SIGDIUS Seminar June

The Special Interest Group Data Infrastructure provides a forum for interested working groups wishing to establish or further develop an RDM infrastructure at working group or institute level. We invite you to a monthly SIGDIUS seminar, to which we invite internal and external experts for presentations and discussions. SIGDIUS members will have the opportunity to exchange their experiences with concrete RDM infrastructures.

We cordially invite all interested parties to our next meeting on 12 June 2024 at 2:00 p.m. For participation, please send an e-mail to Juergen.Pleiss@itb.uni-stuttgart.de.

This seminar will be held as an online seminar with talks from:

Roger Winkler
Darmstadt University

RDM-workflows for construction engineering and architecture

Access to structured research data on the built environment is of fundamental importance to the preservation of knowledge and the development of new methods and technologies in the fields of design and construction. The architectural research community, however, faces significant challenges in efficiently collecting, processing, and FAIR publishing of heterogeneous data collected from diverse sources and media. The NFDI4Ing Seed Fund project “RDM-Workflows for Construction Engineering and Architecture” has explored the challenges in architectural research data management based on a case study carried out with the Institute for Computational Design and Construction at the University of Stuttgart. As a result, the project team developed a “CSV-RDF-Mapper”, an MVP for utilising Application Profiles based on SHACL in order to semantically lift tabular data into FAIR knowledge graphs. The talk will present the background of and work done within the project and aims for the discussion of ideas for further development and applications beyond the architectural domain.

Roger Winkler works in the Department for Information Technology, Research and Development at the University and State Library Darmstadt. He received his Master’s degree in Architecture from the Technical University of Darmstadt and focused on computational and data-driven design technologies. Within the DFG-Project FID BAUdigital, he has been involved in the coordination and development of the workbench since 2020, a service to publish domain-specific research data for civil engineering, architecture and urbanism. The workbench was integrated into the NFDI4Ing as Data Ingest Service in 2024. In this context, he also worked on the project “RDM-Workflows for Construction Engineering and Architecture” together with the Institute for Computational Design and Construction at the University of Stuttgart. He is currently involved in the NFDI4Ing Measure S3 “Metadata & Terminology Services”, focusing on the development and utilisation of SHACL-based application profiles with the NFDI4Ing Metadata Profile Service.

Daniel Appel,
University of Stuttgart, IAG

FAIR-FLEXI – a trustworthy CFD code for simulation and training

Engineering and science disciplines produce an enormous amount of research data every day. However, these data only turn into knowledge and value if they can be readily shared and explored using data analytics and artificial intelligence methods. To this end, a FAIR (findable, accessible, interoperable, reusable) data infrastructure is indispensable. While reproducibility is conceptually supported by open-source software, applying the FAIR principles to complex scientific simulation frameworks extends far beyond publicly providing their source code. This is because such frameworks usually come along with a lot of implicit knowledge on e.g. user-defined parameters, compile options, hardware requirements and third-party dependencies. In light of these uncertainties, impairing the reproducibility and thereby trustworthiness of simulation results, we present FAIR-FLEXI as an open-source CFD framework that implements the FAIR principles along the entire simulation chain, from the first step in a preprocessing tool to the visualisation of results. The framework traces the complete source code, automatically builds configuration and simulation setup for each computation, and generates scripts to readily recreate the simulations. The talk outlines the preliminary work done by coworkers in this regard, but also discusses ideas for extending the usability of the framework and introducing learners more effectively.

Daniel Appel is a PhD student in the Numerics Research Group headed by Prof. Andrea Beck at the Institute of Aerodynamics and Gas Dynamics at the University of Stuttgart. He received his Master's degree in mechanical engineering from the Technical University of Munich with a major in energy technology and engineering fundamentals. His research focuses on high-order numerical methods for compressible multiphase flows, in particular node-level performance optimisations, dynamic load balancing and other HPC-related aspects. He is currently working on the reproducibility of scientific data according to the FAIR principles.