AHP中判断矩阵一致性调整方法研究

提出了两种专家判断矩阵一致性调整的新方法:一般的Hadamard凸组合(Easy-HCC)方法和基于系统聚类分析的Hadamard凸组合(HCC)方法.首先利用判断矩阵的生成元生成一致的正互反生成矩阵,前一种方法对生成矩阵作简单的几何平均;后一种方法通过系统聚类分析,对生成矩阵进行一致性聚类,并以此为基础,按少数服从多数的原则分配权重系数,对生成矩阵进行加权几何平均,获得一致的正互反调整矩阵.然后把这两种方法分别与传统的方法相比较,用同一个算例证明了加法凸组合和前一种方法对判断矩阵调整的无效性,并分析了后一种方法的有效性和实用性.

Study on Consistency Regulation for the Judgment Matrix in AHP

Two new methods for regulating the consistency of the judgment matrix are presented. They are the general Hadamard convex combination (Easy-HCC) and clustering analysis-based Hadamard convex combination (HCC). Firstly, the building elements of the judgment matrix were constructed to positive reciprocal consistent building matrices. The first method was to do the geometric mean for the building matrices, and the later was to cluster the building matrices on the consistency with clustering analyzing to them. Based on this and the rule of majority their weight coefficient assigned. Then did the geometric mean with weight for the building matrices and a consistent transformed matrix was gained. At last, the two methods separately with the traditionary methods were compared. An example was used to explain the invalidity and inaccuracy of the add convex combination and the first method and the validity and usefulness of the later.