Research on Partial Model Extraction of Railway Infrastructure Based on the Industry Foundation Classes Files

Abstract

To effectively cope with the ever-growing intricacy of railway Industry Foundation Classes (IFC) file sizes, the adoption of a partial model approach has consistently been recommended as a highly effective strategy. This approach facilitates the efficient handling of increasing data volumes, enhancing the management capabilities of IFC files for purposes such as data storage, exchange, and transmission. This paper proposes an algorithm that adopts an iterative extraction methodology, grounded on the hierarchical IFC model’s tree structure to produce the desired partial models. Its primary objective is to minimize the size of large IFC files by selectively extracting the indispensable models. By leveraging solely the data structure of the IFC files, this approach circumvents the necessity for file format conversion or reliance on Model View Definitions (MVD). During the extraction of the required models, two types of related attributes are simultaneously extracted, and the extraction of relationship entity properties is optimized to enhance the extraction efficiency. To assess the effectiveness of our algorithmic approach, we conducted an in-depth case study centered around a contemporary high-speed railway project located in the southwestern region of China. The extracted models’ integrity was verified using BIMvision, while semantic syntax verification was performed using Express Engine and IfcObjectCounter. The results demonstrate that our algorithm not only accurately extracts the desired model segments from the IFC file, but also significantly improves model loading efficiency, minimizes memory usage, achieves model miniaturization, and shows promising performance and application prospects.

Keywords

• Industry foundation classes (IFC)
• partial model
• extraction
• building information modeling (BIM)
• intelligent railway