基于不确定优化方法的供应链企业间协同决策研究.doc

文档介绍：
基于不确定优化方法的供应链企业间协同
决策研究
曹鹤婷,左兴权**
5
10
15
20
25
30
35
40
(北京邮电大学计算机学院,北京 100876)
摘要:本论文主要研究不确定需求下供应链的库存协同决策问题。为更准确的模拟供应链末
端的不确定需求,本文采用蒙特卡洛仿真技术建立通用的库存策略评价模型,可以灵活应对
任何类型的不确定需求,极大的克服了前有研究需求局限性。蒙特卡洛仿真模拟是以巨大的
计算消耗为代价的,因此,为平衡蒙特卡洛仿真的计算代价,本论文提出了一种带适应度遗
传的新型粒子群算法,对粒子群算法的适应度遗传技术进行多方面探索,并将其成功应用到
库存协同策略的优化中。实验表明,通过蒙特卡洛,粒子群算法和和适应度技术的融合运用,
能极大提高了供应链库存协同决策的效率,大力提高企业供应链的核心竞争力。
关键词:供应链协同;不确定需求;蒙特卡洛仿真;粒子群算法;适应度遗传
中图分类号:F274
Supply Chain Inventory Collaboration with Uncertain
Demand
CAO Heting, ZUO Xingquan
(Computer School, Beijing University of Posts and munications, Beijing 100876)
Abstract: In this paper, a new algorithm is proposed to model the supply chain inventory
collaboration and find the optimized collaboration scheme with the uncertain customers' demand.
First, Monte Carlo simulation mimicking the behavior of supply chain with uncertain market
demand is used to evaluate a coordination scheme. This evaluation method is able to calculate the
total inventory cost for uncertain demand with any distribution type. Then a fitness inheritance
bined with Monte Carlo simulation is proposed to find an inventory coordination scheme.
Various fitness inheritance techniques are studied to construct an effective fitness inheritance PSO
for inventory coordination. Experiments show that our approach is effective in reducing the
inventory cost of supply chain and sav