Abstract: Rapid identification and driving factor analysis of urban waterlogging risk have been the primary requirements to implement urban waterlogging management.However, the traditional urban drainage modelling method requires high-resolution basic data support and a large modeling cost, which is difficult to meet the demand of urban waterlogging rapid identification.Based on the actual urban waterlogging disaster data of building communities in Luzhou City from 2015 to 2022,the spatial distribution of waterlogging risk in building communities was rapidly identified using kernel density estimation and spatial correlation analysis.Spearman correlation analysis and geodetector approach were used to investigate the waterlogging risk driving factors.The results show that the waterlogging risk of building communities in Luzhou City tends to decrease gradually from the center to the perimeter, and the high-risk areas are mainly located in the Chengbei region, Zhongxinbandao region and Longmatan region.The primary driving factors of the waterlogging risk are soil texture, land use, social factors, and rainfall factors, and they exhibit complex forms of multifactorial synergies.The results can provide a basis for high-resolution modeling of the waterlogging risk in Luzhou City, and the methods can also provide methodological support for a rapid identification of the waterlogging risk and preliminary analysis on factors contributing to waterlogging in urban building communities in hilly cities of southwest China.