基于理想点法的语言评价矩阵评判专家水平研究

Study on the Assessment Level of Experts Based on Ideal Point of Linguistic Assessment Matrices

针对群决策中语言评价矩阵形式偏好信息的评判专家水平问题,提出了一种分析方法.首先,给出了有关语言评价矩阵的若干定义及性质;然后,通过定义专家与专家群体判断之间的偏差距离矩阵、方案集的正理想点和负理想点,根据专家判断与理想点之间的贴近度,给出了基于语言评价矩阵形式偏好信息的评判专家水平的分析方法,最后通过一个算例说明了所提出的分析方法的可行性.本方法使决策结果更加可靠、可信,可作为评判专家水平的备选方法.

With respect to the problem of assessing level of experts based on the linguistic assessment matrices inform of preference information,an analytic method is proposed to define the linguistic assessment matrices with their natures revealed. Then,by defining the deviation distance matrices between individual and grouped experts＇ assessment decision matrices and positive/negative ideal points of alternative set,an analytic method is given to assessment level of experts based on TOPSIS（technique for order preference by similarity to ideal solution）. A numerical example is given to illustrate the feasibility of the proposed approach which will results in more reliable and credible decision-making and can be taken as an alternative way to assess expert level.