Research on IFC instance data based on topology induction network

ABSTRACTBuilding information modeling (BIM) can store a large amount of civil engineering information and has become a high-value economic and social resource. To improve BIM data interoperability, Industry Foundation Class (IFC) was proposed. However, the IFC instance data usually exceeds megabytes, and the logical relationship of the data is complex and changeable. Therefore, how to further enhance the usability of IFC instance data will become very important. To this end, this paper proposes a data deconstruction model ‘Topological Induction Model (TIM)’, which deconstructs the IFC instance data into a network structure ‘Topological Induction Network (TIN)’. Through the analysis and operation of TIN, more effective utilization of IFC instance data can be realized. Based on TIN, the skeleton structure, citation structure and corresponding acquisition algorithm of IFC instance data are discussed, and the hierarchical characteristic, internal aggregation characteristic and data similarity of data topology are found. Subsequently, the working performance of TIM (TIN) is evaluated and compared with existing data deconstruction strategies. The results show that the overall performance of TIM (TIN) is better than traditional methods. Finally, we show how TIM (TIN) can be used in a graph database to enhance the usability of IFC instance data.KEYWORDS: Building information model (BIM)civil engineeringindustry foundation classes (IFC)graph databasecomplex network Disclosure statementNo potential conflict of interest was reported by the author(s).Data availability statementSome or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.Additional informationFundingThis research was financially supported by the National Natural Science Foundation of China [rant No. 52130807], Changjiang Scholar Program of Chinese Ministry of Education, Science Xplore Prize Award. These supporting agencies are gratefully acknowledged.