Many-objective process optimisation with constraints for continuous tableting lines : a case study in lovastatin

Background: Digital design, assisted by data science, experienced significant progress over recent years within the continuous manufacturing in the pharmaceutical sector. Research Objective: The project focuses the fundamental research on robust numerical and visual performance indicators for assessing performance for many-objective optimisation algorithms under multiple constraints Methods: A surrogate model-based machine learning algorithm is used, to train data-driven models that capture the manufacturing process behaviour. Then use optimisation algorithms to get optimal solutions. Results: >75% dissolution release could be achieved in 45 minutes.