Keywords: maintenance strategies, railways, rolling contact fatigue, meta-models, reinforced learning, degradation.

Railways play a vital role in sustainable transportation by offering exceptional energy efficiency and reduced greenhouse gas emissions compared to alternatives like road or air transport. However, the operational and maintenance expenses associated with railway infrastructure are substantial, exceeding 25 billion Euros annually in Europe, with track degradation costs being a significant portion. To address this challenge, an optimal maintenance strategy is essential to reduce costs while maintaining high-quality service standards. This study proposes a methodology that combines meta-models, Petri Nets, and advanced decision-making techniques to optimize maintenance decision-making in railway infrastructure. Specifically, meta-models are used to predict Rolling Contact Fatigue crack initiation, while Petri Nets are employed to model crack propagation and optimize maintenance decisions. The integration of these approaches aims to identify optimal maintenance strategies to mitigate crack progression, minimize operational costs, and ensure safety and reliability in railway systems. A case study demonstrates the effectiveness of the proposed methodology in identifying optimal maintenance strategies. The findings highlight the potential of integrating meta-models, Petri Nets, and advanced decision-making techniques for optimizing maintenance decision-making in railway infrastructure.

download the full-text of this paper (PDF, 12 pages, 686 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description