Flood vulnerability assessment using the triangular fuzzy number-based analytic hierarchy process and support vector machine model for the Belt and Road region

Flood is one of the most serious natural disasters in the world. Flood losses in the developing countries throughout the Belt and Road region are more than twice the global average. However, to date, the extent of the vulnerability of the Belt and Road region remains poorly understood. Therefore, this study sought to address this knowledge gap. In this study, we presented a vulnerability assessment model based on triangular fuzzy number-based analytic hierarchy process (TFN-AHP) and support vector machine (SVM) model. Firstly, a geospatial database including 11 flood conditioning factors was built. Secondly, the exposure and disaster reduction capability were calculated based on TFN-AHP and SVM, respectively. Finally, the spatial distribution of vulnerability throughout the Belt and Road region was generated. According to the results, the exposure and disaster reduction capability in most areas are extremely low, accounting for 86.45% and 80.53%, respectively. Meanwhile, the vulnerability of 47,105,300 km2 areas is low or extremely low, accounting for 93% of the Belt and Road region. The high-vulnerable areas (accounting for 3.54%) are primarily concentrated in the southern and eastern parts of China, northern India, most areas of Bangladesh, the Indus Valley in Pakistan, the Nile River Basin in Egypt, and the central region of Indonesia. Obviously, these regions with high vulnerability are characterized by frequent economic activities and dense populations. As suggested of these results, this study provides scientific and technological evidence for the prevention and mitigation of flood disasters in the countries along the Belt and Road region.