一种多粒度语言的多属性群决策方法  

Approach of multiple attributes group decision-making with multi-granularity linguistics

作　　者：张小刚[1] 张亮[1] 王端民[1] 翟楠楠 

机构地区：[1]空军工程大学装备管理与安全工程学院,陕西西安710051

出　　处：《合肥工业大学学报（自然科学版）》2016年第6期781-785,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(71401174)

摘　　要：针对多粒度语言的多属性群决策问题,文章提出了一种基于二元语义及改进多准则妥协解排序(VIKOR)的群决策方法。首先将不同粒度语言的偏好信息一致化为由基本语言评价集表示的相同粒度二元语义信息;在专家属性未知且方案属性不完全的情形下,分别运用有序加权平均算子(2-tuple ordered weighted averaging,T-OWA)与相对熵从客观角度计算权重;为进一步挖掘决策数据的内在规律,引入灰色关联系数改进评判矩阵,结合该矩阵利用VIKOR方法刻画最优方案。算例结果验证了该方法的有效性和可行性。In view of multiple attributes group decision-making problem with multi-granularity linguis-tics ,a new approach based on two-tuple linguistics and VlseKriterijumska Optimizacija I Kompromis-no Resenje（VIKOR） is proposed .Firstly ,the multi-granularity linguistic preference information is u-niformed into the form of two-tuple linguistic information in basic linguistic term set .The unknown-attribute weight information of different expert and the incomplete weight information of attribute are determined by two-tuple ordered weighted averaging （T-OWA ） operator and relative entropy .A new decision matrix ,which is improved by gray relational coefficient to further investigate the inherent law of decision-making data ,is used to characterize the optimal solution in VIKOR method .Finally ,the feasibility and effectiveness of the proposed method are illustrated by the example .

关 键 词：多属性群决策 二元语义 灰色关联矩阵 相对熵 改进多准则妥协解排序 

分 类 号：N945.25[自然科学总论—系统科学]