Abstract: In this paper, we provide a new decision making method for conflict style group decision making problem with triangular fuzzy preference information. During the conflict resolutive stage, we use triangular fuzzy numbers to express experts' decision preference and define the similarity measure between two triangular fuzzy preference vectors. By computing the similarity measures between each expert's preference vector and the group preference vector to define the conflict measure. We give the threshold and consultation mechanism to regulate conflict measure, until all the experts' conflict measure are less than a given threshold, and then enter the decision-making stage. During the decision-making stage, the weight of each attribute is given by the expected function of the triangular fuzzy numbers. By calculating the weighted similarity measure between each alternative and the ideal alternative, the order of all alternatives can be obtained and the best one can be easily selected. Finally, we apply our method to an examples, we give the curves of conflict measure for each alternative by using Matlab, numerical results show that our method is applicable and effective.

Key words: multiple attribute group decision making, triangular fuzzy numbers, similarity measures, conflict measure, consultation mechanism

摘要： 针对三角模糊偏好下冲突型群决策问题,本文提出一种新的决策方法。在冲突消解阶段,用三角模糊数表示决策专家偏好,定义两三角模糊数型偏好矢量间的相似度,通过计算专家对各个方案的偏好矢量与各方案的群偏好矢量间的相似度,以此为基础定义专家的冲突测度。给出阈值和协商机制调控专家的冲突测度,直到所有的专家的冲突测度都小于给定阈值,进入决策阶段。在决策阶段,利用三角模糊数的期望函数确定属性权重,计算各个方案群偏好矢量与理想方案偏好矢量之间的加权相似度,由加权相似度大小排列决策,选出最优方案。最后给出案例应用,利用Matlab画出各方案的冲突测度图,数值结果表明本文方法的可行性及有效性。

关键词: 多属性群决策, 三角模糊数, 相似度, 冲突测度, 协商机制