Abstract
Bioprocess modelling, optimisation and scale-up are central components for improving sustainable manufacturing within pharmaceutical and chemical industries. However, developing accurate bioprocess digital twins remains a challenging process. Conventional mechanistic models are difficult to construct because of limited mechanistic understanding and large complexity of cellular metabolisms. While data-driven models have gained popularity, they often require large amounts of experimental data that is often time consuming to obtain and lack any quantitative description of the process. Hybrid modelling methods have emerged as promising alternatives however fail to provide physical insight to the root cause of model error. This work therefore presents a promising solution by developing a data-efficient symbolic regression (SR) based framework to enable the automated discovery of interpretable bioprocess models. A universal kinetic model backbone was used to capture overall process behaviour, while SR was applied to strategically uncover the structures of critical kinetic terms within the backbone. Two frameworks, embedding SR directly into the kinetic model backbone or identifying time-varying parameter profiles prior to SR, were benchmarked using an in-silico yeast fermentation case study. The results demonstrated that independently identifying individual kinetic terms was crucial for recovering the ground-truth model, while refining SR-generated candidates through a novel local iterative structural correction strategy significantly improved convergence to the true kinetic expressions, surpassing model-based design of experiments in data efficiency. This study therefore enables automated yet interpretable model construction for small-data bioprocess applications, paving the way towards augmented intelligence driven bioprocess modelling and accelerating digital twin development for process optimisation and control.