基于多目标种群协同算法的板坯入库优化  

Slab location decision optimization based on multi-objective population cooperative algorithm

作　　者：张琦琪[1,2] 张涛[1,3] 刘鹏[1] 

机构地区：[1]上海财经大学信息管理与工程学院,上海200433 [2]上海科学技术职业学院通信与电子信息系,上海201800 [3]上海财经大学上海市金融信息技术研究重点实验室,上海200433

出　　处：《计算机集成制造系统》2015年第7期1820-1828,共9页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金资助项目(71171126);高等学校博士学科点专项科研基金资助项目(20130078110001);上海市哲学社会科学规划资助项目(2011BGL015);上海市金融信息技术研究重点实验室开放课题资助项目~~

摘　　要：针对钢铁企业板坯入库决策问题,基于出库次序A型约束、分散性约束和垛位限高约束等构建了以板坯综合匹配度、垛位利用度和库存均衡度为目标函数的多目标入库决策优化模型。提出一种多目标种群协同粒子群优化算法,并设计了局部搜索策略以提高外部归档集中Pareto解的多样性,同时利用Pareto最优解改进粒子速度更新方式,达到多种群协同优化的目的。仿真实验证明,该算法可以更好地解决多目标板坯入库优化问题。To solve the slab location decision problem in Iron steel enterprise, a multi-objective optimization model by taking slab comprehensive matching degree, stack utilization degree and inventory load balance degree as objective function was constructed based on A-shaped slab discharged order constraints, dispersive constraints and height limit constraints. A multi-objective population cooperative particle swarm optimization algorithm was proposed, and a local search strategy was designed to improve the diversity of Pareto optimal solutions in outside archive set. The velocity update way of each solution inside the groups was changed by using the outside information from Pareto op- timal solutions, thus the purpose of multi-groups collaborative optimization was achieved. A simulation experiment was carried out to illustrate the validity of the proposed model and algorithm.

关 键 词：板坯入库决策 粒子群优化 多目标优化 PARETO最优解 局部搜索策略 

分 类 号：C934[经济管理—管理学]