This paper presents the development and implementation of a multiprocessor system-on-chip solution for fast and real time simulations of complex and nonlinear wheel-rail contact mechanics. There are two main significances in this paper. Firstly, the wheel-rail contact laws (including Hertz and Fastsim algorithms), which are widely used in the study of railway vehicle dynamics, are restructured for improved suitability that can take advantage of the rapid developing multiprocessor technology. Secondly, the complex algorithms for the contact laws are successfully implemented on a medium-sized FPGA (Field-Programmable-Gate-Array) device using six NiosII processors, where the executions of the Hertz and Fastsim parts are pipelined to achieve further enhancement in doing multiple contacts and the operation scheduling is optimized. In the Fastsim part the floating point units with buffering mechanism are efficiently shared by five processors connected in a token ring topology. The FPGA design shows good flexibility in utilizing logic element and on-chip memory resource on the device and scalability for a significant speed-up on a larger device in future work.