摘要
针对混合多准则群决策问题,提出一种考虑决策专家风险偏好的混合多准则群决策方法。该方法在数据处理和群决策两阶段都充分考虑风险偏好对决策结果的影响,使得决策结果在考虑风险偏好方面更加全面。在数据处理阶段,将混合型评价值统一化为直觉模糊数,提出一个考虑风险偏好的新记分函数,将直觉模糊数转化为实数。在群决策阶段,首先,引入改进前景理论,风险偏好系数决定专家风险偏好类型,构建不同偏好的改进前景值矩阵;然后,用综合效用最大化模型求解不同偏好的准则权重;其次,基于专家决策信息犹豫度与前景值偏离度两视角确定专家权重,将专家权重与不同偏好的改进前景值矩阵进行结合,形成专家群体改进前景值矩阵;再次,计算各方案的综合前景值。最后,通过算例验证所提方法的有效性和实用性。
For multiple criteria group decision making problems,this paper proposes a considered decision expert mixed multi-criteria group decision-making method of risk appetite.This method in data processing and group decision two phases are fully considering risk preference on the result of decision-making,which makes more comprehensive results in considering risk preference.In the data processing stage,the mixed evaluation value is transformed into intuitionistic fuzzy number.A new score function based on expert risk preference is proposed to transform intuitionistic fuzzy number into real number.In the group decision making stage,firstly,the improved prospect theory is introduced.The risk preference coefficient is used to determine the type of risk preference of experts,and the improved prospect value matrixs of different preferences are constructed.Then,the comprehensive utility maximization model is used to calculate the criterion’s weights of different preferences.Secondly,the weight of experts is determined from two perspectives:the hesitation degree of expert decision-making information and the deviation degree of prospect value.An expert’s group improved prospect matrixs are constructed by combining the expert’s weight with the improved prospect matrix of different preferences.Thirdly,combined with the improved prospect value matrix of the expert’s group and the weight of criteria,the comprehensive prospect value of each scheme is obtained,which is used to sort and compare the schemes.Finally,a numerical example is given to verify the effectiveness and practicability of the proposed method.
作者
武群丽
赵萌
WU Qun-li;ZHAO Meng(Department of Economics and Management,North China Electric Power University,Baoding 071000,China)
出处
《陕西理工大学学报（自然科学版）》
2022年第1期82-92,共11页
Journal of Shaanxi University of Technology:Natural Science Edition
基金
国家社会科学基金资助项目(17BGL252)。
关键词
混合多准则
群决策
风险偏好
直觉模糊数
新记分函数
改进前景理论
mixed multiple criteria
group decision-making
risk appetite
intuitionistic fuzzy number
new score function
improved prospect theory
