基于改进SPEA2算法的铸造并行车间主生产计划  

Master production plan of parallel casting workshop based on improved SPEA2

作　　者：李海龙 陈发源 计效园[1] 李建斌[2] 周建新[1] LI Hailong;CHEN Fayuan;JI Xiaoyuan;LI Jianbin;ZHOU Jianxin(State Key Laboratory of Material Processing and Die&Mould Technology,Huazhong University of Science&Technology,Wuhan 430074,China;School of Management,Huazhong University of Science&Technology,Wuhan 430074,China)

机构地区：[1]华中科技大学材料成形与模具技术国家重点实验室,湖北武汉430074 [2]华中科技大学管理学院,湖北武汉430074

出　　处：《计算机集成制造系统》2021年第4期1072-1080,共9页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金资助项目(51905188,51775205,71831007)。

摘　　要：针对铸造企业现有人工主生产计划订单交期准确率不高、车间工时利用率低和任务分配公平性不足的问题,建立了以订单提前/拖期惩罚成本、车间完工时间和工作负载均衡为指标的多目标整数规划模型。提出了改进的强度Pareto进化算法,设计了离散形式的编码、交叉与变异操作,并引入模拟退火机制优化环境选择和种群更新方式,求解出并行车间订单排产的Pareto最优解集,然后采用模糊优选法从中选出折中的推荐排产方案。多个规模的仿真实验结果表明,改进后算法所求得的解的支配性明显优于原算法和加权法求得的解,同时最终确定的推荐排产方案与原算法相比,各目标函数值均有效降低,其中惩罚成本和负载均衡目标优化效果显著。In view of the low on-time delivery rate,utilization rate of workshop hours and workload fairness of master production plan with current manual scheduling in casting enterprises,a multi-objective integer programming model with the objectives of order advance/delay penalty,makespan,and workload balance was established.An improved strength Pareto evolutionary algorithm was proposed,the discrete coding,cross and mutation operations were designed,and the simulated annealing mechanism was introduced to optimize the environment selection and population update method,and obtained the Pareto optimal solution set of parallel workshop scheduling.Then the fuzzy optimization method was used to find the recommended compromise solution.Simulation results of multiple scales showed that the dominance of the solution obtained by the improved algorithm was superior to that obtained by the original algorithm and the weighting method,and the final recommended scheduling had achieved a significant improvement in the three objectives,especially for advance/delay penalty and workload balance.

关 键 词：主生产计划 并行车间 强度Pareto进化算法 模拟退火 模糊优选法 铸造 

分 类 号：TP39[自动化与计算机技术—计算机应用技术] TG28[自动化与计算机技术—计算机科学与技术]