X-ray absorption fine structure (XAFS) spectroscopy is an excellent tool to probe the local atomic structure in ordered and disordered materials in a wide range of external conditions. Typically, the information reach region extends up to 3-10 Å around the absorbing atom and is limited by the mean free path of the excited photoelectron and a degree of the thermal and static disorder present in the material. While the multiple-scattering theory provides a solid background for XAFS interpretation, the inclusion of disorder remains a challenging task. In this talk, the use of atomistic simulation methods such as molecular dynamics (MD) and reverse Monte Carlo (RMC) will be discussed to address this issue. The novel approach based on the use of an artificial neural network for quick analysis will also be presented.