论文部分内容阅读
针对风险评估中现有指标定权方法主观性较强的问题,从风险权重的概念出发,以港口自然灾害风险的危险性指标为例,通过设定损失分级和对应的指标损失程度范围,提出了“损失期望”概念;基于可获得的历史资料数据,引入防范因子进行指标的分样本客观定权,提高了权重计算方法的科学性和准确性。利用聚类-判别分析证明了该方法合理性与适用性,并将权重计算结果应用于贝叶斯网络建模过程,实现了港口自然灾害风险的量化评估。
In view of the subjectivity of existing weighting method in risk assessment, from the concept of risk weight, take the risk index of natural disasters in port as an example, by setting the range of loss grade and corresponding index loss, The concept of “loss expectation” was introduced. Based on the available historical data, objective sub-sample weighting was introduced by introducing the factor of prevention factor, which improved the scientificity and accuracy of the weight calculation method. By using cluster analysis and discriminant analysis, the rationality and applicability of this method are proved. The weight calculation results are applied to the Bayesian network modeling process, and the quantitative assessment of port natural disaster risk is realized.