群决策中基于不同偏好信息的相对熵集成方法

发布时间：2018-06-29 15:59

本文选题：群决策 + 偏好信息　；参考：《东南大学学报(自然科学版)》2005年02期

【摘要】：研究了基于不同形式偏好信息的群决策问题.在群决策中专家根据个人的偏好,常常对决策方案集给出 4种不同形式的偏好信息,包括直接反映决策方案优劣次序的序关系值和效用值,以及 2个决策方案成对比较互反判断矩阵和模糊互补判断矩阵.首先给出了序关系值、互反判断矩阵和模糊互补判断矩阵 3种偏好信息均转化为效用值形式的计算公式, 然后从相对熵的概念出发,提出了一种相对熵最优化的集成模型,给出了模型的解.最后进行了 2个实例分析,结果表明所提出集成方法是有效的.
[Abstract]:The problem of group decision making based on different forms of preference information is studied. In group decision making, according to the individual preference, experts often give four different forms of preference information to the decision plan set, including the order relation value and utility value which directly reflect the order of the decision scheme. And two decision schemes are paired to compare the reciprocal judgment matrix and the fuzzy complementary judgment matrix. In this paper, the calculation formulas of order relation value, reciprocal judgment matrix and fuzzy complementary judgment matrix are given. Then, an integration model of relative entropy optimization is proposed based on the concept of relative entropy. The solution of the model is given. Finally, two examples are given and the results show that the proposed integration method is effective.
【作者单位】：南京大学工程管理学院南京大学商学院
【基金】：国家自然科学基金资助项目 (70101003) 中国博士后科学基金资助项目(2004035209)
【分类号】：C934

【相似文献】