Methods and software tools used to identify the material parameters of high-performance cementitious composites are presented. The aim is to provide techniques for the advanced assessment of the mechanical fracture properties of these materials, and the subsequent numerical simulation of components/structures made from them. The paper describes the development of computational and material models utilized for efficient material parameter determination with regards to a studied composite. This determination is performed with the help of experimental data from four-point bending tests. The data is used in inverse analysis based on artificial neural networks. Sensitivity analysis plays an important role in the process. It is a part of a complex methodology for the statistical and reliability analysis of structures made of high-performance cementitious composites. The procedure also utilizes statistical simulation of the Monte Carlo type for the preparation of a training set for the artificial neural network utilized in the material parameter identification process. In the case of fiber-reinforced concrete, the simulation mainly includes tensile strength, modulus of elasticity and the parameters of the tensile softening model.