Data-driven investigation of delay propagation in the UK rail network

The railway network plays a crucial role in supporting sustainable foundation for economic and social activity. Yet, as a complex system of systems, it is continuously affected by propagating disruptions that stall performance and undermine user trust. This study aims to establish novel insights into delay propagation dynamics triggered by local incidents in the UK railway network. To this end, a data analytics framework is proposed to reconstruct past delay propagation events using operational data sourced from the Rail Data Marketplace database. The framework comprises five modules that provide complementary perspectives on delay propagation dynamics. Applied to representative stations and incidents, this framework yields practical diagnostic insights and supports the data-driven identification of network areas that may benefit from targeted interventions. This work primarily aims to support future research on delay propagation forecasting by providing datasets against which simulation models can be validated, with the longer-term objective of enabling data-validated digital twins to support effective rail disruption management. In alignment with this objective, an open-source companion Python toolkit is released to facilitate wider adoption.