Abstract: The traditional single-objective algorithm can only get one solution, but the multi-objective algorithm can get a solution set after every run. The algorithm (DEASA) improved the differential evolution strategy, designed reconstruction, adjusted parameter adaptively, adopted arena's principle to build non-dominating set rules, and added simulated annealing strategies into the differential evolution algorithm which enhanced the ability to further improve the performance of the algorithm and reduced the time complexity to avoid falling into local optimum. The experiments show that this algorithm can effectively solve the union transport problem.

Key words: union transport scheduling, difference algorithm, simulated annealing, non-dominated set, multi-objective optimization

摘要： 传统的单目标算法运行一次只能得到一个解,而多目标算法运行一次可以得到一个解集。文中所提算法(DEASA)通过改进差分进化策略,设计重构,调整自适应参数,并采用擂台法则构建非支配集,将模拟退火策略融入到差分进化算法当中,进一步提高了算法的性能,降低了时间复杂度,增强避免陷入局部最优的能力。通过实验验证表明,该算法能有效地解决联盟运输调度问题。

关键词: 联盟运输调度, 差分算法, 模拟退火, 非支配集, 多目标最优化