Turnouts are a key element of railway system. They are also one part of the railway system with the highest number of degradation modes and failures for a number of reasons, including dynamic loads generated from non-linearities in the rail geometry and track support stiffness. It is therefore necessary to optimise the performance of the system in terms of its dynamic behaviour taking into account the effects on longterm damage evolution. The main aim of the study, described in this paper, is to optimise the rail-pad stiffness in the crossing panel in order to achieve a decrease in the main indicator for ballast settlement, which is ballast pressure. A threedimensional vehicle-track interaction model has been established, considering a detailed description of the crossing panel support structure. A genetic algorithm has been applied to find the optimum rail-pad combination for a specific case where variations in travelling speed and support conditions have been considered.