基于仿真的双目标并行机开放车间自适应动态调度  

Simulation-Based Adaptive Dynamic Scheduling for Bi-objective Parallel Multi-processor Open Shop

作　　者：陈亚绒[1] 管舒晨 黄成军 朱立夏 周富得 Chen Yarong;Guan Shuchen;Huang Chengjun;Zhu Lixia;Chou FuhDer(College of Mechanical and Electronic Engineering,Wenzhou University,Wenzhou 325000,China)

机构地区：[1]温州大学机电工程学院,浙江温州325000

出　　处：《系统仿真学报》2023年第1期69-81,共13页Journal of System Simulation

基　　金：国家自然科学基金(51705370)。

摘　　要：针对工件到达时间与加工时间不确定,且存在紧急工件的并行机开放车间调度问题,以TWC(total weighted completion time)与TWT(total weighted tardiness)为优化目标,设计了一种集成FlexSim仿真模型与NSGA-Ⅱ算法的自适应动态调度方法。该方法以FlexSim模型仿真工件的生成和加工为基础,根据车间实时负荷确定动态调度周期,对紧急工件进行右移重调度,利用NSGA-Ⅱ算法生成双目标优化的调度方案。某晶粒拣选车间生产数据的实验结果表明,相较于利用规则的实时动态调度与固定周期重调度,提出的方法能够在最小化调度偏离度的同时获得满意解。Aiming at the parallel multi-processor open shop scheduling problem with uncertain job’s release time, processing time and urgent jobs, an adaptive dynamic method integrating Flex Sim simulation model and NSGA-Ⅱ algorithm is designed to optimize the bi-objectives of TWC(total weighted completion time)and TWT(total weighted tardiness). By using the Flex Sim simulation model, this method determines the adaptive scheduling cycle according to the dynamic workload of the open shop, and conducts right-shift rescheduling to the urgent jobs. NSGA-Ⅱ algorithm is used to generate the bi-objective optimization scheduling scheme. Experimental results of a grain sorting shop show that compared with the rule-based real-time dynamic scheduling and periodic rescheduling, the proposed method can obtain a satisfactory solution while minimizing the deviation degree between the dynamic scheduling and the original scheduling.

关 键 词：并行机开放车间 NSGA-Ⅱ算法 动态调度 工件随机到达 仿真模型 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]