基于拉格朗日松弛算法的分布式供应链优化  

Coordination Method for Distributed Supply Chain Planning Based on Lagrangian Relaxation

作　　者：周威[1] 金以慧[1] 

机构地区：[1]清华大学自动化系,北京100084

出　　处：《控制工程》2006年第2期130-134,共5页Control Engineering of China

基　　金：国家自然科学基金资助项目(60174046)

摘　　要：为解决分布环境下的无协调中心的供应链生产计划的协调问题,提出了一种基于拉格朗日松弛算法的折扣价格协调优化策略。针对企业计划只能基于本地信息的特点,利用拉格朗日松弛算法将企业之间的物料耦合约束松弛掉,从而把整个供应链计划问题分解为多个可利用本地信息求解的企业生产计划子问题。通过上下游企业之间对折扣价格(拉格朗日算子)的异步更新,可以逐步获取整个供应链生产计划的优化解,从而实现分布环境下的供应链生产计划的异步协调。仿真实验证明了该方案的可行性。To the supply chain planning problem without a coordination center, a decentralized asynchronous coordination method based on Lagrangian relaxation algorithm is presented. By relaxing the material flow balance constraints among the enterprises, the whole supply chain planning problem is decomposed into multiple single-enterprise planning sub-problems, which can be solved with the local information, So, each enterprise production planning model can be set up and solved independently with the discount-prices, i.e. Lagrangian multipliers, which are given by the upstream or downstream enterprise. To obtain the feasible solution, a distributed heuristics algorithm is proposed. During the coordination process, through iteratively updating the discount-prices among the enterprises, the near optimal solution can be achieved. The computational experiments show that the. method can solve the supply chain planning problem eflqciently.

关 键 词：供应链 协调 生产计划 拉格朗日松弛 代理次梯度 

分 类 号：TP14[自动化与计算机技术—控制理论与控制工程]