一种基于信息熵的混合型多属性决策方法  

Hybrid multi-attribute decision making method based on entropy

作　　者：付沙[1] 肖叶枝[1] 

机构地区：[1]湖南财政经济学院,长沙410205

出　　处：《齐齐哈尔大学学报（自然科学版）》2014年第6期19-23,共5页Journal of Qiqihar University(Natural Science Edition)

基　　金：湖南省情与决策咨询研究课题(2013BZZ78);湖南省教育厅科学研究项目(14C0184)

摘　　要：针对属性值为精确数型、区间型和模糊型的混合多属性决策问题,提出了一种混合型多属性决策方法。该方法首先采用熵权系数法确定各性能指标的权重,以减少传统权重确定方法产生的主观偏差;然后,利用逼近理想解排序法模型对方案优劣排序;最后,假设某一最优产品所对应装备制造商的生产能力有约束等状况,采购方可根据自身情况选择若干家制造商并对其建立单目标的线性规划模型,确定生产数量在所选制造商之间的分配,以期达到总耗资最低、选择价值最大的目的。实例结果验证了该决策方法的可行性和有效性。To the hybrid multi-attribute decision making problems of attribute values were exact number type, interval tape and fuzzy type, we proposed a hybrid multi-attribute decision making method. This method use Method of Entropy Coefficient to determine the performance indexes weights, in order to reduce the subjective bias generated by traditional weights determining method. Then, use TOPSIS sorting method model sort the programs on strengths and weaknesses. Finally, assume that an optimal product that corresponds to equipment manufacturers have production capacity constraints and other conditions, purchaser may choose several manufacturers according to their own situation and establish the single-objective linear programming model to determine the production quantity distribution among the selected manufacturers, to achieve the lowest total cost, select the purpose of largest value. The numerical results demonstrate the feasibility and effectiveness of this decision-making method.

关 键 词：信息熵 混合型多属性决策 装备制造商 区间数 模糊语言 

分 类 号：C934[经济管理—管理学] N949[自然科学总论—系统科学]