Abstract: The traditional inspection method of steel structure bridge manufacturing quality is time-consuming and inefficient. In order to improve the inspection efficiency, a method based on point cloud data is proposed. 3D laser scanner was used to obtain the point cloud data of the steel beam and preprocess it. The improved region growing algorithm was used to segment each face of the steel beam, and the detection location was found based on the properties of the Hash grid. The singular value decomposition algorithm was used to fit the plane, and the plane equation was obtained. The three manufacturing quality parameters of beam height, flatness and perpendicularity were calculated by using the spatial geometric relationship between the planes. Based on the pluggable development mode of the open-source point cloud processing software CloudCompare, the development of the plug-in for automatic calculation of manufacturing quality parameters was completed using C++ programming language, and the I-shaped steel beam of Yangshan Bridge was tested and verified. The experimental results show that all the parameters meet the fabrication specifications of I-shaped steel beams, which proves the feasibility of this method.