多产品采购条件下的供应商选择与订购量分配问题研究  

Research on vendor selection & order quantity allocation problem under multi-product purchases

作　　者：关志民[1] 周宝刚[1] 马钦海[1] 

机构地区：[1]东北大学工商管理学院,辽宁沈阳110004

出　　处：《计算机集成制造系统》2005年第11期1626-1632,共7页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金资助项目(70372061)~~

摘　　要：为了解决随机性需求和价格折扣并存条件下的多产品采购供应商选择问题,建立了相应的多目标混合整数随机规划模型。该模型的特点是:①模型的约束条件中兼具确定性和随机性;②通过约束条件方程式准确地表现随机性需求和价格折扣两大假设条件。根据该模型的特殊结构,提出了一种适用的求解策略:首先,通过把机会约束转化为确定性等价类,而将多目标混合整数随机规划模型转化为多目标混合整数规划模型;然后,采用最大满意度法,将体现决策者偏好和目标模糊性的加权模糊多目标混合整数规划模型转化为求解等价的多个单目标混合整数规划问题;最后,在确定每个模糊目标的隶属度函数表达式的基础上求得问题的最优解。另外,通过应用算例说明了模型的有效性和可行性。In order to solve the Vendor Selection Problem （VSP） with stochastic demand and price discount under multi-products purchases, a multi-objective mixed-integer stochastic programming model was constructed. The proposed model had the following characteristics ： constraints of model with both deterministic and stochastic characteristics, and the assumption of stochastic demand and price discount were expressed by constraint equations. According to the special features of the model, an applicable solution strategy was proposed, which involved three steps： （1） by means of transforming stochastic constraints into deterministic equivalents, the multi-objective mixed -integer stochastic programming model was converted into a multi-objective mixed-integer programming model, （2） a weighted fuzzy multi-objectives mixed-integer programming model, which represented decision makers＇ preference and objective fuzziness, was decomposed into several single-objective mixed-integer programming models through maximum satisfaction degree method, and （3） the optimal solution for VSP was derived based on getting membership function of each fuzzy objective. An application example was also given to testify the feasibility and effectiveness of the proposed method.

关 键 词：供应商选择 多产品采购 随机需求 价格折扣 多目标混合整数随机规划 

分 类 号：F273[经济管理—企业管理]