摘要
建立了基于MTO-MTS的钢厂合同计划的整数规划模型,模型同时考虑库存余材匹配和生产计划,以提前/拖期惩罚、交货时间窗内拖后惩罚、生产费用、库存匹配费用、合同违约惩罚总额最小为目标.根据模型特点,构造了对非可行解进行启发式修复的改进粒子群算法求解策略.仿真实验首先对参数设置进行分析,然后对多组数据进行了结果分析,并在相同条件下,对比了本文模型与分阶段考虑库存匹配/合同计划方法的实验结果,验证了本文模型和算法的有效性.
Based on MTO-MTS ideas for the steel-iron enterprise, this paper constructs an integer programming model for order planning, which considers inventory matching and production planning simultaneously. The objective is to minimize the total cost including earliness-tardiness penalty, later delivery penalty in delivery time window, production cost, inventory matching cost, order cancellation penalty. According to the characteristics of the model, a PSO algorithm with heuristic repaired strategy for infeasible solutions is designed. Using several sets of practical data as instances, this paper analyzes the influence on the results brought by the different parameters in the algorithm and compares the solutions obtained by this method with those obtained by the method considering inventory matching and production planning in different phases. The numerical analysis shows that the model and the algorithm are valid.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2008年第11期85-93,共9页
Systems Engineering-Theory & Practice
基金
国家自然科学基金资助项目(70501018,60773124)
上海市自然科学基金(08ZR1407400)
上海财经大学211三期重点学科项目
关键词
面向订单
面向库存
合同计划
粒子群算法
碾合整数规划
make to order（MTO）
make to stock （MTS）
order planning
particle swarm optimization（PSO）
mixed integer programming
