Seminar by Aline de Mitri

On July 16, 2025, Aline de Mitri from our partner Técnico Lisboa delivered a seminar entitled “From Data to Membranes: Predicting Performance and Guiding Material Selection with Machine Learning”. The talk explored how machine learning can support and accelerate membrane research by enabling data-driven material selection and performance prediction.

The seminar began with an introduction to the growing role of machine learning in membrane development, highlighting its potential to complement experimental work and improve efficiency in identifying promising materials. Key concepts were presented in an accessible way, including common approaches used for regression and classification tasks.

Two practical case studies illustrated the application of these methods. The first focused on predicting gas permeability in polymeric membranes, demonstrating how machine learning models can help identify promising candidates for gas separation. The second case study showed how machine learning can be used to analyze membrane pore structure and predict the long-term performance of membranes in membrane distillation systems. These approaches offer valuable tools for monitoring membrane behavior and supporting optimization.

The seminar provided valuable insight into how machine learning can enhance membrane research by combining experimental knowledge with predictive modeling. It also highlighted the importance of interdisciplinary approaches in advancing membrane technologies and developing more efficient and reliable systems for future applications.