带市场选择的联合库存选址模型  

Joint Inventory-Location Models with Market Selection

作　　者：唐凯[1] 杨超[1] 杨珺[1] 

机构地区：[1]华中科技大学管理学院,湖北武汉430074

出　　处：《工业工程与管理》2007年第5期5-10,共6页Industrial Engineering and Management

基　　金：国家自然科学基金资助项目(7047104270601011);教育部新世纪优秀人才支持计划项目(NCET-06-0653)

摘　　要：假设存在一个有供应能力限制的供应商和一个需求量和收益都不确定的潜在需求市场集。选择进行服务的需求市场,然后根据风险分摊的效益原理,确定将哪些需求市场建立为配送中心,如何分配这些配送中心对其他被选择的需求市场进行服务。其目标为使损失的总收益、配送中心的建立成本、运输成本和库存成本之和最小。所以,构建了一个非线性的整数规划模型。由于该模型属于NP-Hard问题,因此提出了一种基于拉格朗日松弛和子梯度法的启发式算法,并通过算例对该算法进行了研究,得到的最优解的上下界相对误差最大不超过5%,大多数小于1%。Instead of the traditional model, given a single capacitated supplier and a set of markets, each specified by a stochastic demand and associated with expectation revenue. Decisions were made in two stages： first made a decision of what markets to select, where all other markets were rejected. Next in terms of risk-pooling benefits theory, determined which markets should serve as distribution centers and how to allocate the other selected markets to these distribution centers. The objective was to minimize the total losing revenue, inventory, transportation, and facility location costs. This problem could be formulated as a nonlinear integer-programming model. This model is NP-Hard, so a heuristic solution approach was presented, based on Lagrangian relaxation and the subgradient method. Finally, Numerical examples were given to illustrate the application of this algorithm, and show that the upper bounds of errors are all lower than 5%, and most even lower than 1% .

关 键 词：选址 库存 市场选择 拉格朗日松弛 

分 类 号：F224[经济管理—国民经济]