Live
From In-line Data to Process Insights: PAT Meets Hybrid Modeling | Part 2 - Yeast: Holistic Monitoring, Forecasting and Efficient Scale-Up
Process Analytics
Part 2 | Yeast Cultures: Holistic Monitoring, Forecasting and Efficient Scale-Up
Building on Part 1, this second session in the webinar series shifts focus to yeast fermentation, exploring how inline sensing and AI-driven modeling can transform monitoring and scale-up across both small-scale and pilot-scale bioreactors. Experts from Hamilton, TH OWL, and DataHow share real experimental data and practical insights into making yeast bioprocesses smarter and more predictable.
What you'll learn:
- How to combine inline sensors (VisiFerm, EasyFerm, Incyte, Dencytee, CO2NTROL) with offline measurements to build a holistic picture of the fermentation process
- How to build real-time soft sensors and model forecasts for key yeast culture parameters — including viable cell density (VCD), total cell density (TCD), viability, substrate levels, and metabolites
- How AI-powered models can predict critical process events such as peak cell density, substrate limitation, viability decline, and product inhibition — supporting smarter harvest and feeding decisions
- How the same PAT and modeling approach scales from small lab reactors to pilot-scale bioreactors, with results from 16 real-world experiments across two yeast strains (W-68 and W-34/70)
- How anomaly detection can flag issues
Speakers
-
Dr. Katharina Dahlmann
PAT & Application Expert, Hamilton -
Prof. Dr. Björn Frahm
Professor for Biotechnology and Bioprocessing, TH OWL -
Miguel Pupo Correia
Process Modelling Engineer, DataHow