Prediction & Simulation

The ability to model bioprocesses that predict manufacturing performance should dramatically improve the efficiency of R&D. Predicting manufacturing performance from the target molecule sequence will accelerate R&D during discovery and development. Guided decision tools will discard undesired molecule candidates, and rapidly focus resources to a predicted scale-up success via minimal R&D experimentation. 

This vision will require the combination of predictive models alongside miniaturized tools (nano and microliter scale) that confidently mimic manufacturing bioprocesses. The miniaturized tools will allow hundreds of thousands of experiments to be executed, in an automated approach, to generate true big data sets that boost the modeling capability.

Sartorius is driving this journey by creating the hybrid mechanistic models, designing miniaturization concepts of a complete bioprocess and building the concepts for the digital infrastructure foundation. The creation of the bioprocess digital twin will drive R&D efficiency in multiple ways, including faster decision making such as molecule selection and process optimization. It will also enable advanced control strategies to improve manufacturing process reliability and efficiency, including autonomous type self-regulating process control.