Bi-objective optimisation for scheduling the identical parallel batch-processing machines with arbitrary job sizes, unequal job release times and capacity limits
Bi-objective optimisation for scheduling the identical parallel batch-processing machines with arbitrary job sizes, unequal job release times and capacity limits
复制标题
DOI：
10.1080/00207543.2014.952795
复制
发表时间：
2015-01
影响因子：
9.2
通讯作者：
M. Abedi;Hany Seidgar;H. Fazlollahtabar;Rohollah Bijani
中科院分区：
文献类型：
--
作者：
M. Abedi;Hany Seidgar;H. Fazlollahtabar;Rohollah Bijani
本文研究了每台并行批处理机（bpm）的调度问题，即每台并行批处理机可以同时处理一组作业。本文提出了一种新的BPM双目标混合整数线性规划模型，该模型考虑了制造环境中存在的任意作业大小、不等释放时间和容量限制等现实假设。目标是最小化完工时间和工作的提前和延迟的总加权（及时）。在建立了新的双目标模型的基础上，提出了一种求解该问题的约束方法。这个问题被称为Np-hard。因此，采用快速非支配排序遗传算法（NSGA-II）和多目标帝国主义竞争算法（MOICA）两种多目标优化方法寻找大型问题的帕累托最优前沿。利用响应面法（RSM）对所提算法的参数进行了标定，并对所提算法在不同规模问题上的性能进行了分析，计算结果表明MOICA在解质量和计算时间上优于NSGA-II。
This paper deals the scheduling identical parallel batch-processing machines (BPMs) that each machine can be process a group of jobs as a batch simultaneously. The paper presents a new bi-objective-mixed integer linear programming model for BPM in which arbitrary job size, unequal release time and capacity limits are considered as realistic assumptions occur in the manufacturing environments. The objectives are to minimise the makespan and the total weighted earliness and tardiness of jobs (just in time). After developing a new bi-objective model, an ɛ-constraint method is proposed to solve the problem. This problem has been known as Np-hard. Therefore, two multi-objective optimisation methods, namely, fast non-dominated sorting genetic algorithm (NSGA-II) and multi-objective imperialist competitive algorithm (MOICA) are employed to find the pareto-optimal front for large-sized problems. The parameters of the proposed algorithms are calibrated using Response surface methodology (RSM) and the performances of the proposed algorithms on the problems of various sizes are analysed and the computational results clarify that MOICA outperform than NSGA-II in quality of solutions and computational time.