Jul 2 – 3, 2025
Ruhr-Universität Bochum
Europe/Berlin timezone

HZB’s Tailored Digital Lab Workflows Towards AI-ready Datasets

Not scheduled
25m
UFO (Ruhr-Universität Bochum)

UFO

Ruhr-Universität Bochum

Querenburger Höhe 283, 44801 Bochum

Speakers

Dr Ana Velazquez (Helmholtz-Zentrum Berlin (HZB)) Carla Terboven (Helmholtz-Zentrum Berlin (HZB))

Description

Digital transformation in experimental Science is driven by efficient data management and workflow integration across all processes, from sample synthesis to specialized measurements, allowing for structured, machine-readable, AI-ready datasets that are aligned with FAIR principles.
At HZB, we focus on driving a paradigm shift by supporting the transition from analog workflows to seamless digital integration using NOMAD Oasis as Data Management platform. Across HZB’s diverse research landscape (e.g. Perovskite Solar Cells, Low Temperature Electrocatalysis, Thermocatalysis, and Thin Films for all these applications), Data Stewards and Scientists closely collaborate on developing tailored, seamless digital workflows that link data across the full sample life cycle (i.e. from synthesis to advanced synchrotron-based characterization) using NOMAD and tailored Jupyter notebooks. Our digital lab workflows integrate heterogeneous data sources (e.g. raw data from devices, manually digitized entries, and pre-processed data) into well-defined data structures, annotated with community-driven vocabularies and ontologies (i.e. voc4Cat and TFSCO). The result is metadata-rich, machine-readable datasets, that are accessible through API, supporting reuse, automation, and AI-readiness.
Here, we present two customized digital workflows developed for our Thin Film-Catalysts laboratories focusing on Thermocatalysis and Electrocatalysis. These workflows manage research data across multiple laboratory processes, linking information from sample synthesis to advanced synchrotron measurements. This systematic approach enables the generation of large, structured, machine-readable datasets that are AI-ready, which in turn, facilitates advanced data analysis, high-throughput analysis, AI-driven insights, and machine learning-based optimization.

Primary authors

Dr Ana Velazquez (Helmholtz-Zentrum Berlin (HZB)) Carla Terboven (Helmholtz-Zentrum Berlin (HZB))

Co-authors

Dr Daniel Amkreutz (Helmholtz-Zentrum Berlin (HZB)) Dr Lukas Thum (Helmholtz-Zentrum Berlin (HZB)) Dr Maddalena Zoli (Helmholtz-Zentrum Berlin (HZB)) Dr Marcel Risch (Helmholtz-Zentrum Berlin (HZB)) Dr Michael Götte (Helmholtz-Zentrum Berlin (HZB))

Presentation materials

There are no materials yet.