属性权重动态更新的自适应群体共识决策方法  

Adaptive group consensus decision-making method with dynamic update of attribute weights

作　　者：庞继芳[1] 侯治国 宋鹏[2] 张超[1] PANG Jifang;HOU Zhiguo;SONG Peng;ZHANG Chao(School of Computer and Information Technology,Shanxi University,Taiyuan 030006;School of Economics and Management,Shanxi University,Taiyuan 030006,China)

机构地区：[1]山西大学计算机与信息技术学院,山西太原030006 [2]山西大学经济与管理学院,山西太原030006

出　　处：《智能系统学报》2024年第4期941-951,共11页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金项目(62006148,72171137,62272284).

摘　　要：为了提高不确定语言环境下异构多属性群决策的质量、效率及可解释性,提出一种属性权重动态更新的自适应群体共识决策方法。首先,定义不确定语言变量与中间值之间的转换函数,发展多级共识测度,建立计算属性初始权重的双目标优化模型;进而,构建包含群体共识自动达成规则和属性权重动态更新机制的自适应共识模型,实现待调整值的精准定位和自动修改,在优化属性权重的同时进一步提升群体共识水平;然后,对于达成共识的群体决策矩阵,先利用转换函数将中间值转换为不确定语言变量,再使用属性权重和集结算子得到各方案的综合评价结果;最后,通过供应商选择实例和实验比较分析验证所提方法的有效性和可行性。本文研究结果为灵活、高效地求解复杂环境下的多属性群决策问题提供了有效途径。To improve the quality,efficiency,and interpretability of heterogeneous multi-attribute group decision making in an uncertain linguistic environment,an adaptive group consensus decision-making method with dynamic update of attribute weights is proposed.It first defines the conversion functions between uncertain linguistic variables and median values,and develops the multi-level consensus measures,then calculates the initial weights of attributes by establishing a bi-objective optimization model,and further establishes the group consensus automatic reaching rules and attribute weight dynamic update mechanism to achieve accurate positioning and automatic modification of the values to be adjusted,further improving the level of group consensus while optimizing attribute weights.Furthermore,for the group decision matrix that has reached consensus,the median values in the matrix are first converted into corresponding uncertain linguistic variables by using the conversion function,and then the optimized attribute weight and the uncertain linguistic weighted average operator(ULWA)are used to calculate the overall evaluation results,and the alternatives are ranked and selected by using the dominance index.Finally,the effectiveness and feasibility of the proposed method are verified through the case of supplier selection and experimental comparative analysis.The proposed method has the characteristics of wide application scope,high consensus efficiency,flexibility,practicality,and strong interpretation,which provides an effective approach for solving multi-attribute group decision-making problems in complex environments.

关 键 词：异构多属性群决策 不确定语言变量 转换函数 多级共识测度 属性权重 双目标优化 自适应共识模型 可解释性 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]