management systems > Database administration > Data dictionaries; Theory of computation; Theory of computation > Design and analysis of algorithms; Theory of computation > Design and analysis of algorithms > Mathematical optimization
• ISBN: 9781424405015; 1424405017
• DOI: 10.5555/1218112.1218170

In this paper, we consider a multi-objective simulation optimization problem with three features: huge solution space, high uncertainty in performance measures, and multi-objective problem which requires a set of nondominated solutions. Our main purpose is to study how to integrate statistical selection with search mechanism to address the above difficulties, and to present a general solution framework for solving such problems. Here due to the multi-objective nature, statistical selection is done by the multi-objective computing budget allocation (MOCBA) procedure. For illustration, MOCBA is integrated with two meta-heuristics: multi-objective evolutionary algorithm (MOEA) and nested partitions (NP) to identify the nondominated solutions for two inventory management case study problems. Results show that, the integrated solution framework has improved both search efficiency and simulation efficiency. Moreover, it is capable of identifying a set of non-dominated solutions with high confidence.