LDAC2024 12th Linked Data in Architecture and Construction Workshop PieterPauwels Eindhoven University of Technology
Netherlands
MaríaPoveda-Villalón Universidad Politécnica de Madrid
Spain
WalterTerkaj Consiglio Nazionale delle Ricerche
Italy
June 13-14 2024 Bochum Germany
LDAC2024 12th Linked Data in Architecture and Construction Workshop 1613-0073 73F618DCC253CBF71ADC5E0C5904F59D GROBID - A machine learning software for extracting information from scholarly documents
Silver sponsors:

Organisational support:

Preface

The LDAC workshop series provides a focused overview on technical and applied research regarding the usage of semantic web, linked data and web of data technologies for architecture and construction (design, engineering, construction, operation, etc.). The workshop aims at gathering researchers, industry stakeholders, and standardization bodies of the broader Linked Building Data (LBD) community. The aim of the workshop is to present current developments in research and development, coordinate efforts, gather stakeholders, and extend industry uptake and collaboration.

We are pleased to collect in this volume the papers that were submitted and presented during the 12th Linked Data in Architecture and Construction (LDAC) Workshop. The workshop took place on 13 and 14 June 2024, as a stand-alone event in Bochum, Germany. The workshop was preceded by an extensive 3-day summer school. These proceedings present the eleven full papers and two short papers that were accepted after the peer-reviews by the members in the program committee.

The workshop also included three excellent keynotes on three diverse topics, one of them hosted during the SSoLDAC Summer School. The first keynote, by Ekaterina Petrova (TU Eindhoven) was titled "Symbolic, Neural or Neuro-Symbolic Artificial Intelligence for Architecture, Engineering and Construction?".

Embedding intelligence in Artificial Intelligence (AI) systems requires enabling both the analysis of vast amounts of raw data and the use of background knowledge to perform higher-level functions such as abstraction, analogy, reasoning, and planning. While sub-symbolic (i.e., neural) AI approaches such as Artificial Neural Networks are powerful when it comes to processing data and discovering patterns in it, they are usually data-hungry and lack explicit representations of background knowledge. Symbolic structures, on the other hand, represent this background knowledge explicitly (e.g., semantic graphs). Neuro-symbolic AI blends the processing and approximation capabilities of neural methods with the knowledge representation and reasoning abilities of symbolic approaches to improve the overall performance of AI systems and address challenges at both algorithm and application levels. This talk will discuss the untapped potential of neuro-symbolic AI in Architecture, Engineering and Construction and highlight recent implementations and ongoing research aimed at improving the performance of the built environment, focusing particularly on smart building applications.

The second keynote was given by Raúl García-Castro (Universidad Politécnica de Madrid), with the title "Lessons learnt from researching on semantic interoperability".

Seamless interoperability is still a challenge in the current landscape of information systems, which need to be aware of cross-sectorial information coming not only from other information systems but also from people or diverse entities in the world. This talk presents some lessons learnt from researching and implementing semantic interoperability. On the one hand, it will cover the semantic aspect of interoperability, giving guidelines on the development of ontologies and on their role for sharing consensus. On the other hand, it will deal with the interoperability perspective, giving hints on engineering semantically interoperable systems and presenting challenges in decentralised interoperability, such as the privacy one.

The third and last keynote was presented by Martin Voigt (elevait GmbH), with the title "Overcoming boundaries: How Linked Data and Machine Learning are Transforming Enterprise Software".

In a world driven by data, the ability to effectively structure, process and utilize information is more crucial than ever. However, traditional enterprise software and its technological paradigms often reach their limits. In this keynote, we outline how the elevait suite pushes the boundaries of traditional data processing by applying linked data principles and machine learning. Our platform is domain and use case agnostic, but is widely used in the construction industry to automate business processes and increase efficiency. Learn how we use advanced workflows to semantically annotate data, train ML models and generate data according to linked data principles and use it in process automation. We also show how no-code tools are used to define business rules and how automatically generated user interfaces make it easier to interact with this data. Using concrete examples from our platform, we will demonstrate how these technologies are implemented in real-life scenarios to create precise, scalable and efficient solutions. We will also share valuable lessons learned that can help you drive your development and research projects forward.

In the event, we also had a full industry marketplace, as well as a number of posters. These posters have not been included in this Proceedings volume, yet are available via the LDAC workshop website1 . The list of posters presented in the event is:

• Lasitha Chamari,
Shalika Walker, Ekaterina Petrova, Pieter Pauwels. Linked Data Service for Improving Portability of a Model Predictive Controller. • Sebastian Blechmann, Hannah Görigk, Rita Streblow and Dirk Müller. Lu Wan, Ferdinand Rossa, Torsten Welfonder, Ekaterina Petrova and Pieter Pauwels. A web application for automated building energy analyses and monitoring using semantic web technologies. • Sergio Acero González, James Allan, Hashem Birahjakli, Emanuele Laurenzi, Wolfram Willuhn, Edrisi Munoz and Sascha Stoller. The development of an application layer connected to a knowledge graph for the continuous calculation of energy performance indicators. • Zehor Hounas, Maxime Lefrançois, Antoine Zimmermann and Bruno Traverson. A hybrid approach of semantic modeling and co-simulation for a better consideration of the physics of physical phenomena in a smart building. • Jonas Maibaum and Philipp Hagedorn. Knowledge modeling and provision for circular economy aspects of municipal infrastructure projects. • Daniel Napps. Knowledge Graphs for Building Design Decisions. • Dimitris Mavrokapnidis, University College London, UK • Claudio Mirarchi, Politecnico di Milano, Italy • Jyrki Oraskari, RWTH Aachen, Germany • Nicolas Pauen, RWTH Aachen, Germany • Pieter Pauwels, Technical University Eindhoven, The Netherlands • Ekaterina Petrova, Technical University Eindhoven, The Netherlands • María Poveda-Villalón, Universidad Politécnica de Madrid, Spain • Dimitrios Rovas, University College London, United Kingdom • Ana-Maria Roxin, University of Burgundy, France • Oliver Schulz, RWTH Aachen, Germany • Madhumitha Senthilvel, RWTH Aachen, Germany • Alvaro Sicilia, La Salle Barcelona, Spain • Ranjith Soman, Imperial College London, UK • Daniele Spoladore, Consiglio Nazionale delle Ricerche, Italy • Walter Terkaj, Consiglio Nazionale delle Ricerche, Italy • Edlira Vakaj, Birmingham City University, UK • Jeroen Werbrouck, Ghent University, Belgium • Sven Zentgraf, Ruhr-Universität Bochum, GermanyLocal Organising Committee• Philipp Hagedorn, Ruhr-Universität Bochum, Germany• Sonja Pohlmann, Ruhr-Universität Bochum, Germany• Sven Zentgraf, Ruhr-Universität Bochum, Germany• Markus König, Ruhr-Universität Bochum, Germany• Marlena Block, Ruhr-Universität Bochum, GermanyLDAC Committee• Philipp Hagedorn, Ruhr-Universität Bochum, Germany• Pieter Pauwels, Eindhoven University of Technology, Netherlands• María Poveda-Villalón, Technical University of Madrid, Spain• Walter Terkaj, Consiglio Nazionale delle Ricerche, Italy• Alex Donkers, Eindhoven University of Technology, Netherlands • Madhumitha Senthilvel, RWTH Aachen, Germany based approach for fault detection and diagnosis and fault location assessment Ontology-in air handling units. • Jeroen Werbrouck, Ghent University, Belgium• Pille-Riin Peet, Ergo Pikas and Aime Ruus. Semantic Web for StreamliningBuilding Design and Permitting Processes.• Julia Kaltenegger, Ekaterina Petrova, André Borrmann and Pieter Pauwels. Aconceptual system architecture for enriching Digital Twins with materialperformance data using symbolic and sub-symbolic Artificial Intelligence.
•
Table of contents Regular PapersStandard-Oriented Ontology Export of Domain Catalogues from Data DictionariesSebastian Schilling, Christian Clemen10 -22Semantic Interoperability using Ontologies and Standards for Building ProductPropertiesHe Tan, Rahel Kebede, Annika Moscati, Peter Johansson23 -35Advanced Process Representation for Semi-Automated Linking betweenConstruction Schedules and IFC filesJonas Schlenger, André Borrmann36 -49Web of Simulation Ontology (WoSO): Integration of Building PerformanceSimulations in IoT SystemsZehor Hounas, Maxime Lefrançois, Antoine Zimmermann, Bruno Traverson50 -62infraspatialOT: An Ontology for the Representation of Spatial Relationships in RoadInfrastructureIna Heise, André Borrmann63 -76A Standard-Based Ontology Network for Information Requirements in DigitalConstruction ProjectsMartina Mellenthin Filardo, Liu Liu, Philipp Hagedorn, Sven Zentgraf, JürgenMelzner, Markus König77 -90Relative Location Ontology: An Ontological Model for Representing DirectionalTopological Relationships between Spatial Entities in Oriented SpaceAnne Göbels, Jakob Beetz91 -104Aligning openCDE APIs with Linked Building Data through Constrained Containersin Common Data EnvironmentsOliver Schulz, Jakob Beetz105 -117
https://linkedbuildingdata.net/ldac2024/
Acknowledgments

This event is only possible by dedicated and committed support from several organisations and people. In particular, we thank the authors and presenters for their high-quality contributions, the Program Committee who reviewed the papers presented in this volume, the keynote speakers for their excellent contributions, and the local organisers, in particular Philipp Hagedorn for all organizational efforts. Special thanks go to the sponsors of this workshop, with Pallas and Tennet as diamond sponsors, elevait, Neanex, and Building Digital Twin Association (BDTA) as gold sponsors, and bSDD, BimConnected, Schüßler-Plan digital, XRtech, NMBU, and QAECY as silver sponsor.

Program Committee

• Calin Boje, Luxembourg Institute of Science and Technology, Luxembourg

Using ICDD for BIM and GIS Integration in Infrastructure Judith Krischler, Paul-Christian Schuler Jakob Taraben Defining Semantics for Digital Twins of Façade Component Testing Facilities CalinBoje NicoMack SylvainKubicki AntoineDugué Pascale Brassier an Extension of the WoT Thing Description Ontology for Digital Twins in the Construction Domain Salvador González-Gerpe Wotdt Socorro Bernardos, María Poveda-Villalón, Raúl García-Castro AndreaCimmino Short Papers How Much OWL Do You Need to Know to Make Sense of Building Ontologies García-Castro María Poveda-Villalón SergioCarulli-Pérez Raúl A Method to Unify Custom Properties in IFC to Linked Building Data Conversion Jyrki Oraskari LukasKirner MaritZöcklein Sigrid Brell-Cokcan