基于对抗交叉评价模型的中国自然灾害区域脆弱性评价

摘要

The Aggressive-Cross-Evaluation model was first proposed for the assessment of regional vulnerability to natural disasters in this article. This paper constructed the assessment model of regional vulnerability to natural disasters from three aspects, including the danger of regional disasters, the exposure of regional hazard-affected body and the regional natural disaster losses. The Chinese regional vulnerability to natural disasters was analyzed by using the Aggressive-Cross-Evaluation model. In order to ensure the effectiveness of the algorithm, the small amount of input-output indicators were selected in traditional DEA models. The defects of traditional DEA methods were avoided by the way of Aggressive-Cross-Evaluation. This method evaluated regional vulnerability to natural disasters comprehensively and objectively.%首次提出使用对抗交叉评价(Aggressive-Cross-Evaluation)模型对自然灾害的区域脆弱性程度进行研究.从区域自然灾害的危险性,区域承灾体暴露性和区域自然灾害造成的损失程度构建了区域自然灾害系统的脆弱性评价模型,并通过算例对中国自然灾害的区域脆弱性进行了评价.传统的DEA方法为了保证算法的有效性,一般选取的输入和输出指标的个数较少.对抗交叉评价模型回避了传统的DEA方法的缺陷,继承了DEA方法不需要预先估计权重参数和函数模型的优点,可以全面、客观地评价自然灾害的区域脆弱性.