A statistical model of filled polymers is constructed based on computer imitational modeling. It can be used to construct model macrostructures for a wide class of materials to synthesize resistive composites with the desired properties.Progressively increasing requirements for products from multicomponent materials are met mostly through the development of theoretical and methodical background for structural analysis that allows key relationships describing the formation of the material properties to be established. In the applied aspect, this provides the stability of the characteristics and improves the methods and means of designing the composite materials.In the process of synthesis of new materials, the design decisions are often made based on regression models. They consider the influence of an arbitrary number of input factors and their possible combinations on an output parameter, but frequently cannot predict the behavior of materials without experimental investigations. It is well known that to regulate the properties of the material and to design it, a physical and mathematical model is preferable. However, many unsolved problems make the development of theoretical approaches to an analysis of the properties of multicomponent materials difficult. Therefore, model systems, despite certain limitations imposed on their properties, have the advantage of predicting the behavior of new materials and methods of their improvement. An important role is played by models of the macrostructure geometry that provide insight into the nature of the processes controlling the characteristics of the material [1]. A model multicomponent material is considered to be either homogeneous and anisotropic with average effective characteristics or two-component under assumption that the homogeneous and isotropic matrix is arranged inside a periodicity cell [1,2]. Many model data have already been accumulated. The problem existing in this approach is insufficiently scrutinized universal dependences that can be used to estimate the properties of designed multicomponent materials and to control their parameters. Most studies do not use the notion of random variable and the mathematical apparatus of random processes, which does not allow a great variety of statistical factors to be taken into account to obtain the desired characteristics. At the same time, the inhomogeneous structure of multicomponent materials causes the statistical distribution of their parameters that substantially influence the properties of the end product.Rubbers filled with technical carbon are widely used to manufacture the resistive products due to their considerable opportunities of profiling the end products, convenience of assembling, anticorrosive properties, and well-developed industrial production processes [2]. The main disadvantage of these products is an inhomogeneous distribution of the conducting component throughout the matrix volume due to imperfections in the technological production process, inhomogeneity of the supermolecular polymer struc...