Stochastic Evaluation of Railway Track Buckling Using Monte-Carlo Simulations

Abstract

The necessity of railway track buckling assessment stems from the critical need to mitigate the risks associated with track buckling, which can lead to considerable track system damage and pose significant risks to operational safety and efficiency. Given the challenges in accurately determining the track parameters that influence buckling, the inherent uncertainties in these parameters introduce additional complexity in the evaluation of railway track buckling. Consequently, a stochastic approach to buckling analysis becomes necessary for a more robust and realistic management of buckling risks. This paper introduces a stochastic methodology for evaluating track buckling, leveraging Monte-Carlo simulations and parallel computing to process track parameters as random variables across a huge number of simulations utilizing a dynamic three-dimensional (3D) track model. By conducting about 67,000 simulations, buckling probabilities are calculated based on the frequency of buckling occurrences, offering a probabilistic perspective on track stability management. The findings highlight the effectiveness of the stochastic evaluation method in promoting a risk-based approach to maintaining track stability, improving the precision and reliability of maintenance strategies of railway engineering.