EURO 2025 Leeds
Abstract Submission

1077. Game-theoretic Tree Regression for Surrogate-based Optimisation of Active Pharmaceutical Ingredient Manufacturing

Invited abstract in session MB-4: Interpretable Optimization Methods and Applications, stream Data Science meets Optimization.

Monday, 10:30-12:00
Room: Rupert Beckett LT

Authors (first author is the speaker)

1. Artemis Tsochatzidi
Chemical Engineering, University College London
2. Francesca Cenci
GlaxoSmithKline (GSK)
3. Magdalini Aroniada
GlaxoSmithKline (GSK)
4. Lazaros Papageorgiou
Chemical Engineering, University College London

Abstract

Modern industries rely on advanced modelling techniques to enhance process efficiency, yet the computational complexity of these models often limits their direct use in optimisation. Surrogate-based approaches provide an effective alternative, enabling efficient exploration of the solution space. In this work, we introduce a game-theoretic tree regression approach for surrogate-based optimisation, integrating a multi-target tree regression model to approximate complex process dynamics. The proposed approach formulates optimisation as a strategic decision-making problem, to reveal trade-offs between conflicting objectives such as yield and purity in Active Pharmaceutical Ingredient (API) manufacturing. By leveraging Pareto fronts and Nash equilibrium, the methodology provides a structured mechanism for navigating these competing goals and deriving optimal operational strategies. The proposed approach enhances interpretability, computational efficiency, and practical applicability, offering a powerful tool for decision-making in pharmaceutical manufacturing and beyond.

Keywords

Status: accepted


Back to the list of papers