Predicting the remaining useful life of metro pantograph sliding strips using gamma processes and its implications for maintenance scheduling

Abstract

The progressive wear of pantograph sliding strips on metro trains necessitates timely replacement to ensure safe and reliable operations. This study proposes an adaptive, data-driven framework for predicting the remaining useful life (RUL) of these components, leveraging operational data from Chongqing Metro Line 6. A Gamma-process model is employed to capture the wear behavior under real-world operating conditions, integrating historical records and new observations through Bayesian inference. Markov chain Monte Carlo (MCMC) sampling is then applied to solve the posterior distribution, with three parameter-estimation approaches compared and the model’s predictive accuracy evaluated across different life-cycle stages. The results demonstrate that incorporating prior knowledge significantly improves prediction accuracy. To showcase practical utility, the study devises a maintenance-scheduling strategy that integrates RUL forecasts with regular vehicle-maintenance intervals, thereby extending service life and reducing costs. Validated using real-world data, the proposed methodology offers a pragmatic tool for predictive maintenance in metro systems and can be adapted to similar engineering applications.