Abstract: The evaluation targets of geological hazard susceptibility and risk assessment are the same but the evaluation contents including time, space and intensity information of geological disasters are different. This paper combines the statistical model in the susceptibility with the physical model in the risk assessment on collapse landslides, and combines the advantages of statistical model predicting spatial location information objectively with the advantages of physical model simulation including geological disaster occurrence mechanisms. It makes up for the insufficiency of regional statistical model predicting the disaster intensity information. It also effectively controls or corrects the spatial position of the physical model simulation. Then a comprehensive analysis is completed on the susceptibility and risk level of the collapse and landslide, which achieves a refined assessment of the potential high-risk location of the regional collapse and landslide. This paper takes the Longshan community in Fuding County, Fujian Province as an example. It uses data from high-definition images, terrain, boreholes and geological disasters acquired in the field. With the combination of statistical model and physical model, the refined evaluation of potential high-risk locations is completed. The research results show that:the area that needs to be focused on governance is about 26.92% of the total area of the mountain near the community. There are five areas in the community that require centralized investigation and governance. Three of them need to be focused on key governance. The potential high-risk areas coincide with the hidden danger points in the investigation area. The five high-risk areas directly threaten the safety of 180 buildings(about 360 householders). Refining the large-scale high-risk areas is delineated in field research. The feasibility of the evaluation method system is verified. The evaluation method system provides working ideas and guidance for the refined investigation and governance of regional collapse and landslide disasters.