A linked Data approach for information integration between BIM and sensor data

Recent years, the trend to connect data from various sources becomes more and more obvious. The Semantic Web is developed to facilitate information linking, sharing and reusing across application, enterprise, and community boundaries. Linked Data, which lies at the heart of the Semantic Web, is able to make the Semantic Web a reality through creating interrelated data. In the building domain, Building Information Modelling (BIM) is popularly used to integrate building data among different life phases and different sub-disciplines. BIM is recognized as having a good effect in containing various building related information. However, the current integration between BIM and sensor data has some problems, like data interoperability troubles between data input and output. Since there already exist experiences in processing sensor data with the Semantic Web technologies, and the integrated result “Semantic Sensor Web” is argued for bring the usefulness of the sensor data to its full potential, this integration approach could also be implied in BIM and sensor data integration.

In this thesis, a Linked Data approach is developed to achieve information integration between BIM and sensor data for the facility managers’ building performance analysing needs. The thesis is constructed of six chapters to clarify and verify this integration method. Chapter 1 discusses the research design background, the research questions and the research process. Chapter 2 is the glossary of this article. Chapter 3 holds the literature review of the current researches on BIM, sensor data, the Semantic Web and their integrations. Chapter 4 describes the methodology to integrate the BIM and sensor data by using Linked Data. Chapter 5 testifies this methodology through a case study from TU/e’s Vertigo building. Last, chapter 6 gives the reflections of this integration method and provides some future research recommendations.

The methodology of this BIM and sensor data integration process is divided into six parts. The main flow path of the integration is collecting data, analysing data, transforming data, linking data, querying data and visualizing data. These technical processes are examined through the case study, and several query result visualizations are presented to help the facility managers analysing the building environment performances.

In the case study, there are four different data sources. Through the Linked Data approach, these four data sources with different data processing systems are integrated into one RDF model, and are convenient to be synthetically processed and queried without format or domain boundaries. It is testified that the Linked Data approach developed in this thesis, is effective to achieve the information integration between sensor data and BIM model, and could help the facility managers analysing the building operation performances through comprehensive information backup from different disciplines and data formats.