基于正态分布区间数的信息不完全的群决策方法  

Approach of group decision making based on normal distribution interval number with incomplete information

作　　者：汪新凡[1] 肖满生[1] 

机构地区：[1]湖南工业大学理学院,湖南株洲412008

出　　处：《控制与决策》2010年第10期1494-1498,1506,共6页Control and Decision

基　　金：国家自然科学基金项目(60874025)

摘　　要：针对属性值为正态分布区间数而属性权重信息不完全的多属性群决策问题,定义了一些新的集成算子,即正态分布区间数的加权算术平均(NDINWAA)算子、正态分布区间数的有序加权平均(NDINOWA)算子和正态分布区间数的混合加权平均(NDINHA)算子,进而提出一种基于正态分布区间数的信息不完全的多属性群决策方法.该方法利用NDINWAA算子和NDINHA算子对正态分布区间数属性值进行集成,利用正态分布区间数属性值的方差,通过建立优化模型确定最优属性权重,利用期望-方差准则对方案进行排序并择优.实例分析表明了该方法的可行性和有效性.For multiple attribute group decision making problems,in which the attribute values are normal distribution interval numbers and the attribute weight information is incomplete,some new aggregation operators are defined,such as the normal distribution interval number weighted arithmetic averaging（NDINWAA） operator,the normal distribution interval number ordered weighted averaging（NDINOWA） operator and the normal distribution interval number hybrid weighted averaging（NDINHA） operator.Then an approach is developed for solving multiple attribute group decision making based on normal distribution interval number with incomplete information.In this method,normal distribution interval number attribute values are aggregated by the NDINWAA operator and the NDINHA operator,some optimal models are constructed to determine the optimal attribute weights by using the variance of normal distribution interval number attribute values,and ranking of alternatives is performed by using expectation-variance principle.Finally,an example shows the effectiveness of this method.

关 键 词：群决策 正态分布区间数 信息不完全 集成算子 

分 类 号：C934[经济管理—管理学]