Tuning Evolutionary Multiobjective Optimization for Closed-Loop Estimation of Chromatographic Operating Conditions


Abstract

Purification is an essential step in the production of biopharmaceuticals. Resources are usually limited during development to make a full assessment of operating conditions for a given purification process commonly consisting of two or more chromatographic steps. This study proposes the optimization of all operating conditions simultaneously using an evolutionary multiobjective optimization algorithm (EMOA). After formulating the closed-loop optimization problem, which is subject to constraints and resourcing issues, four state-of-the-art EMOAs — NSGAII, MOEA/D, SMS-EMOA, and ParEGO — were tuned and evaluated on test problems created from real-world data available in the literature. The simulation results revealed that the performance of an EMOA depends on the setting of the population size, and constraint and resourcing issue-handling strategies adopted. Tuning these algorithm parameters revealed that the EMOAs, in particular SMS-EMOA and ParEGO, are able to discover reliably within 100 evaluations operating conditions that lead to high levels of yield and product purity.