Guidance in evolutionary multi-objective optimization

Many real world design problems involve multiple, usually conflicting optim ization criteria. Often, it is very difficult to weight the criteria exactl y before alternatives are known. Multi-Objective Evolutionary Algorithms ba sed on the principle of Pareto optimality are designed to explore the compl ete set of non-dominated solutions, which then allows the user to choose am ong many alternatives. However, although it is very difficult to exactly de fine the weighting of different optimization criteria, usually the user has some notion as to what range of weightings might be reasonable. In this pa per, we present a novel, simple, and intuitive way to integrate the user's preference into the evolutionary algorithm by allowing to define linear max imum and minimum trade-off functions. On a number of test problems we show that the proposed algorithm efficiently guides the population towards the i nteresting region, allowing a faster convergence and a better coverage of t his area of the Pareto optimal front. (C) 2001 Elsevier Science Ltd. All ri ghts reserved.