The operation of PSA (Pressure Swing Adsorption) processes is a highly nonlinear and challenging problem. We propose a systematic procedure to achieve the optimal operation of a PSA process. The model of the PSA process for CO2 separation and recovery is developed first and optimization is performed to identify optimal operating conditions based on the model. The effectiveness of the model developed is demonstrated by numerical simulations and experiments using CO2 and N2 gases and zeolite 13X. Breakthrough curves and temperature changes in the bed are computed from the model and the results are compared with those of experiments. The effects of the adsorption time and reflux ratio on the product purity and the recovery are identified through numerical simulations. The optimization problem is formulated based on nonlinear equations obtained from simulations. The optimal operating conditions identified are applied to experiments. The results show higher recovery of CO2 under optimal operating conditions.