摘要
针对区间直觉模糊环境下,识别和修正决策者的不自信行为,以及群体意见相似度不足时的共识优化问题,提出一种考虑直接与间接隶属度和交叉因子的距离测度,并借助距离测度与角度来计算认知偏差,定义了一个基于区间重叠率与面积重叠率的双重相似度。使用K均值聚类将专家集合分成几个子集群,针对子集群中个别专家的不自信行为,提出了一种识别与管理方法。在达成共识过程中,对于相似度低的专家提出了一种调整机制,由此形成了一种多属性群决策方法。通过算例验证了所提方法的可行性。
This study focuses on identifying and correcting non-confident behavior of decision-makers within an interval-valued intuitionistic fuzzy environment,while simultaneously addressing the consensus optimization problem when the similarity of group opinions is insufficient.Firstly,a distance measure incorporating direct and indirect membership degrees and a cross-factor is proposed.Utilizing this distance measure and angles,cognitive deviations are calculated.A dual similarity measure based on interval overlap rate and area overlap rate is defined.Secondly,the expert set is partitioned into several sub-clusters using K-means clustering.A method for identifying and managing non-confident behavior of individual experts within sub-clusters is presented.An adjustment mechanism is proposed for experts exhibiting low similarity during the consensus-reaching process.This forms a multi-attribute group decision-making method.Finally,a numerical example is provided to verify the feasibility of the proposed method.
作者
马东灿
冯涛
张少谱
MA Dongcan;FENG Tao;ZHANG Shaopu(Department of Mathematics and Physics,Shijiazhuang Tiedao University,Shijiazhuang 050043,P.R.China;School of Sciences,Hebei University of Science and Technology,Shijiazhuang 050018,P.R.China)
出处
《重庆邮电大学学报(自然科学版)》
北大核心
2025年第6期857-869,共13页
Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金
河北省自然科学基金项目(A2025208003)
国家自然科学基金项目(62076088)。
关键词
多属性群决策
区间直觉模糊集
距离测度
相似度
共识达成
multi-attribute group decision-making
interval-valued intuitionistic fuzzy set
distance measure
similarity measure
consensus reaching
