Self-Driving Bioprocess Scale-Up (SELF-SCALE)

Industrial biotechnology plays a central role in the sustainable production of chemicals, materials, and pharmaceuticals. However, over 90% of these processes fail when scaled up to the industrial level. The main reason is the discrepancy in scale: biocatalysts (e.g. bacteria or enzymes) are developed on a tiny laboratory scale (microliters), but their behavior under real production conditions (> 100 liters) is not tested until very late in the process. As a result, critical problems – such as those caused by poor mixing – often don’t become apparent until changes are costly and time-consuming to implement.

This is where SELF-SCALE comes in, completely rethinking bioprocess development. Instead of carrying out design, process optimization, and scale-up sequentially, a self-regulating development cycle is created: From the very beginning, experiments are linked across different scales. Data from automated laboratory platforms continuously feeds into models, which in turn specifically suggest the most informative next experiments.

This is achieved by combining bioprocess engineering, laboratory automation, and AI. As a result, scaling issues are identified early, costly missteps are reduced, resources are conserved, and the path to industrial application is significantly accelerated.