ABSTRACT
Multiobjective optimization (MOO), or so-called multicriteria optimization, Pareto optimization, involves a problem solution with more than one objective function to be optimized simultaneously. The solutions and/or decisions taken for MOO problems need to consider the trade-off between two or more conflicting criteria. Pareto-based approaches can be historically studied as covering two generations. The first generation is characterized by the use of fitness sharing and niching combined with Pareto ranking. The most representative algorithms from the first generation are the following: Nondominated Sorting Genetic Algorithm (NSGA), Niched-Pareto Genetic Algorithm (NPGA), and Multi-objective Genetic Algorithm (MOGA). The second generation of Multiobjective evolutionary algorithms was born with the introduction of the notion of elitism. In multiobjective extremal optimization, the dominance ranking method is used to evaluate the special fitness, that is, the fitness value of one solution equals to the number of other solutions by which it is dominated, to determine the fitness value for each solution.
