摘要
多属性决策环境下决策属性之间难以保持绝对独立,属性关联可能会对决策结果产生影响,辨识其影响过程和影响结果可为科学决策提供支撑。为此,首先分析属性关联模糊测度模型,采用菱形成对比较(diamond pairwise comparison,DPC)法确定新的属性关联系数。其次,基于ELECTRE(elimination et choice translating reality)方法,分析属性关联对非一致性指数影响的几种典型情形。再次,全面分析3对属性两两关联情形下非一致性指数的变化条件,并给出DPC-ELECTRE多属性决策步骤。最后,通过算例证明该方法过程及其应用的有效性和优越性。
In multi-attribute decision-making environment,it is difficult to maintain absolute independence among decision attributes,and attribute correlation may affect the decision-making result.Identifying the influence process and result can provide support for scientific decision-making.Therefore,a fuzzy measurement model of attribute association is analyzed firstly.And uses diamond pairwise comparisons(DPC)is used to determine new attribute association coefficients.Secondly,based on the elimination et choice translating reality(ELECTRE)method,several typical situations of the impact of attribute association on non-consistency index are analyzed.Thirdly,the change condition of the non-consistency index value under the pairwise association of three pairs of attributes is analyzed comprehensively,and the multi-attribute decision steps of DPC-ELECTRE are given.Finally,an example is given to illustrate the effectiveness and superiority of the method and its application.
作者
徐天琦
江文奇
朱力
杨珊珊
王嘉丽
XU Tianqi;JIANG Wenqi;ZHU Li;YANG Shanshan;WANG Jiali(School of Economics and Management,Nanjing University of Science and Technology,Nanjing 210094,China;Jiangsu Industrial Cluster Decision-Making Consulting and Research Base,Nanjing 210094,China)
出处
《系统工程与电子技术》
北大核心
2025年第1期202-209,共8页
Systems Engineering and Electronics
基金
国家自然科学基金(71971117)
教育部人文社会科学研究规划基金(17YJA630035)资助课题。
关键词
多属性决策
属性关联
菱形成对比较方法
ELECTRE方法
multi-attribute decision-making
attribute associations
diamond pairwise comparison(DPC)method
elimination et choice translating reality(ELECTRE)method
作者简介
徐天琦(1990-),女,博士研究生,主要研究方向为多属性决策、模糊决策;通讯作者:江文奇(1975-),男,教授,博士研究生导师,博士,主要研究方向为复杂决策;朱力(1986-),男,博士研究生,主要研究方向为物流与供应链管理、管理评价与决策技术;杨珊珊(1993-),女,博士研究生,主要研究方向为多属性群决策;王嘉丽(1994-),女,博士研究生,主要研究方向为多属性决策、模糊数据集。