The composite polymeric electrolytes (CPE) are well known systems in which improvement in the electrical conductivity of the polymer is achieved by the addition of a nonconductive additive (ceramic powders, immiscible polymers). The filler forms small grains (characterized by the average radius R) randomly distributed in the polymer matrix in a uniform way. The currently widely accepted mechanism of enhanced conductivity was previously discussed in the literature and is based on the assumption of the formation of amorphous highly conductive grain shells of a particular thickness (t) on the polymer-filler interface due to the inhibition of the polymer crystallization process on the grain surface. The system is complicated and multiphase, thus very difficult for theoretical studies. In this paper a conductivity model based on the application of the Random Resistor Network approach is employed to study CPE. The first stage of the simulation is based on the generation of a virtual composite sample and its transformation into a three-dimensional resistor network. Two different algorithms were developed to find the electrical properties of the resulting network. One is leading to the calculation of the number of percolation paths in the sample and the other is giving directly the conductivity value. The dependency of conductivity on the grain filler concentration was studied for varying grain sizes, changing t/R ratio and for different statistical distributions of the t/R parameter. The obtained results of the simulation are in general agreement with the experimental data. (c) 2005 Published by Elsevier B.V.