随机需求条件下生产-库存系统优化与仿真  

Optimization and Simulation of Production-Inventory System with Stochastic Demand

作　　者：田俊峰[1] 杨梅[2] 

机构地区：[1]西南交通大学交通运输学院,四川成都610031 [2]铁道第二勘察设计院线路处,四川成都610031

出　　处：《系统仿真学报》2004年第11期2522-2524,共3页Journal of System Simulation

摘　　要：针对多周期、多产品、有能力约束动态制造系统的生产-库存问题,考虑随机需求条件和产品的需求满足率,建立以系统总成本最低为目标的二级随机线性规划模型,通过随机模拟法将原问题转化为等价的确定性问题,运用对偶理论和Benders分解法把等价问题分解为相互关联的主问题和子问题,然后分别进行求解。最后的实例仿真结果验证了模型和算法的合理有效性,表明了它们在生产实践中的应用性。The paper studies production-inventory problem in multi-period multi-product capacitated dynamic manufacture system. Considering stochastic demand and satisfying demand ratio of product, it puts forward a two-stage stochastic linear programming model, with the aim of minimizing summation of cost in the system. The primal stochastic problem is transformed into equivalent deterministic problem by stochastic simulation method. Then the equivalent problem is decomposed into interactional master problem and subproblem in accordance with dual theory and Benders decomposition algorithm. Next both decomposed problems are respectively solved. In the last simulation sample, the result validates that the model and algorithm are reasonable and effective, and shows that they are applicable in production.

关 键 词：生产—库存系统 二级随机线性规划 Benders分解算法 对偶理论 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]