The present work reports new experimental CO2 solubility data of CO2 absorption in aqueous mixture of 1-(2-aminoethyl)piperazine (AEP) and N-methyldiethanolamine (MDEA) over the temperature and pressure range of (303.2 -> k323.2) K and (3-300) kPa, respectively. The total mass fraction for AEP and MDEA was fixed to be y(AEP) + y(MDEA) = 0.4 in the aqueous solution. In the experimental measurement, the samples with the five different composition of AEP: MDEA (y(AEP) : y(MDEA)) have been considered as 0.05:0.35; 0.10:0.30; 0.15:0.25; 0.20:0.20 and 0.25:0.15 respectively. The experimental solubility data is modeled using modified Kent-Eisenberg (KE) equilibrium model. The equilibrium constants related to AEP deprotonation, and carbamate formation reactions are regressed as a function of CO2 partial pressure, amine concentration and temperature to fit the equilibrium CO2 solubility data. The model predicted VLE results from KE model are in good agreement with the experimental data within an average absolute deviation (MD) of 5.94%. The solubility data is also correlated using Artificial Neural Network (ANN) approach. The predicted value from the feed forward neural network model, which involves Levenberg-Marquardt algorithm, shows very good agreement with the experimental data having mean square error (MSE) of 0.000413 and correlation coefficient (R-2) of 0.997. Thermophysical properties such as density, viscosity and surface tension of aqueous (AEP + MDEA) have been also measured in the temperature range of (303.2-343.2) K. The density and viscosity data is modeled as a function of temperature and amine composition using Redlich-Kister and Grunberg-Nissan models, respectively. The model results of these temperature dependent physical properties are also in good agreement with the experimental data. (C) 2018 Elsevier B.V. All rights reserved.