Identification of the critical factors in flood vulnerability assessment based on an improved DEMATEL method under uncertain environments

With accelerated urbanization, the expansion of urban construction land and the increase in concrete pavement have led to a continuous decline in the natural permeability of the urban surface. This makes it difficult for rainwater in urban areas to infiltrate quickly into the ground, which may lead to urban flooding disasters in cases of heavy rainfall. In recent years, the frequency of urban flooding disasters has been increasing, posing a great safety threat to communities and residents as well as property damage. Therefore, it is important to build a scientific and reasonable community flooding vulnerability assessment system to improve the community emergency response capacity and reduce the losses caused by flooding disasters. In this paper, 94 indicators for community flooding vulnerability assessment are collected by combining literature research and field surveys. The newly proposed fuzzy language expression method based on pentagonal fuzzy numbers (PFN) and the proportional two-tuple linguistic representation (P2TLR) are used as the first step of indicator evaluation and screening, and PFN-P2TLR with the two-step DEMATEL method are combined as the second step of indicator screening. Finally, 23 representative evaluation indicators belonging to the dependent variable are selected. Following a thorough investigation, it was found that “resident characteristics”, “regional economic strength”, “community precipitation status” and “institutional soundness” are the four most influential factors for community resilience to flooding, and these indicators have a significant impact on other indicators. The research results provide a reliable basis and specific guidance suggestions for community flooding vulnerability assessment, which is crucial for further reducing the occurrence of flooding disasters and mitigating the social and economic losses caused by them.