Data Fabric is the podcast from cts Group Industrial Informatics. We talk to the people who connect shop floors to the cloud — system integrators, consultants, researchers and plant engineers who deal with OT/IT convergence, industrial data platforms and digital twins every day. This podcast is about having conversations about what it actually takes to turn scattered production data into a reliable foundation for real-world use cases. New episodes drop regularly.
Site
RSS
Apple
Classements récents
Dernières positions dans les classements Apple Podcasts et Spotify.
Liens partagés entre épisodes et podcasts
Liens présents dans les descriptions d'épisodes et autres podcasts les utilisant également.
Découvrez des podcasts liées à Data Fabric. Explorez des podcasts avec des thèmes, sujets, et formats similaires. Ces similarités sont calculées grâce à des données tangibles, pas d'extrapolations !
Il n'y a pas de contenu associé à ce podcast.
Aucun lien partagé par ce podcast
Qualité du flux RSS
À améliorer
Score global : 53%
Digital Twins for Direct Air Capture — How TU Vienna Built a Live Research Platform with cts DataFabric
mercredi 2 septembre 2026 • Durée 52:51
In this episode, Clemens talks with Stefan Jankovic, postdoctoral researcher at TU Wien, about a 2.5-year collaboration that produced a fully operational digital twin for a direct air capture pilot plant. Stefan explains how his research group combines experimental work, measurement technology, and process simulation to better understand and optimize carbon capture — and why centralized data management turned out to be the critical foundation for everything else.
The conversation covers how the team previously built a digital twin on Microsoft Azure, what worked and what didn't, and why a colleague's recommendation of inmation changed their approach entirely. Stefan walks through the concrete architecture: multiple vendor PLCs, external measurement devices with different protocols (OPC UA, MQTT, Modbus), and how DataFabric serves as the single point of truth that makes dashboards, automated reporting, and simulation model integration possible without custom programming overhead.
The episode also looks ahead: real-time optimization based on environmental and electricity market conditions, the vision of shipping a DataFabric Core with every containerized DAC unit as a Digital Twin as a Service, and Stefan's upcoming course at TU Wien where chemical engineering students will get hands-on experience building digital twins with industry software.
In this episode, Florian Seidl, Business Unit Manager for Industrial Informatics at cts Group, sits down with Florian Werner, Principal Consultant at Xenium, to discuss what it actually takes to turn fragmented shop floor data into usable, scalable information in pharmaceutical manufacturing.
The conversation starts with what both see in reality when they walk into a pharma production site: package-unit driven organizations where every line comes from a different vendor, data lives in isolated PLCs with inconsistent naming, and the focus is on keeping production running — not on data strategy. From there, they work through the entire arc of a data-driven transformation: why a unified namespace based on ISA-95 matters, how to structure a project from strategy through rollout, and why the most critical phase — change management and user adoption — is consistently underestimated.
A recurring theme throughout the conversation is trust: trust on the shop floor from operators who have worked with the same system for 15 years, trust from management who needs clear risk assessment and budget planning, trust from IT who needs scoping transparency, and trust from OT who operates in 10-year lifecycle windows with validated systems. Both speakers share concrete examples of implementations that failed because change management was skipped, and how "front runner sites" and working demos build the credibility needed for enterprise-wide adoption.
The episode also covers the practical architecture of an industrial data fabric in pharma: standardized message brokering between MES and OT systems, data harmonization across 50+ devices with different naming conventions, DQ/P system environments for GxP-compliant validation, and why the platform becomes an enabler for use cases that don't exist yet — including AI initiatives that require harmonized data in one place. The conversation closes with both looking ahead to joint conversations at Pharma Manufacturing Europe in Berlin.