摘要
针对多品种小批量模式下离散制造车间调度问题,以柔性流水车间为例,建立了多品种柔性流水车间调度模型,结合模型特点,提出了一种改进的候鸟算法。设计了一种基于产品种类编号的单层编码规则,提出了基于反向学习的种群初始化策略,同时保证了初始种群的多样性及质量,针对插入和交换产生邻域结构的不足,提出了一种基于最优插入和最优交换方式的邻域结构,并采用了基于动态邻域搜索的局部搜索策略,增强IMBO算法的局部搜索能力,最后,构建了多组适用于多品种柔性流水车间调度模型的算例,通过对比实验,验证了所提的改进候鸟算法的有效性和高效性。
Aiming at the scheduling problem of the discrete manufacturing workshop under the multi-variety and small-batch mode,considering the characteristics of the model,an improved migratory bird algorithm and a scheduling model have been brought up.A single-layer coding rule based on product category numbers is designed,a population initialization strategy based on reverse learning is proposed.The diversity and quality of the initial population are guaranteed at the same time.Due to the insufficient neighbourhood structure which generated by insertion and exchange,a new strategy is proposed.A neighbourhood structure based on the optimal insertion and optimal exchange method,and a dynamic neighbourhood searching strategy have been developed to enhance the local searching capability of the IMBO algorithm.Finally,multiple groups were constructed for multi-variety flexible flow shop scheduling.Through comparative experiments,the models verified the effectiveness and efficiency of the developed migratory bird algorithm with improved performance.
作者
唐红涛
杨源
闻婧
TANG Hongtao;YANG Yuan;WEN Jing(School of Mechanical and Electronic Engineering,Wuhan University of Technology,Wuhan 430070,China;Hubei Key Laboratory of Digital Manufacturing,Wuhan 430070,China)
基金
国家自然科学基金资助项目(51705384,52075401)
关键词
多品种小批量
离散制造车间
反向学习
候鸟算法
multi-variety and small-batch
discrete manufacturing workshop
reverse learning
migratory bird algorithm
