Multiple Criteria Performance Analysis of Non-dominated Sets Obtained by Multi-objective Evolutionary Algorithms for Optimisation

摘要: The paper shows the importance of a multi-criteria performance analysis in evaluating quality non-dominated sets. sets are generated by use evolutionary algorithms, more specifically through SPEA2 or NSGA-II. Problem examples from different problem domains analyzed on four criteria quality. These namely cardinality set, spread solutions, hyper-volume, and set coverage do not favour any algorithm along examples. In Multiple Shortest Path (MSPP) examples, solutions is decisive factor for 2S|1M configuration, 3S configuration. differences values between NSGA-II MSPP small since both algorithms have almost identical solutions. Decision Tree factors hyper-volume. computations show that criterion vary all except criterion. This binary measure sets, as itself tests dominance. various confronted means decision tool.