当前位置: X-MOL 学术J. Comb. Optim. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Rescheduling problems with allowing for the unexpected new jobs arrival
Journal of Combinatorial Optimization ( IF 1 ) Pub Date : 2021-08-27 , DOI: 10.1007/s10878-021-00803-4
Xingong Zhang , Win-Chin Lin , Chin-Chia Wu

In modern manufacturing and services, rescheduling frequently happens in manufacturing practice, which a disruption may happen. In this paper, the rescheduling problem with the unexpected new jobs arrival is studied, and the effect of time disruptions is taken into account on a previously planned optimal schedule. Rescheduling means that a set of original jobs has already given jobs sequence by minimizing some classical objective, then a set of new jobs arrived will create a time disruption for the original jobs. Our focus is to find a feasible rescheduling to minimize maximum weighted tardiness costs under a limit of time disruptions from the original schedule. Two time disruption problems will be involved: under an upper restriction of the total time disruption and relaxing time disruption restriction. For the former case, the proof of strongly NP-hard by reducing 3-Partition problem is presented. For latter case, a strongly NP-hard proof is also given,and then a two-approximation polynomial time algorithm is presented to confirm that the latter problem has not an approximation polynomial-time algorithm, which a performance ratio is less than 2 unless \(P=NP\). Finally, one of the proposed algorithms is improved by three local searches, respectively, as three seeds used in simulated annealing for approximate solution. Furthermore, we also present some extensive numerical experiment to evaluate their performance.



中文翻译:

重新安排问题以允许意外的新工作到来

在现代制造业和服务业中,制造实践中经常发生重新安排,这可能会发生中断。在本文中,研究了意外新工作到达的重新调度问题,并考虑了时间中断对先前计划的最佳调度的影响。重新调度意味着一组原始作业已经通过最小化某些经典目标给出了作业序列,然后一组新作业的到来将对原始作业造成时间中断。我们的重点是找到一种可行的重新安排,以在原始时间表中断的时间限制下最小化最大加权延迟成本。将涉及两个时间中断问题:在总中断时间的上限下和放松中断时间限制。对于前一种情况,提出了通过减少 3-Partition 问题的强 NP-hard 证明。对于后一种情况,也给出了一个强NP-hard证明,然后提出了一个二逼近多项式时间算法来证实后一问题没有一个性能比小于2的逼近多项式时间算法,除非\(P=NP\)。最后,提出的算法之一分别通过三个局部搜索进行了改进,分别作为模拟退火中用于近似解的三个种子。此外,我们还提供了一些广泛的数值实验来评估它们的性能。

更新日期:2021-08-27
down
wechat
bug