AHP 中判断矩阵一致性修正的可信度法

Credit degree methods for correcting judgement matrix into one with complete uniformity

定义了判断矩阵的可信度；基于关联度分析和模糊聚类分析的思想，分别给出了可信度的计算公式，从而提出了将判断矩阵修正成满足完全一致性的可信度法．该法是用公式计算修正后的判断矩阵及其对应于最大特征值ｎ的正特征向量．理论分析和实例计算表明：该法简便、实用。

By defining the definition of the judgement matrix's credit degree and based on the correlate degree analysis and fuzzy clustering analysis, the paper presents new methods for correcting a judgement matrix into a reciprocal matrix with complete uniformity. Both the matrix to be obtained and its positive eigenvector corresponding to the eigenvalue n can be calculated by formulae. The theoretical analyses and the numerical results show that the above methods are effective, and they are the natural generalizations of one of the methods usually used to calculate the eigenvector corresponding to the maximum eigenvalue.