Knowledge Graphs for Multidisciplinary Co-Design of Buildings
Semantic and Geometric Representation in the AEC Industry Using Semantic Web Technologies
Project Description
This PhD project explores how Semantic Web Technologies (SWTs) and knowledge graphs can improve data interoperability, validation, and geometric modeling across Architecture, Engineering, and Construction (AEC) disciplines. The research addresses current limitations in BIM and IFC-based workflows, where data fragmentation and lack of reasoning capabilities hinder integrated co-design processes. The project introduces a co-design workflow that integrates SWTs into collaborative AEC tools, enabling cross-database querying and 3D geometric reasoning. It develops semantic representations using ontologies and validates design data using standards like OWL, SHACL, and SWRL. Furthermore, it investigates efficient methods for modeling and querying geometry in RDF graphs, including linking to native geometry stores. By enabling reasoning across disciplinary models and embedding 3D spatial logic directly into queryable knowledge graphs, the project supports earlier and more informed decision-making during building design.
PhD Topic, Context, and Aim
This research is conducted within the Cluster of Excellence Integrative Computational Design and Construction for Architecture (IntCDC) – specifically Research Project RP20: Knowledge Representation for Multidisciplinary Co-Design of Buildings. The work is carried out at the Chair for Computing in Architecture (ICD/CA) at the University of Stuttgart and involves active collaboration with Buro Happold and Gropyus AG, who contribute industrial use cases and access to interoperable frameworks like BHoM. The aim of the PhD is to explore how Semantic Web Technologies can enhance interoperability, data validation, and the modeling of semantic and geometric information in the multidisciplinary design of buildings. By integrating symbolic reasoning, modular ontologies, and 3D spatial logic into knowledge graphs, the project seeks to overcome existing data silos and support seamless collaboration between architectural, engineering, and construction stakeholders.
Research Questions
-
How can SWTs support AEC design tools with a focus on interoperability, enabling cross-database querying and reasoning?
-
How can SWTs validate building data and support reasoning during design phases, and what challenges arise in real-world applications?
-
What are the most efficient methods for modeling and accessing geometry in RDF graphs, and how can geometric functions be applied in semantic queries?
Methodology
The research follows a Design Science Research (DSR) methodology and is evaluated through industrial case studies. It includes:
-
Developing a semantic co-design workflow that integrates SWTs with AEC design tools. This includes ontology alignment across architectural and engineering domains, federated querying, and bidirectional integration with object-oriented frameworks like BHoM.
-
Implementing inferential reasoning and rule-based validation using OWL, SHACL, and SWRL. This enables validation of design constraints and inference of new knowledge across heterogeneous AEC models, integrated with semantic knowledge graphs and industrial workflows.
-
Designing and evaluating approaches for geometric data modeling and processing in RDF, including the development of over 20 custom SPARQL geometry functions for reasoning over 3D building models directly in RDF graphs. This is complemented by a comparative analysis of RDF-native, extended triplestore, and hybrid (e.g. PostGIS) solutions, examining their expressiveness, processing capabilities, validation, interoperability, and scalability in co-design workflows.
Expected Results and Contributions
This PhD project is expected to result in a knowledge graph-based co-design workflow that effectively integrates SWTs with 3D geometric reasoning for the AEC industry. By embedding semantic and spatial logic into collaborative design processes, the workflow will enable more informed and efficient decision-making across disciplines. One of the key contributions will be the extension of SPARQL with custom geometrical functions that allow querying and analysis of 3D models in RDF-based environments. These functions will enhance the capability of knowledge graphs to reason about design constraints and spatial relationships in co-design contexts. Additionally, the project will provide validated methods for federated design data integration, demonstrated through real-world use cases. These methods will support seamless linking of data from different AEC tools and disciplines, enabling holistic project views and cross-domain reasoning. Overall, the research will advance semantic interoperability in the AEC domain, particularly during early design stages, and will foster better collaboration across architecture, engineering, and construction through machine-readable, queryable, and logic-enforced data representations.
PROJECT TEAM
PhD Student: Diellza Elshani
Institute for Computational Design and Construction, Department for Computing in Architecture (ICD/CA), University of Stuttgart
First Supervisor: Prof. Dr. Thomas Wortmann
Institute for Computational Design and Construction (ICD), University of Stuttgart
Second Supervisor: Prof. Dr. Steffen Staab
Department for Analytic Computing (AC), Institute for Artificial Intelligence, University of Stuttgart & Electronics and Computer Science, University of Southampton
PROJECT FUNDING
Funded by the Cluster of Excellence IntCDC, University of Stuttgart (EXC 2120/1, RP20), and the Zukunft Bau research program (BBSR/BMI), project no. 10.08.18.7-24.25.
RELATED PUBLICATIONS
- Elshani, D., Lombardi, A., Hernández, D., Staab, S., & Wortmann, T. (2025). AEC co-design workflow for cross-domain querying and reasoning using Semantic Web Technologies. Automation in Construction, 176, 106226.
https://doi.org/10.1016/j.autcon.2025.106226 -
Elshani, D., Lombardi, A., Fisher, A., Staab, S., & Wortmann, T. (2022). Knowledge graphs for multidisciplinary co-design: Introducing RDF to BHoM. In Proceedings of the LDAC Workshop on Linked Data in Architecture and Construction at ESWC 2022, Hersonissos, Greece.
https://ceur-ws.org/Vol-3243/paper-03.pdf -
Elshani, D., Lombardi, A., Fisher, A., Staab, S., & Wortmann, T. (2022). Inferential reasoning in co-design using Semantic Web standards alongside BHoM. In Proceedings of Forum Bauinformatik 2022, Technische Universität München, Germany.
https://doi.org/10.13140/RG.2.2.23271.39845 -
Elshani, D., Hernández, D., Lombardi, A., Siriwardena, L., Schwinn, T., Fisher, A., Staab, S., Menges, A., & Wortmann, T. (2023). Building information validation and reasoning using Semantic Web technologies. In M. Turrin, C. Andriotis, & A. Rafiee (Eds.), CAAD Futures 2023: Interconnections – Co-computing Beyond Boundaries (pp. 470–484). Springer.
https://doi.org/10.1007/978-3-031-37189-9_31 -
Elshani, D., Dervishaj, A., Hernández, D., Gudmundsson, K., Staab, S., & Wortmann, T. (2024). An ontology for the reuse and tracking of prefabricated building components. In Knowledge Graphs for Sustainability (KG4S) Workshop at ESWC 2024, Crete, Greece.
Elshani, D., Nahkaee, A., Arrascue, A., Isakovic, H., Staab, S., & Wortmann, T. (2025). Comparative analysis of approaches for geometric data representation in RDF. In LDAC 2025 – Linked Data in Architecture and Construction, Porto, Portugal.