Importance measures-based prioritization for improving the performance of multi-state systems: application to the railway industry

The railway industry is undertaking significant efforts in the application of reliability-based and risk-informed approaches for rationalizing operation costs and safety requirements. In this respect, importance measures can bring valuable information for identifying the actions to take for most effective system improvement. In this paper, the railway network is modelled within a multi-state perspective in which each rail section is treated as a component, which can stay in different discrete states representing the speed values at which the section can be travelled, depending on the tracks degradation and on the traffic conditions. The Monte Carlo method is used to simulate the complex stochastic dynamics of such multi-state system. A prioritization of the rail sections based on importance measures is then used to most effectively improve the performance of the rail network, in terms of a decrease in the overall trains delay. High-importance sections, i.e. with highest impact on the overall delay, are considered for a relaxation of their speed restrictions and the proposed changes are then verified, from the risk-informed perspective, to have negligible impact on the risk associated to the rail infrastructure.