...
首页> 外文期刊>International Journal of Production Research >Hybrid simulated annealing with memory: an evolution-based diversification approach
【24h】

Hybrid simulated annealing with memory: an evolution-based diversification approach

机译:带有内存的混合模拟退火:基于进化的多样化方法

获取原文
获取原文并翻译 | 示例
           

摘要

This study presents an efficient metaheuristic approach for combinatorial optimisation and scheduling problems. The hybrid algorithm proposed in this paper integrates different features of several well-known heuristics. The core component of the proposed algorithm is a simulated annealing module. This component utilises three types of memories, one long-term memory and two short-term memories. The main characteristics of the proposed metaheuristic are the use of positive (reinforcement) and negative (inhibitory) memories as well as an evolution-based diversification approach. Job shop scheduling is selected to evaluate the performance of the proposed method. Given the benchmark problem, an extended version of the proposed method is also developed and presented. The extended version has two distinct features, specifically designed for the job shop scheduling problem, that enhance the performance of the search. The first feature is a local search that partially explores alternative solutions on a critical path of any current solution. The second feature is a mechanism to resolve possible deadlocks that may occur during the search as a result of shortage in acceptable solutions. For the case of job shop scheduling, the computational results and comparison with other techniques demonstrate the superior performance of the proposed methods in the majority of cases.
机译:这项研究为组合优化和调度问题提出了一种有效的元启发式方法。本文提出的混合算法融合了几种著名的启发式算法的不同功能。该算法的核心部分是模拟退火模块。该组件使用三种类型的存储器,一种是长期存储器,两种是短期存储器。拟议的启发式方法的主要特征是使用正面(增强)记忆和负面(抑制性)记忆,以及基于进化的多元化方法。选择作业车间调度以评估所提出方法的性能。给定基准问题,还开发并提出了所提出方法的扩展版本。扩展版本具有两个独特的功能,这些功能专门为解决车间调度问题而设计,可增强搜索的性能。第一个功能是本地搜索,部分搜索任何当前解决方案的关键路径上的替代解决方案。第二个功能是解决由于可接受的解决方案短缺而在搜索过程中可能发生的死锁的机制。对于车间调度的情况,计算结果以及与其他技术的比较证明了在大多数情况下所提出方法的优越性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号