A hybrid predictive model has been developed for accurate prediction of thermodynamics of carbon dioxide separation by aqueous alkanolamines. The model incorporates equation of state/excess Gibbs energy model into Kent–Eisenberg approach to predict carbon dioxide–alkanolamine–water equilibria. The approach imparts theoretical corrections to Kent–Eisenberg approach and significantly extends their range of application for monoethanolamine, diethanolamine, methyldiethanolamine, and 2-amino-2-methyl-1-propanol solutions. The proposed model suitably predicts thermodynamics of carbon dioxide separation, well beyond the regressed range of parameters. The results are in excellent agreement with experimental data for a wide range of process parameters and found superior to existing thermodynamic approaches.