Fundamental understanding of transport properties of fibrous porous media is contingent upon in depth knowledge of their internal structure at the micro-scale. In this work computer simulations are explicitly coupled with X-ray micro-computed tomography (mu CT) to investigate the effect of micro-structure on permeability of fibrous media. In order to reach this aim, samples of layered nonwoven fabrics were produced and realistic 3D images of their structure were prepared using X-ray mu CT. A series of algorithms were developed to extract micro-structural parameters of fibrous media, including fibers population, orientation and diameter of each fiber as well as the local porosity of structure from high-resolution realistic 3D images. A Matlab-based program capable of producing fibrous structures with various fiber diameters, porosities, thicknesses and 3D fiber orientations was developed. The obtained parameters from mu CT images were then implemented into the simulation code to generate virtual fibrous structures. Prediction of permeability in realistic and virtual structures was done by fluid flow simulation through the micro-structure of porous media. The results indicated that both through and in-plane permeabilities are strongly dependent on the porosity of structure. It was established that the anisotropic nature of the geometry creates anisotropic permeability, with a ratio of 1.8. The anisotropy effect was found to be more profound at higher porosity values. Comparison of numerical results with experimentally obtained data and those of empirical, analytical, numerical, and experimental models were made. Considering the porosity of structures, acceptable agreement between the results and previously published findings was observed. (C) 2017 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.