基于熵权TOPSIS的风险矩阵改进方法研究  

An improved method for the risk matrix research based on the entropy weight TOPSIS

作　　者：张青松[1] 郑薇[1] 

机构地区：[1]中国民航大学安全科学与工程学院,天津300300

出　　处：《安全与环境学报》2013年第6期225-228,共4页Journal of Safety and Environment

基　　金：中央高校基本科研业务费项目(ZXH2012L003)

摘　　要：针对传统风险矩阵法计算结果存在风险结的不足,通过熵权法分别对风险发生可能性和严重性指标赋予不同的权重,引入TOPSIS法对得出的风险值进行精确排序,给出针对风险矩阵的熵权TOPSIS建模方法及计算步骤,提出基于熵权TOPSIS的风险矩阵改进方法。实例分析结果表明,风险结密度由原来的7.5降为0,从而可依据风险精确排序结果合理分配安全资源。The present paper is aimed at improving the risk matrix method for risk analysis by proposing the necessary calculation pro cess of entropy weight and establishing a standardized ranking matrix model. As is known, the risk matrix method is one of the many com monly adopted approaches to assessing the risk liabilities for its being simple in theoretical principle and application in addition to its intu itive results. All of the abovesaid advantages have made it widely used in the military, chemical, transportation and other scientific and industrial fields. However, we have found that there exist three as pects of defects with the method to be overcome. First of all, the tra ditional risk matrix used to treat the risk liabilities of equal importance without discriminating any priority and various weights. Secondly, the traditional method ignores the heterogeneity of the liabilities and severity but merely taking into account the direct likelytobe influen tial factors. And, thirdly, when adopting the method, the assessors of the risk may find it difficult to separate the most important security risks from the junctions of a number of highrisk due to the presence of the risk knot. Seeing the above mentioned defects of the traditional method, we have proposed a revised version of the method by taking into account the necessary inclusion of processing the entropy weight TOPSIS. First of all, it is necessary to set up weights for possibility and severity indicators affecting the safety risk by entropy weights, and establish a standardized ranking matrix. And, next, it is needed to work out the relative distance of the risks and most likely points of the risk indicators with the TOPSIS method, which would finally be converted into the close degrees of the safety risk. After the conver sion, the safety risks could then be ranked according to the evaluation of the close degrees. In so doing, the heterogeneity of the indicators would be no longer ignored as was done in the traditional risk matrix method, hut turned to be a

关 键 词：安全科学技术基础学科 风险矩阵 熵权法 TOPSIS法 风险结 序值 

分 类 号：X913.4[环境科学与工程—安全科学]