This research addresses the challenge of accurately estimating the liquid vapor thermodynamic equilibrium in mixtures of carbon dioxide and isopropyl acetate using artificial neural networks. The main goal is to develop a model that surpasses the limitations of traditional methods, providing more precise and efficient estimates. A methodology involving the design, training, and validation of a neural network, using experimental data for model adjustment, was employed. The findings indicate a significant improvement in the precision of thermodynamic equilibrium estimations compared to conventional approaches. The conclusions highlight the feasibility of artificial neural networks as an advanced tool for prediction in chemical engineering, offering valuable implications for the design and optimization of industrial processes. This study contributes to the advancement of knowledge in the mo...