Research on Semi-Automated Creation of BIM Models for Existing Buildings Based on Bi-Directional Information

Building Information Modelling (BIM) is widely used in construction engineering because of its ability to integrate geometric, spatial, and operational information. However, BIM reconstruction for existing buildings still faces limitations in practical implementation, especially low modelling efficiency and insufficient model usability during operation and maintenance. These problems are also relevant to laboratories and indoor facilities used for electromagnetic testing, antenna measurement, and propagation experiments, where accurate spatial information is required. This paper proposes a semi-automated BIM modelling method based on Dynamo programs and manually labelled point clouds. The method integrates point cloud data with drawing information to extract complete component details and generate BIM models of existing buildings through specialized modules. Case studies show that the proposed approach improves both modelling efficiency and information accuracy compared with conventional manual modelling. The resulting BIM models can support facility operation, maintenance, and spatial management, providing practical support for the digital reconstruction of existing buildings and engineering spaces related to electromagnetic wave and antenna applications.