A Statistical Comparison of Multiobjective Evolutionary Algorithms Including the MOMGA-II
- 3.8k Downloads
Many real-world scientific and engineering applications involve finding innovative solutions to “hard” Multiobjective Optimization Problems (MOP). Various Multiobjective Evolutionary Algorithms (MOEA) have been developed to obtain MOP Pareto solutions. A particular exciting MOEA is the MOMGA which is an extension of the single-objective building block (BB) based messy Genetic Algorithm. The intent of this discussion is to illustrate that modifications made to the Multi-Objective messy GA (MOMGA) have further improved its efficiency resulting in the MOMGA-II. The MOMGA-II uses a probabilistic BB approach to initializing the population referred to as Probabilistically Complete Initialization. This has the effect of improving the efficiency of the MOMGA through the reduction of computational bottle-necks. Similar statistical results have been obtained using the MOMGA-II as compared to the results of the original MOMGA as well as those obtained by other MOEAs as tested with standard generic MOP test suites.
KeywordsPareto Front Test Suite Generational Distance Pareto Optimal Solution Multiobjective Problem
Unable to display preview. Download preview PDF.
- 1.David A. Van Veldhuizen. Multiobjective Evolutionary Algorithms: Classifications, Analyses, and New Innovations. PhD thesis, Air Force Institute of Technology, Wright-Patterson AFB, 1999.Google Scholar
- 2.David A. Van Veldhuizen and Gary B. Lamont. On measuring multiobjective evolutionary algorithm performance. In 2000 Congress on Evolutionary Computation, pages 204–211, 2000.Google Scholar
- 3.David A. Van Veldhuizen and Gary B. Lamont. Multiobjective optimization with messy genetic algorithms. In 2000 Symposium on Applied Computing, pages 470–476, 2000.Google Scholar
- 5.David E. Goldberg, Kalyanmoy Deb, H. Kargupta, and G. Harik. Rapid, accurate optimization of difficult problems using fast messy genetic algorithms. Technical Report 93004, University of Illinois at Urbana-Champaign, 1993.Google Scholar
- 7.Eckart Zitzler and Lothar Thiele. An evolutionary algorithm for multiobjective optimization: The strength pareto approach. Technical report, Computer Eng. and Comm. Networks Lab, Swiss Federal Institute of Technology (ETH), 1998.Google Scholar
- 8.Kalyanmoy Deb, Samir Agrawal, Amrit Pratap, and T Meyarivan. Fast elitist non-dominated sorting genetic algorithm for multi-objective optimization: Nsga-ii. In Parallel Problem Solving from Nature-PPSN VI, 2000.Google Scholar
- 9.David A. Van Veldhuizen and Gary B. Lamont. Multiobjective evolutionary algorithm research: A history and analysis. Technical Report TR-98-03, Air Force Institute of Technology, Wright-Patterson AFB, OH, 1998.Google Scholar
- 10.David E. Goldberg, Bradley Korb, and Kalyanmoy Deb. Messy genetic algorithms: Motivation, analysis, and first results. In Complex Systems, pages 493–530, 1989.Google Scholar
- 11.Hillol Kargupta. SEARCH: Polynomial Complexity, And The Fast Messy Genetic Algorithm. PhD thesis, University of Illinois at Urbana-Champaign, 1995.Google Scholar