Does Laziness Pay Off? - A Lazy-Constraint Approach to Timetabling
- Timetabling is a classical and complex task for public transport operators as well as for railway undertakings. The general question is: Which vehicle is taking which route through the transportation network in which order? In this paper, we consider the special setting to find optimal timetables for railway systems under a moving block regime. We directly set up on our work of [8 ], i.e., we consider the same model formulation and real-world instances of a moving block headway system. In this paper, we present a repair heuristic and a lazy-constraint approach utilizing the callback features of Gurobi, see [3]. We provide an experimental study of the different algorithmic approaches for a railway network with 100 and up to 300 train requests. The computational results show that the lazy-constraint approach together with the repair heuristic significantly improves our previous approaches.
| Author: | Torsten Klug, Markus Reuther, Thomas SchlechteORCiD |
|---|---|
| Document Type: | In Proceedings |
| Parent Title (English): | 22nd Symposium on Algorithmic Approaches for Transportation Modelling, Optimization, and Systems (ATMOS 2022) |
| Volume: | 106 |
| First Page: | 11:1 |
| Last Page: | 11:8 |
| Series: | Open Access Series in Informatics (OASIcs) |
| Publisher: | Schloss Dagstuhl -- Leibniz-Zentrum für Informatik |
| Date of first Publication: | 2022/09/06 |
| DOI: | https://doi.org/10.4230/OASIcs.ATMOS.2022.11 |

