Abstract: BIM technology has been widely used in the whole life cycle management of buildings. However, each participant in a construction project usually creates their own BIM model. At the same time, due to the influence of design changes and construction errors and other factors in the construction process, there is a large deviation between the BIM model and the actual building entity. These problems not only destroy the consistency of information, but also seriously hinder the cooperation of project participants. Previous methods of using point cloud data to generate or adjust BIM models can reflect the actual condition of buildings, but they often rely on specific algorithms or tools, which are complex and lack generality. This paper proposes a new solution, combining IFC file and point cloud data, to realize efficient update and display of BIM models. The BIM model attributes and geometric data in IFC file are mapped to the relational database and registered with the collected point cloud data. After calculating the deviation, the data is updated to the database. Finally, the BIM model on the Web side is displayed using the data in the database. This method effectively solves the problem of deviation between BIM model and building entity, and improves the efficiency and accuracy of BIM model updating.