Efficient measurement of gas permeation through high-barrier materials requires effective theoretical and numerical tools for data reduction and parameter extraction, to reduce the measurement period within error tolerance. An empirical permeation model was developed in this regard, by comprehensively considering gas–surface interactions as well as time and concentration-dependent diffusivity of the permeant particle. An algorithm was developed based on particle swarm optimization (PSO), to numerically fit the experimental data to the permeation model and to extract the characteristic permeation parameters. Compared with the traditional models, the comprehensive model is more accurate to predict the actual permeation phenomena, allowing 75% reduction of the measurement period within 10% tolerance on the extracted steady-state permeability. In addition, the PSO algorithm demonstrates extended applications in other non-steady-state transport problems such as convective–diffusive heat transfer.