基于Kriging模型的供应链优化设计

供应链优化的目的之一是确定使得总成本最小的最佳运作水平.供应链系统是复杂的动态系统,由于库存系统的复杂性和供应链本身的不确定性,利用传统优化方法往往需要耗费一定的计算成本和经济成本.而元模型则能以简单的数学表达式较精确地刻画仿真系统的输入输出关系,为研究者分析复杂系统提供了一种分析方法.针对供应链优化问题,给出了一种基于计算机试验设计中的元模型一Kriging模型的供应链优化方法,并通过一个三级供应链问题对所提方法进行了实证研究.研究结果表明了所提方法的有效性和可用性,为供应链优化提供了一种新的研究思路.

Optimization Design of the Supply Chain Network Based on Kriging Meta-model

One of the most important dimensions of the supply chain optimization is to minimize the total costs via determining its optimal operating conditions.However,supply chain network is a complex and dynamic system due to the complexity of inventory system and the underlying uncertainty,the traditional optimization methods are often requires too much calculation cost and economic cost.In order to resolve such difficulties,the optimization method based on meta-model appears.Meta-model is approximated to the simulation model,which can show the basic properties of the target system and expose accurately the system's input-output relationship through simple mathematical expression.In this paper,Kriging model is used to expose the function relationship between input variables (or controllable factors),e.g.Capacity,reorder policy,lead time,reliability of suppliers and output variables (or responses) such as the total cost.A three-echelon supply chain is given to illustrate the efficiency of the proposed approach,it is implemented more simply and more practically.It can also amplify the idea of optimization of supply chain and provide certain theoretical basis and decision-making basis for managers.