IFC-Based Partial Data Model Retrieval for Distributed Collaborative Design

Abstract

To deal with the growing complexity of Industry Foundation Classes (IFC) models, a partial model has always been proposed as an effective way to satisfy users’ specific requirements by tailoring the central model. The traditional approaches with IFC-validated partial models have the problems of high structural complexity and low data density. This paper proposes an IFC property–based model generation method to support domain-specific data extraction with a property-based model view definition (MVD). This approach removes data-irrelevant IFC definitions while keeping data links between the view and the IFC central model. Based on the MVD, an algorithm is designed to extract the required properties from the central IFC model by recursively iterating through IFC data instances for two types of properties. The proposed approach has been implemented on the top of the open source BIMserver platform. Its feasibility and efficiency have been validated through a set of practical examples. Compared to central IFC model and IFC-validated partial model approach, the experiment results show that our proposed approach achieves the smallest model size, the fastest speed for model loading, the least memory usage, and the lowest network transmission overhead.