EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

2038. A real-world locomotive scheduling problem with stochastic travel times and buffers

Invited abstract in session TC-54: Disruption management and recovery, stream Public Transport Optimization.

Tuesday, 12:30-14:00
Room: S01 (building: 101)

Authors (first author is the speaker)

1. Kanchan Joshi
Business Decisions and Analytics, University of Vienna
2. Jan Fabian Ehmke
Business Decisions and Analytics, University of Vienna

Abstract

Railways, offering environmental efficient transport services, are a potential alternative to roadways. However, the increasing complexity of rail networks and interactions between infrastructure, resource planning, particularly locomotive scheduling, and personnel deployment, poses significant challenges. Railway companies use optimization techniques to generate circulation plans, promising efficient resource utilization. Nonetheless, these plans are often deterministic and ignore information on disruptions arising from delayed crew, crowded infrastructure and breakdowns. As a consequence, ad hoc short-term rescheduling and adjustments are necessary, resulting in economic penalties and a negative impact on perceived service quality. In this paper, we aim to determine reliable locomotive schedules that minimize the number of locomotives and empty-run distances using a time-space network formulation. To generate reliable plans, we design and include buffers considering stochastic travel times. We evaluate the effect of inserting these buffers on the number of locomotive and empty-run distances for real-world Austrian railway use cases. At the same time, we assess schedule reliability through feedback from simulations. The trade-off between reliability, the number of locomotives, and empty-run distances can offer valuable insights to planners seeking to address disruptions without stressing already constrained infrastructure.

Keywords

Status: accepted

Back to the list of papers