考虑属性关联的TOPSIS语言群决策方法

New TOPSIS Linguistic Group Decision Making Method Considering Interactions between Attributes

针对在实际的多属性语言群决策问题中属性之间往往具有相互关联作用这一问题,提出了一种考虑属性关联的TOPSIS语言群决策方法.该方法中决策者给出对方案的语言评价信息,并采用二元语义概念对语言评价信息进行处理.在传统理想点法的基础上引入λ模糊测度的概念,根据属性间的关联作用确定每个方案属性的权重并得到加权决策矩阵.然后计算每个方案与正、负理想点间在n维二元语义空间内的二元语义欧氏距离并得到相对贴近度,确定最优方案.该方法有效地避免了决策过程中语言评价信息的扭曲和损失,并能够考虑属性间的关联作用对于属性权重的影响,使得决策结果更加合理可信.通过一个实例验证了方法的有效性和实用性.

Interactions often exist between some attributes in multi-attribute linguistic group decision making. To solve this problem, a new TOPSIS linguistic group decision making method was proposed. In this method, decision makers give linguistic assessment information to different alternatives, and the two- tuple linguistic representation is used to aggregate the linguistic assessment information. The concept of λ fuzzy measures is introduced to the traditional TOPSIS method to describe the interactions between attributes. For each alternative, the weights of attributes are determined by considering the influence of interactions between them. Then weighted decision making matrix is obtained, and the two-tuple linguistic eu- clidean distance between each alternative and positive/negative ideal solution is calculated. Then the rela- tive closeness coefficients can be obtained, and the optimal alternative can be determined. The new method not only avoids information distortion and loss which occur formerly in the linguistic information processing, but also considers the influence of interactions between attributes on their weights. Consequently, more reasonable results can be obtained on the basis of this algorithm. An example was given to validate the proposed method.