A Lightweight Approach to Railway Infrastructure BIM Models Considering Geometric Detail Features

Objectives: Railway infrastructure is diverse in type, large in data volume, and complex in structure. Lightweighting its Building Information Modeling (BIM) models is crucial for efficient data processing. However, existing lightweighting algorithms face issues such as poor retention of geometric features and significant loss of detailed information. Methods: This paper proposes a lightweighting method for railway infrastructure BIM models that considers geometric detail features. The method first focuses on the geometric detail features and lightweighting needs of track, roadbed, tunnel, and bridge BIM models, identifying the challenges faced by lightweighting different models. Secondly, based on the geometric and detail features of different types of railway infrastructure, it establishes rules such as boundary preservation, angle control, and neighborhood area. The boundary preservation factor helps retain the contour features of the model to a greater extent and reduces the occurrence of hollow parts in the model. The angle error control factor prevents the loss of triangular faces caused by large rotation angles during edge folding. The neighborhood triangular average area factor helps retain more local detail features of the model. Finally, it designs simplification methods for various types of railway infrastructure based on different model needs and analyzes the method's effectiveness using five evaluation criteria. Results: Comparative analysis of the three algorithms shows that the method proposed in this paper simplifies the railway infrastructure model more effectively. The method retains more geometric detail features, effectively reduces the geometric error of the lightweight model, and provides high-quality visual effects. Furthermore, the test results of the simplification factors demonstrate that the boundary preservation, angular error control, and average area of neighboring triangles introduced in this paper effectively reduce model voids and maximize the retention of model contours, surface features, and local details. Conclusions: The methodology proposed in this paper effectively simplifies the model and is well-suited for lightweighting railway infrastructure models.