The dynamical response of a train rolling on a real track depends on several parameters. Most of them cannot be accurately identified and have to be considered as uncertain. The aim of this thesis is the construction of a probabilistic model of the rail fatigue life considering the track geometry and the rail wear as random fields modelled with the Karhunen-Loève expansion. The multivariate distributions of the random projection coefficients are characterized using a Polynomial Chaos Expansion (PCE) calibrated on measurements data of the track irregularities. The curve radius, the rail age and the train operational velocity introduce non-stationary effects that have to be taken into account to model the track. A validation of the random models is therefore performed using a set of measurements of the wheel-rail contact forces. A global sensitivity analysis is performed on some dynami-cal quantities of interest in order to quantify the impact of the random fields on the vehicle dynamics and the rail fatigue initiation. Since this step is computationally expensive, a PCE-based meta-modelling technique is employed to estimate the fatigue index.