Graphbasierte Algorithmen und gesamtheitliche Repräsentation von Systemen der TGA mit BIM und Linked Data

• Graph based algorithms and holistic representation of technical systems with BIM and Linked Data

Pauen, Nicolas Patrick Maria; van Treeck, Christoph Alban (Thesis advisor); Monsberger, Michael (Thesis advisor)

Aachen : RWTH Aachen University (2022)
Dissertation / PhD Thesis

Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2022

Abstract

Technical systems in the Architecture, Engineering, Construction and Operations (AECO) industry are complex interconnected structures that are interdependent with each other and with spatial entities. Over the life cycle of technical systems, different stakeholders with various information needs and software applications work together in temporary organizations. To cope with this high complexity, large amounts of data have to be processed and exchanged. Building Information Modeling (BIM) makes it possible to represent this data as structured, digital information in machine-readable representations. However, not all hierarchical, structural, and functional aspects of technical systems can be represented in current data models and linked with information on further domains. Therefore, Semantic Web Technologies (SWT) offer the possibility to link information web-based on data level and to define a common terminology for the description of resources in the form of knowledge representations with the help of ontologies. Against this background, the current state of the art for the representation of technical systems in the AECO industry, based on SWT, is analyzed. Furthermore, shortcomings with respect to the holistic representation according to general systems theory are identified. Building on this, TSO is explained as a knowledge representation for an integrated explicit representation of linked technical systems, their hierarchical subdivision, structural and functional interconnections, and relationships to spatial entities. As such, TSO supports the effort to represent linkable information in a future semantic web of building data. In addition, a method and its modular implementation for informational enrichment, topological complexity reduction, and transfer of information from Industry Foundation Classes (IFC) models to TSO are presented. The modular design enables the usage of the individual process steps independently and, thus, also for review processes and quality assurance of technical systems in BIM models. For validation purposes, a fictitious and a real project are used to show that the information required for an holistic representation can be enriched and transferred from IFC models to TSO as well as represented in TSO. With respect to the representation, use cases from the planning and operation phases are considered in detail. Finally, the findings are critically discussed and corresponding limitations as well as the need for further research are pointed out.