Smart Biopharmaceutical Process Development
Application of machine learning and integrated process modelling
Biopharmaceutical production process development is typically sequential and time intensive. Traditional methods, which optimize each step individually, overlook potential interactions between steps, compromising overall efficiency.
To address this, Boehringer Ingelheim has implemented an innovative approach using integrated process models using machine learning and genetic algorithms. These data-driven models, applied to solubilization and refolding operations, predicted a twofold productivity increase. When extended to include capture chromatography, productivity increased by 50% to 100%, depending on the baseline process.
These results underscore the value of machine learning and optimization algorithms in process development and the impact of integrated process models across the whole process chain, including all unit operations.
Learn more about this innovative approach in process development by click the following link to read the scientific paper by Boehringer Ingelheim’s experts.