Assessment of geo-disaster risk levels induced by extreme rainfall using integrated FCM-VIKOR approach
Xiang Liu, Hai‐Min Lyu, Shui‐Long Shen
Abstract
This study introduces a novel method for assessing the risk of flooding and associated geo-disasters triggered by extreme rainfall events. The method incorporates clustering algorithms into the analytic hierarchy process (AHP) and VIKOR methodology. Risk assessment is enabled by VIKOR, which calculates the Q value for each grid within a geographic information system (GIS) framework. AHP is employed to assign weights to each influencing factor in the VIKOR model. The utilised clustering algorithms include fuzzy c-means (FCM), density-based spatial clustering of applications with noise (DBSCAN) and K-means, which categorise the risk factors of 2300 grids into four distinct clusters. The Q values derived from the results of the three clustering algorithms display a gradient distribution, with the FCM algorithm showing the most defined gradient boundaries. Consequently, risk levels are determined based on the Q values obtained from FCM. The methodology was applied in Meizhou, Guangdong, using GIS to map the risk levels of flooding disasters induced by heavy rainfall. Field observations confirm the applicability of this approach, highlighting its robust predictive and emergency response capabilities.