互联网 qkzz.net
全刊杂志网:首页 > 女性 > 文章正文
刊社推荐

求解多目标job-shop生产调度问题的量子进化算法覃朝勇 刘 向 郑建国


摘 要:基于量子计算理论和进化理论,提出了用于多目标job-shop优化的量子进化算法(QEA-MOJSP)。QEA-MOJSP采用量子比特来表示工序对加工顺序的优先概率,利用量子叠加和相干机理,通过更新和交叉操作完成进化过程。对所有机器上工序对优先概率进行观测可得到一个调度方案,修补算子被用于不可行调度方案的修补。设计了局部搜索算子用于开采当代最优个体周围的解空间,以提高算法的收敛速度。实验结果表明,对于测试算例,QEA-MOJSP的解接近Pareto最优解集前沿,并具有很好的多样性。
  关键词:多目标作业车间调度; 量子进化算法; 局域搜索
  中图分类号:TP18 文献标志码:A
  文章编号:1001-3695(2010)03-0849-04
  doi:10.3969/j.issn.1001-3695.2010.03.011
  
  Quantum-inspired evolutionary algorithm for multi-objective job-shop scheduling
  
  QIN Chao-yong1,2, LIU Xiang1, ZHENG Jian-guo2
  
  (1.School of Mathematics & Information Science, Guangxi University, Nanning 530004, China; 2.School of Business & Management, Donghua University, Shanghai 200051, China)
  
  Abstract:This paper proposed a quantum-inspired evolutionary algorithm for multi-objective job-shop scheduling problems(QEA-MOJSP). In the QEA-MOJSP, employed a quantum bit to represent processing priority of two operations executed on the same machine. Used updating operator of quantum gate to speed up individuals converge toward the current best solution. Performed conventional crossover as well. Employed quantum computation mechanics to accelerate evolution process.A scheduling solution could be obtained by observing quantum chromosome on all machines. To repair illegal solution, employed harmonization algorithm. At last, designed local search operator to exploit the space around the current best solution. Experiments are conducted on benchmark test problems, the results show that the proposed approach can search for the near-optimal and non dominated solutions by optimizing the makespan and mean flow time. ......
很抱歉,暂无全文,若需要阅读全文或喜欢本刊物请联系《计算机应用研究》杂志社购买。
欢迎作者提供全文,请点击编辑
分享:
 

了解更多资讯,请关注“木兰百花园”
分享:
 
精彩图文


关键字
支持中国杂志产业发展,请购买、订阅纸质杂志,欢迎杂志社提供过刊、样刊及电子版。
关于我们 | 网站声明 | 刊社管理 | 网站地图 | 联系方式 | 中图分类法 | RSS 2.0订阅 | IP查询
全刊杂志赏析网 2017