A Novel Weight-Based Immune Genetic Algorithm for Multiobjective Optimization Problems

The weight-based multiobjective evolutionary algorithms have been criticized mainly for the following aspects: (1) difficulty in finding Pareto-optimal solutions in problems having nonconvex Pareto-optimal region, and (2) non-elitism approach for most; cases, and (3) difficulty in generating uniformly distributed Pareto-optimal solutions. In this paper, we propose a weight-based multiobjective immune genetic algorithm(MOIGA), which alleviates all the above three difficulties. In this proposed algorithm, a randomly weighted sum of multiple objectives is used as a fitness function. An immune operator is adopted to increase the diversity of the population. Specifically, a new mate selection approach called tournament selection algorithm with similar individuals (TSASI) and a new environmental selection approach named truncation algorithm.,with similar individuals (TASI) are presented. Simulation results show MOIGA outperforms NSGA-II and RWGA.