Construction projects are challenging because of significant investment from various investors. The completion of construction work also requires the collaboration among experts from several disciplines. However, previous works on the integration of information available in the design, construction, and maintenance phases are insufficient. In this study, we initially reviewed the previous studies in this area and then discussed the possible challenges and demands. Afterward, we presented a thorough comparison of different models (i.e., Tekla model as architectural model and Midas Civil model as structural analysis model) and software applications. Then, we proposed a novel approach in which a unified information model and a region-based feature point matching algorithm (RFPMA) were employed to enhance the integration of information available in different phases of a construction project. Furthermore it has been achieved to dynamically integrate the monitoring data and analysis results into the unified information model. Finally, we proposed the method and the overall workflow that show how the available information in different phases is integrated. The proposed approach was implemented in a Client Server (C/S) platform and applied in a large real project located at Jinan, Shandong Province, China, which is called Jinan Yellow River Bridge. The application involved several scenarios, including information sharing between Tekla model and Midas Civil model, synchronous model updating and data visualization, and collaborative analysis of the safety level during the construction process. This research has made a breakthrough in the theory, technology, and application of model integration and has established the foundation for safety management.
引用方式: Zhang, X., Lin, J.*, Hu, Z., Zhang, J., Fang, J. and Du, S. (2016). "Towards BIM-based model integration and safety analysis for bridge construction" in Proceedings of the 16th International Conference on Computing in Civil and Building Engineering. 1091-1097. Osaka, Japan.