Computing Pareto fronts using distributed agents

Abstract Many problems that face a business decision maker are most accurately formulated as multi-objective optimization problems. However, actually solving these problems is a difficult and computationally expensive process. In this paper, we develop and use an agent-based optimization system for efficiently generating the non-dominated solution set to a non-convex multi-objective optimization problem. Through simulation, we highlight the benefits that arise from allowing collaboration among agents embodying a diverse set of algorithms. Further, we demonstrate the ability of the agent system to use large-scale parallel resources efficiently. We also propose two new metrics for quantifying the performance of multi-objective optimization systems.

[1]  John L. Gustafson,et al.  Reevaluating Amdahl's law , 1988, CACM.

[2]  Ignacio E. Grossmann,et al.  Systematic Methods of Chemical Process Design , 1997 .

[3]  Arthur C. Graesser,et al.  Is it an Agent, or Just a Program?: A Taxonomy for Autonomous Agents , 1996, ATAL.

[4]  Kalyanmoy Deb,et al.  Multi-objective optimization using evolutionary algorithms , 2001, Wiley-Interscience series in systems and optimization.

[5]  Jason R. Schott.  Fault Tolerant Design Using Single and Multicriteria Genetic Algorithm Optimization. , 1995 .

[6]  Craig W. Reynolds.  Flocks, herds, and schools: a distributed behavioral model , 1987, SIGGRAPH.

[7]  Gary B. Lamont,et al.  Multiobjective evolutionary algorithms: classifications, analyses, and new innovations , 1999 .

[8]  Sarosh N. Talukdar,et al.  Cooperation schemes for autonomous agents , 1996 .

[9]  Lothar Thiele,et al.  Multiobjective Optimization Using Evolutionary Algorithms - A Comparative Case Study , 1998, PPSN.

[10]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[11]  Melanie Mitchell,et al.  An introduction to genetic algorithms , 1996 .

[12]  S.N. Talukdar,et al.  Scale efficient organizations , 1992, [Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics.

[13]  G. Amdhal,et al.  Validity of the single processor approach to achieving large scale computing capabilities , 1967, AFIPS '67 (Spring).

[14]  Arthur W. Westerberg,et al.  Optimization for design problems having more than one objective , 1983 .

[15]  Allen Newell,et al.  SOAR: An Architecture for General Intelligence , 1987, Artif. Intell..

[16]  Marco Dorigo,et al.  Ant system: optimization by a colony of cooperating agents , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[17]  David E. Goldberg,et al.  Genetic Algorithms with Sharing for Multimodalfunction Optimization , 1987, ICGA.

[18]  Laurent Chaudron,et al.  Conflicting agents: conflict management in multi-agent systems , 2001 .

[19]  Eckart Zitzler,et al.  Evolutionary algorithms for multiobjective optimization: methods and applications , 1999 .

[20]  Kaisa Miettinen,et al.  Nonlinear multiobjective optimization , 1998, International series in operations research and management science.

[21]  Les Gasser,et al.  Perspectives on Organizations in Multi-agent Systems , 2001, EASSS.

[22]  D. Himmelblau,et al.  Optimization of Chemical Processes , 1987 .

[23]  Michael Allen,et al.  Parallel programming: techniques and applications using networked workstations and parallel computers , 1998 .

[24]  Steinar Hauan,et al.  Toward agent-based process systems engineering: proposed framework and application to non-convex optimization , 2003, Comput. Chem. Eng..

[25]  Panos Y. Papalambros,et al.  Principles of Optimal Design: Modeling and Computation , 1988 .

[26]  Peter J. Fleming,et al.  On the Performance Assessment and Comparison of Stochastic Multiobjective Optimizers , 1996, PPSN.