摘要
为了求解最小化问题的最大完工时间(Makespan)和最大延迟时间的多目标零等待作业车间调度问题,提出了一种混合分布估计算法。首先,提出了NEH与随机生成并用的初始化种群机制,从而提高初始解的质量。其次,根据概率矩阵模型生成新的个体并通过非劣解集中的非受支配解更新概率矩阵。然后,在局部搜索部分,设计了一种变邻域的搜索机制,从而加强局部搜索能力,提高算法的搜索性能。最后,在仿真实验部分,通过对不同规模标准测试问题的测试,以及与其他算法的对比结果,验证了混合EDA求解多目标零等待作业车间调度问题的有效性。
In order to solve the multi-objective no-wait job shop scheduling problem with makespan and maximum tardiness criteria,a hybrid estimation of distribution algorithm is proposed.Firstly,the initialization population mechanism of NEH and random generation are proposed to improve the quality of the initial solution.Secondly,new individuals can be generated through the probability model and a mechanism is provided to update the probability model by using the non-dominated solutions.Then,a variable neighborhood search mechanism is designed to enhance the local search ability and improve the search performance of the algorithm.Finally,through the test of different scale standard test problems,simulation results and comparisons show the effectiveness of the hybrid estimation of distribution algorithm for the multi-objective no-wait job shop scheduling problem.
作者
姚友杰
钱斌
胡蓉
YAO You-jie;QIAN Bin;HU Rong(Department of Automation,Kunming University of Science and Technology,Kunming 650500,China)
出处
《控制工程》
CSCD
北大核心
2020年第3期418-423,共6页
Control Engineering of China
基金
国家自然科学基金资助项目(51665025,61963022,60904081)。
关键词
零等待作业车间调度
分布估计算法
概率模型
局部搜索
No-wait job shop scheduling problem
estimation of distribution algorithm
probability model
local search
