Regional disaster risk assessment of china based on self-organizing map: Clustering, visualization and ranking

Abstract

Regional disaster risk assessment is of practical value for disaster prevention and mitigation in emergency management. It remains an ongoing challenge due to the multiple criteria associated with the decision making problem. In this paper we propose a new multi-criteria decision making method to evaluate the natural disaster risk of China at the region scale. The evaluation index system is composed of 28 indicators reflecting the danger of natural disasters and the vulnerability of the affected body. The method involves a combination of Analytic Hierarchy Process (AHP), Self-Organizing Map (SOM), Isometric Feature Mapping (Isomap), and Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) to achieve the clustering, visualization, and ranking of Chinese regions. Compared with the traditional ranking methods in regional disaster risk assessment, the proposed method is able to recognize the similarity of regions, exploit the embedded clustering structures, visualize the high dimensional regional data, examine the influence of indicators, and rank the regions in terms of the disaster risk. It offers a comprehensive and visual way for policy makers or related organizations to understand the studied data from different perspectives and thus mitigate the disaster risk in the Chinese regions.