计算机应用 ›› 2017, Vol. 37 ›› Issue (2): 540-545.DOI: 10.11772/j.issn.1001-9081.2017.02.0540

• 人工智能 • 上一篇    下一篇

基于后悔理论和证据理论的犹豫模糊决策方法

朱轮   

  1. 常州大学 信息科学与工程学院, 江苏 常州 213016
  • 收稿日期:2016-07-11 修回日期:2016-09-08 出版日期:2017-02-10 发布日期:2017-02-11
  • 通讯作者: 朱轮,3278191370@qq.com
  • 作者简介:朱轮(1974-),男,江苏常州人,高级工程师,硕士,主要研究方向:电子信息技术、智能决策。
  • 基金资助:
    江苏省科技支撑计划(工业)重点项目(BE2013005-3)。

Hesitant fuzzy decision-making method based on regret theory and evidence theory

ZHU Lun   

  1. School of Information Science and Engineering, Changzhou University, Changzhou Jiangsu 213016, China
  • Received:2016-07-11 Revised:2016-09-08 Online:2017-02-10 Published:2017-02-11
  • Supported by:
    This work is partially supported by the Jiangsu Province Science and Technology Support Program (Industrial) Key Project (BE2013005-3).

摘要: 针对属性值为犹豫模糊信息、属性权重和自然状态发生概率完全未知的多属性群决策问题,考虑决策者心理行为,提出一种基于后悔理论和证据理论的多属性群决策方法。首先,运用证据理论计算各自然状态发生的概率;然后,基于区间模糊矩阵、t-分布估计以及得分函数矩阵确定属性信息的效用值,进而依据后悔理论得到每个自然状态下的感知效用矩阵;通过加权算术平均得到综合感知效用矩阵,并依据方案综合感知效用的大小确定方案优劣排序;最后,将所提方法运用于对投资公司的选择实例中。实验结果表明,虽然所提方法与现有方法得到的决策结果相同,但是所提方法在决策过程中只需考虑较少数量的参数。对比分析实验表明,所提方法得到的决策结果合理、可靠,且能反映实际决策情况。

关键词: 犹豫模糊集, 后悔理论, 证据理论, 综合感知效用, 多属性群决策

Abstract: Under the hesitant fuzzy environment, considering the decision makers' psychological behavior, a method based on regret theory and evidence theory was proposed to cope with Multi-Attribute Group Decision Making (MAGDM) problems that the attribute value is hesitant fuzzy information, the attribute weights and probability information of situation are completely unknown. First, evidence theory was utilized to calculate the probability information of the states. Then, based on the interval fuzzy matrices, the estimation of t-distribution and the score function matrices, the utility values of attribute values were determined. Moreover, by using regret theory, the decision makers' perceived utility values were obtained. The overall perceived utility of each alternative was acquired on the basis of the weighted arithmetic mean. After that, all the alternatives were ordered. Finally, the proposed approach was applied to a numerical example about the selection of enterprise. The experimental results show that the proposed method can get the same results as the existing methods, but only a small number of parameters needed to be considersed in the decision process. The results of comparative analysis domonstrate that the decision making results obtained by the proposed method are reasonable and reliable, which can reflect the actual situation.

Key words: Hesitant Fuzzy Set (HFS), regret theory, evidence theory, overall perceived utility, multi-attribute group decision making

中图分类号: