A Method for Risk Assessment to Deep-Seated Catastrophic Landslides caused by Heavy Rain based on Artificial Intelligence and Mathematical Statistics

摘要

Sediment disasters caused by deep-seated catastrophic landslides havebecome a serious problem in Japan. Thus, it is necessary to identify theslopes with higher risk to deep-seated catastrophic landslides in order toestablish disaster prevention planning. In this study, a method based onboth self-organizing map and Hayashi's second method ofquantification is proposed. Slopes with higher risk to deep-seatedcatastrophic landslides are then identified by applying the proposedmethod to 142 slopes in Nara Prefecture. Furthermore, the slopes areprioritized according to the sample scores of Hayashi's second methodof quantification.