SMS-EMOA – Effektive evolutionäre Mehrzieloptimierung (SMS-EMOA – Effective Evolutionary Multiobjective Optimization)

Bei der mehrkriteriellen Pareto-Optimierung wird zu konfliktären Anforderungen eine Menge von Kompromisslösungen gesucht, die die bestmöglichen Lösungen approximieren. Evolutionäre Algorithmen haben sich hierbei als effektive und robuste Verfahren bewährt. Die Güte einer Approximation lässt sich durch das von ihr dominierte Hypervolumen im Zielraum, der sogenannten S-Metrik, quantifizieren. Die Maximierung der S-Metrik ist also ein erstrebenswertes Ziel und gleichzeitig eine adäquate einkriterielle Ersatzzielfunktion. Ein evolutionärer Algorithmus setzt diese innerhalb der Selektion ein und erreicht dadurch hervorragende Ergebnismengen. Wir zeigen anhand von Benchmarkproblemen und realen Anwendungen aus der Flugzeugtechnik und anderen Industriezweigen, dass dieser Algorithmus außerordentlich effektiv ist. Dies zeigt sich insbesondere für den Fall, dass mehr als drei Ziele zu optimieren sind, weil dann andere populäre mehrkriterielle evolutionäre Algorithmen versagen. In multiobjective optimization there are conflicting demands for which a set of compromise solution is searched which approximates best possible solutions. Evolutionary algorithms established as effective and robust methodologies for this task. The quality of an approximation may be evaluated by its hypervolume dominated in the objective space, called S-metric. Thus, the maximization of the S-metric is a desirable aim and an adequate single-objective substitute function. An evolutionary algorithm applies this measure within its selection operator and reaches excellent results. We demonstrate the algorithm´s outstanding effectiveness by academic benchmark problems and real-world problems from aeronautics and other industries. The performance is especially striking in the case of more than three objectives, when other popular methods fail completely.

[1]  Susana Cecilia Esquivel Evolutionary algorithms for solving multi-objetive problems . Carlos A. Coello Coello, David A. van Veldhuizen and Gary R., Lamont , 2002 .

[2]  Evan J. Hughes,et al.  Evolutionary many-objective optimisation: many once or one many? , 2005, 2005 IEEE Congress on Evolutionary Computation.

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

[4]  Lothar Thiele,et al.  Comparison of Multiobjective Evolutionary Algorithms: Empirical Results , 2000, Evolutionary Computation.

[5]  Nicola Beume,et al.  S-Metric Calculation by Considering Dominated Hypervolume as Klee's Measure Problem , 2009, Evolutionary Computation.

[6]  Nicola Beume,et al.  An EMO Algorithm Using the Hypervolume Measure as Selection Criterion , 2005, EMO.

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

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

[9]  Günter Rudolph,et al.  Solving multimodal problems via multiobjective techniques with Application to phase equilibrium detection , 2007, 2007 IEEE Congress on Evolutionary Computation.

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

[11]  Kalyanmoy Deb,et al.  Towards a Quick Computation of Well-Spread Pareto-Optimal Solutions , 2003, EMO.

[12]  Nicola Beume,et al.  Pareto-, Aggregation-, and Indicator-Based Methods in Many-Objective Optimization , 2007, EMO.

[13]  Nicola Beume,et al.  Design and Management of Complex Technical Processes and Systems by means of Computational Intelligence Methods Gradient-based / Evolutionary Relay Hybrid for Computing Pareto Front Approximations Maximizing the S-Metric , 2007 .

[14]  Zbigniew Michalewicz,et al.  Handbook of Evolutionary Computation , 1997 .

[15]  Nicola Beume,et al.  Multi-objective optimisation using S-metric selection: application to three-dimensional solution spaces , 2005, 2005 IEEE Congress on Evolutionary Computation.

[16]  Nicola Beume,et al.  SMS-EMOA: Multiobjective selection based on dominated hypervolume , 2007, Eur. J. Oper. Res..

[17]  Eckart Zitzler,et al.  Indicator-Based Selection in Multiobjective Search , 2004, PPSN.

[18]  Marco Laumanns,et al.  SPEA2: Improving the Strength Pareto Evolutionary Algorithm For Multiobjective Optimization , 2002 .

[19]  Marco Laumanns,et al.  Scalable multi-objective optimization test problems , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[20]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems , 2002, Genetic Algorithms and Evolutionary Computation.

[21]  Marco Laumanns,et al.  Performance assessment of multiobjective optimizers: an analysis and review , 2003, IEEE Trans. Evol. Comput..