1295. Real-time crew rescheduling for disruption management in railway systems
Invited abstract in session WC-20: Topics in Combinatorial Optimization 2, stream Combinatorial Optimization.
Wednesday, 12:30-14:00Room: Esther Simpson 2.11
Authors (first author is the speaker)
| 1. | Manuel Schlenkrich
|
| Institute for Transport and Logistics Management, WU Vienna University of Economics and Business | |
| 2. | Valentina Cacchiani
|
| DEI, University of Bologna | |
| 3. | Vera Hemmelmayr
|
| Vienna University of Economics and Business (WU) |
Abstract
A well-functioning transport system is the backbone of economic activity. Disruptions due to extreme weather conditions or technical failures happen frequently and reduce the efficiency of the railway system. Since it is mostly impossible to avoid disruptions, a fast and adequate response is necessary. Therefore, advanced decision support systems with the potential to increase the efficiency of the railway system are required.
In this work we focus on real-time railway crew rescheduling. The goal is to quickly find a new crew schedule that does not differ too much from the original crew schedule and avoids trip cancellations as much as possible, while still being feasible according to the labor rules and operational constraints. We consider railway crew rescheduling for freight trains along multiple days in real-time, which limits the computation time to only a few minutes.
We develop a metaheuristic method, namely a Variable Neighborhood Search, to quickly generate high quality crew plans to recover from a disruption. We propose several neighborhood structures and investigate different types of disruptions regarding their duration and severeness. We test the algorithm on real-world data from Rail Cargo Austria to support the railway operator's decision making.
Keywords
- Railway Applications
- Scheduling
- Metaheuristics
Status: accepted
Back to the list of papers