The degree of consistency of judgment matrix can reflect the level of consistency of expert thinking, which is an impor-tant factor to determine the weight of experts based on judgment matrix. Aiming at the problem that the traditional Euclidean dis-tance method is not concerned with the relative difference between elements, and the measure of similarity degree ( or degree of difference) is easy to be affected by the 1-9 scale, the expert weight determination method based on the consistency degree of judgment matrix was studied. The expert weight determination method based on Euclidean distance and grey correlation degree was proposed, and the method of expert weight determination based on relative distance was also proposed. In the former, the rel-ative value of the consistency degree of the expert judgment matrix was calculated by introducing the grey correlation degree;and in the latter, the relative distance method was used to reduce the influence of 1-9 scale. In the example analysis, it is proved that the order of expert weighting obtained by the two proposed methods is reasonable compared with the traditional Euclidean distance method, and the different characteristics of the proposed methods are analyzed, which provides reference for the method selection in practical application in the future.%判断矩阵的一致性程度可以反映专家思维的一致性水平,是基于判断矩阵确定专家权重的一个重要因素.针对传统欧氏距离方法对元素间相对差异关注不够,以及判断矩阵相似度(或差异程度)的度量易受1-9标度影响的问题,研究基于判断矩阵一致性程度的专家权重确定方法,分别提出基于欧氏距离与灰色关联度的专家权重确定方法和基于相对距离的专家权重确定方法.前者通过引入灰色关联度,计算得到的是各专家判断矩阵一致性程度比较意义上的相对值;后者通过采用相对距离方法来减弱1-9标度的影响.算例分析中,基于一致性检验指标CR验证了本文2种方法得到的专家权重排序较传统欧氏距离方法得到的排序结果合理,并分析了各种方法在专家权重具体取值上的不同特点,为今后实际应用中的方法选择提供参考.
展开▼
