The purpose of this thesis was to verify the simulated wheel-rail forces from the simulation soft-ware Simpack by comparing the results with on-track measurements. The report constitutes a literature study on wheel-rail contact modelling and techniques for measurements of track forces and track irregularities. Furthermore, it describes the verification procedure itself and how the model was set up, and subsequently presents the results from the actual comparisons. Three versions of Simpack have been compared that utilize different types of contact models. Investigations on alternative ways of filtering track irregularities were conducted, and how the simulated forces were influenced by using worn wheel and rail profiles was also studied.The method was to realistically model the track sections in nine test cases with varying curve radii, considering the nominal track geometry and track irregularities. A vehicle model for the Regina 250 train was used for the simulations, which were then compared with the correspond-ing on-track measurements. Time histories and frequency contents were compared.It was found that the simulated wheel-rail forces correspond well to the measurements, at least for frequencies up to 5-10 Hz, and that no considerable deviations could be found when comparing the contact models. However, the simulation times differed and the discrete contact model introduced in Simpack 9.8 was particularly time demanding.The results shows that it is beneficial to filter the track irregularities before conducting simula-tions, since it saves simulation time while retaining an adequate accuracy of the track forces.The use of worn wheel and rail profiles could in this report not be shown to have any ma-jor impact on the wheel-rail forces, which is in contradiction with the hypothesis. However, it should be noted that the scope for this part was narrow and it merits a more thorough investigation.