A deployed data-cataloging and sharing platform built for an EU Horizon 2020 project on microwave processing of composites in the aerospace and automotive sectors (SIMUTOOL, 2015–2019). It let eight companies with incompatible systems, proprietary simulation tools, and differing confidentiality and legal constraints describe, discover, and exchange data assets across organizational boundaries - without integrating their underlying systems.

This is the inter-organizational data-sharing problem now addressed by initiatives like Catena-X and the EU Data Act, tackled here hands-on, in the field, years before those existed.

System-design deep-dive: see ARCHITECTURE.md.

Research context: dissertation · CORDIS · project page, Uni Bamberg

Eight geographically dispersed partners, each with distinct specializations and proprietary systems, many of whom had never worked together. The hard part was never the variety of data formats - it was the dense web of data-driven dependencies between activities, especially the simulations: process control depended on process simulation of the part (and vice versa), and was further shaped by oven simulations predicting microwave behavior. Coordinating these hand-offs, across confidentiality boundaries, was the core challenge.

A platform centered on describing and exchanging data assets rather than moving or integrating systems:

• Catalog & metadata service - a semantic metadata layer (controlled vocabularies over a property-graph store) managing descriptions, discovery, and access for heterogeneous assets.
• Blob store - heterogeneous files kept in their original formats.
• End-user clients - an upload client for adding assets and metadata, and a client for visualizing and comparing manufacturing sensor data.

The system was deployed and used across the consortium for the duration of the project. The repositories in this organization are an archived snapshot of the project's prototype implementation, preserved for reference.

1. Accelerating R&D Processes: Shortening turnaround time between R&D, engineering, and production by simplifying the sharing of data across orgranizations.

2. Enhancing Data Discovery: Easily find data assets within a shared catalog.

3. Streamlining Data Management: Eliminating inefficiencies associated with requesting participants to re-upload or re-share data, thus saving valuable time.

4. Promoted Data Asset Reusability: Leveraging controlled vocabulary and semantic technologies, thereby extending their lifespan beyond their original uploader's involvement.

The software-solutions team at the University of Bamberg (Chair of Mobile Systems, Prof. Daniela Nicklas). I was the project's architect and lead developer and the one constant member across its full duration - owning the architecture, building hands-on, and coordinating the data-sharing design across partners.

• Project lead / architect / developer: Nasr Kasrin
• Research associates: Maliha Qureshi, Simon Steuer
• Developers: Adrian Lengenfelder, Valentina Veloso, Lukas Genssler, Katharina Broswik, Harshit Gupta
• Supervisor: Prof. Daniela Nicklas

Circular identifiers in the diagram denote the external-facing systems of participating companies. A fuller list is in the dissertation (p. 66).