Based On Genetic Algorithm For Multi-objective Optimization And Decision-making Methods

Most optimization problems involve multiple conflicting objectives. So solutions of multi-objective optimization problem are a number of trade-off optimal solutions. An ideal multi-objective optimization procedure first find multiple trade-off optimal solutions with a wide range of values for objectives, then choose one of the obtained solutions using higher-level information. Compared with the classical method, the ideal approach is more methodical, more practical, and less subjective.Genetic algorithms (GAs) originate from simulating and researching biology system by computer. GAs can find multiple optimal solutions in one single simulation run. Thus, GAs are ideal candidates for solving multi-objective optimization problems.The main work of this paper is the- investigation of some multi-objective evolutionary algorithms. Early multi-objective evolutionary algorithms do not use any elite-preserving operator, which may violate the elites. In order to carry over the elites to the next generation, we present a number of elitist multi-objective evolutionary algorithms. In the light of the constraints in the multi-objective optimization problem, we present a few of constraint handling strategy. Moreover, we argue some kinds of multi-objective optimization and decision-making technology. At last, we design a compound gear train using some methods presented in the paper.