摘要
在置换流水加工环境下,以最小化生产流程时间为目标制定的初始加工方案,由于新工件的到达变得不再最优或不可行,为了降低对原始加工方案的影响,在权衡生产成本和扰动成本的情况下,建立双目标重调度干扰管理模型,对初始最优方案进行调整。针对该模型的特点和问题复杂度,结合微粒群算法强大的全局搜索能力,以及非支配排序遗传算法(NSGA-Ⅱ)获得的Pareto解优良的综合性能,提出了一种混合微粒群算法来对问题求解。通过求解经典文献中置换流水车间双目标问题和随机生成的置换流水车间新工件到达问题,结果表明混合算法要优于NSGA-Ⅱ和多目标微粒群算法(MOPSO),同时验证了求解置换流水车间干扰管理问题的有效性。
In Permutation flowshop scheduling, the initial schedule is obtained via minimizing the makespan. A set of new arrival iobs make the initial schedule not optimal or feasible. Initial scheduling should be revised to trade off the original objective and deviation cost, and a hi-objective disruption management model is built up. Given the characteristics of the model and the complication of the problem, with concern of the Particle Swarm Algorithm with strong global search ability and the Pareto solutions with excellent comprehensive properties obtained by the Non-dominated Sorting Genetic Algorithm Ⅱ （NSGA-Ⅱ ）, we propose the Hybrid Particle Swarm Optimization algorithm （HPSO） to obtain the （ap- proximate） optimal solutions. By solving bi-objective flow shop problems in classic literatures and ran- domly generated flowshop problems with new arrival orders, the proposed HPSO outperforms NSGA-Ⅱ and MOPSO is verified as an effective approach to coping with disruptions.
出处
《石家庄铁道大学学报（自然科学版）》
2016年第1期86-92,共7页
Journal of Shijiazhuang Tiedao University(Natural Science Edition)
基金
国家自然科学基金(71271039)
