In this work, we aim to develop a validated device-scale CFD model that can quantitatively capture both hydrodynamics and CO2 capture efficiency for an amine-based solvent absorber column with a random Pall ring packing. Eulerian porous-media approach and two-fluid model were employed, in which the momentum and mass transfer equations were closed by literature-based empirical closure models. We proposed a hierarchical approach for calibrating the parameters in the closure models to make them accurate for the studied packed column. Specifically, a parameter in the closure for momentum transfer was first calibrated based on a single experimental data. With this calibrated parameter, a parameter in the closure for mass transfer was next calibrated at a single operating condition. Last, the closure of wetting area was calibrated for each gas velocity at three different liquid flow rates. For each calibration, cross validations were pursued using the experimental data under operating conditions different from those used for calibrations. This hierarchical approach can be generally applied to develop validated device-scale CFD models for different absorption columns.