We present an efficient computational procedure for the rapid prediction of the self-diffusivity of gas molecules in nanoporous materials by a combination of the Knudsen model, Rosenfeld's excess-entropy scaling method, and a classical density functional theory (DFT). The self-diffusivity conforms to the Knudsen model at low density, and the effects of intermolecular interactions at higher densities are accounted for by Rosenfeld's excess-entropy scaling method. The classical DFT provides a convenient way to calculate the excess entropy used in the scaling analysis. The hybrid computational procedure has been calibrated with MD simulation for the adsorption of H-2, He, Ne, and Ar gases in several nanoporous materials over a broad range of pressure. It predicts adsorption isotherms and different types of diffusion behavior in excellent agreement with the simulation results. Although the simulation of gas diffusion in nanoporous materials is extremely time-consuming, the new procedure is computationally very efficient because it uses only single molecular and thermodynamic parameters.