Abstract: Aimed at the problem of comprehensive evaluation to pull the greatest degree of difference between evaluation objects, through analyzing the limitations existing in linear deviation maximization method, this paper proposes a nonlinear deviation maximization method based on comprehensive judgment of subject and object information,in order to further enrich and improve the theoretical knowledge of deviation maximization method. Firstly, we give a brief introduction to the linear deviation maximization method and analyze the limitation. Secondly,we develop the research hypotheses and give the nonlinear factor selection principle and the method to determine the index order,and give the method to select the adjacent index important degree ration after ordering. According to the principle of evaluation, we use the planning model to calculate every index’s weight,and also use the nonlinear aggregation model to calculate the final evaluation results. Finally, we give a numerical example to verify the effectiveness of the method.

Key words: comprehensive evaluation, linear deviation maximization method, nonlinear factors, planning model

摘要： 针对以最大程度的拉开被评价对象间差异的综合评价问题,通过分析线性拉开档次法存在的局限性,本文提出了一种基于主客观信息综合判断的非线性拉开档次法,旨在进一步丰富和完善拉开档次法的理论知识。首先对线性拉开档次法进行简单的介绍,并分析了线性拉开档次法的局限性;然后提出前提假设,给出非线性因子选取的原则以及确定指标序关系的方法,并给出排序后相邻指标重要程度比值的选取方法,根据评价原则利用规划模型求解各评价指标的权重系数,利用求得的指标权重采用非线性集结模型计算最终的评价结果;最后用一个算例验证了方法的有效性。

关键词: 综合评价, 线性拉开档次法, 非线性因子, 规划模型