Can AI‐Based Surrogate Models Offer Added Value for Process Optimization?—Investigations Into Practical Application Examples in the C4 Verbund

On‐stream optimizing the performance of chemical processes requires accurate modeling and real‐time problem‐solving. Exploiting artificial neural networks as surrogate models offers a promising solution due to their capability of capturing complex correlations efficiently. Although artificial intelligence (AI)‐based surrogate models are theoretically well‐established and have been the subject of extensive research, practical applications in industrial settings are still rarely reported in the literature. This article therefore demonstrates the effectiveness of AI surrogate‐based optimization through two real‐world case studies in the field of C4 chemistry.