DoE, Mechanistic Modeling, And Artificial Intelligence In Chromatography

Modeling can make process development more efficient by reducing experimental effort, improving decision-making, and building a deeper understanding of complex processes. Statistical modeling through design of experiments (DoE), mechanistic modeling, and machine learning or artificial intelligence each offer distinct advantages. Selecting the right approach requires an understanding of its fundamental principles, data needs, technical requirements, and potential limitations. Martin Sichting, Senior Product Manager at Cytiva, examines how these methods differ, where they are currently being applied, and how their role may evolve as modeling capabilities advance.

Explore the practical considerations for applying modeling strategies throughout process development.