摘要
为了利用粒子群优化算法解决作业车间调度问题,提出了将调度问题转化为连续优化问题的有效策略;设计了Pareto档案粒子群算法(PAPSO),该算法将档案维护和全局最好位置选取结合在一起,在档案维护过程中为每个粒子选取全局最好位置;给出了变异与PAPSO的结合新策略.将PAPSO和带变异的PAPSO应用于15个调度实例,以最小化总拖后时间和最大完成时间,与强度Pareto进化算法2等进行比较,结果验证了PAPSO在作业车间调度方面的良好性能.
To solve job shop scheduling problem by using particle swarm optimization, this paper proposed an effective strategy to convert the scheduling problem into the continuous optimization problem and designed Pareto archive particle swarm optimization （PAPSO） in which the archive maintenance is combined with global best position selection and the global best position is selected for each particle in the process of archive maintenance. The incorporation of mutation operator into PAPSO was also discussed, Finally, it applied the PAPSO and the PAPSO with mutation to 15 scheduling instances for simultaneously minimizing makespan and total tardiness and compared the above two algorithms with the strength Pareto evolutionary algorithm 2 etc. The computational results demonstrate the good performance of the PAPSO in job shop scheduling.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2007年第11期1796-1800,共5页
Journal of Shanghai Jiaotong University
基金
湖北省自然科学基金资助项目(2007ABA332)
关键词
粒子群优化
多目标
作业车间调度
particle swarm optimization （PSO）
multi-objective
job shop scheduling
