TOPSIS中不同规范化方法的研究

针对不同规模、不同属性值数据范围和类型的多属性决策问题,采用排序一致性指标(RCI)定量地衡量了向量规范化法、极差变换法、线性比例变换法、比重变换法和指数变换法这5种规范化方法对TOPSIS排序一致性的影响.排序一致性指标越大,对应的规范化方法越好.研究结果表明,向量规范化方法能有效处理不同规模、不同属性值数据范围和类型的多属性决策问题,是TOPSIS排序中一种有效的规范化方法.

Investigation of Different Normalization Methods for TOPSIS

Normalizing the decision matrix is an important procedure in the technique for order preference by similarity to ideal solution (TOPSIS). The five well known normalization methods are investigated in this paper, including vector normalization method, max-min method, max/min method, sum method and exponent transformation method. Those methods were compared in terms of their ranking consistency index (RCI) when used with TOPSIS to solve the general multi-attribute decision making (MADM) problem with various problem sizes and data ranges. The higher the RCI is, the better normalization method. The results show that, among the five normalization methods, the vector normalization method is the best one for TOPSIS. It could deal with the general multi-attribute decision making (MADM) problems with various problem sizes, data ranges and attribute types effectively.