Abstract: During the assembly of prefabricated metro components, poor installation quality and accuracy problems often arise due to heavy quality inspection works and large errors. Thus, building information modeling(BIM) and threedimensional (3D) laser scanning technology are adopted to mitigate these problems. Using BIM and 3D point cloud modeling for prefabricated components, virtual preassembly is performed to check the processing accuracy to avoid unnecessary waste due to component quality problems in the onsite installation process. Then, the gray prediction method is used to construct the gray model GM (1, 1) to predict the cumulative error of the assembly of multiple prefabricated components in the prefabricated metro station and employed to compare and verify the virtual preassembly inspection results. The findings of this study are as follows. (1) The processing errors of seven components in the quality inspection are all within the specification range. (2) The cumulative assembly error, assembly error between blocks C1 and D, and the gap between the longitudinal seams of other adjacent blocks of the entire ring are13.920, 4.150, and <3 mm. (3) The virtual preassembly error test results fit well with the gray prediction results of the GM (1, 1) model, and the maximum error is only 0.407mm.

Key words: prefabricated metro station, prefabricated components, building information modeling + threedimensional laser scanning, quality inspection, gray prediction, installation accuracy