The optimal design of grade crossings is a major challenge, as repairing grade crossings causes major delays to both rail and highway transportation systems. To improve the current state-of-the-art, this research project will focus on the development of a finite element model which can be used to simulate the interface between rail track and the highway pavement at grade crossings.
The model will be based on the state-of-the-art including international experience with track designs such as asphalt underlayment and overlayment. The multiphysics model will consider climatic loading, as well as loading caused by rolling stock and highway vehicles. The predicted model responses will then be incorporated into a physics-informed neural network which uses data from simulations and literature, as well as concepts of mechanics to predict mechanical responses in the track and ensure that predicted responses are realistic. The predicted responses will then be used as a framework for designing optimal grade crossings for the railroad industry.