Comparing Lazy Constraint Selection Strategies in Train Routing with Moving Block Control

Stefan Engels,Robert Wille

2024-07-03

Abstract:Railroad transportation plays a vital role in the future of sustainable mobility. Besides building new infrastructure, capacity can be improved by modern train control systems, e.g., based on moving blocks. At the same time, there is only limited work on how to optimally route trains using the potential gained by these systems. Recently, an initial approach for train routing with moving block control has been proposed to address this demand. However, detailed evaluations on so-called lazy constraints are missing, and no publicly available implementation exists. In this work, we close this gap by providing an extended approach as well as a flexible open-source implementation that can use different solving strategies. Using that, we experimentally evaluate what choices should be made when implementing a lazy constraint approach. The corresponding implementation and benchmarks are publicly available as part of the Munich Train Control Toolkit (MTCT) at <a class="link-external link-https" href="https://github.com/cda-tum/mtct" rel="external noopener nofollow">this https URL</a>.

Systems and Control,Computational Engineering, Finance, and Science

What problem does this paper attempt to address?

### Problems the paper attempts to solve

This paper aims to solve the problem of how to optimize train paths in railway transportation with a moving - block control system. Specifically:

1. **Background**:
- Railway transportation is crucial for sustainable transportation systems and the demand is constantly increasing.
- Adding new tracks to increase capacity is both resource - consuming and time - consuming, so it is necessary to use modern train control systems to increase capacity.
- Modern train control systems (such as moving - block) allow trains to follow more closely on the existing infrastructure, thus significantly increasing capacity.

2. **Deficiencies in existing research**:
- Although path optimization for traditional train control systems has been well - studied, there is relatively little research on path optimization under moving - block control.
- There is a lack of detailed evaluation of the so - called "lazy - constraint" strategy, and there is no public implementation.

3. **Research objectives**:
- Provide an extended method and a flexible open - source implementation that can use different solution strategies.
- Experimentally evaluate which strategies should be selected when implementing the lazy - constraint method.
- The proposed model extends previous solutions, allowing for more general timetable requests and more accurate modeling of train separation, especially in scenarios near stations.

### Main contributions

1. **Model improvement**:
- A new mixed - integer linear programming (MILP) model is proposed, which can more accurately model train separation on short track segments, taking into account the actual train length.
- The model allows for more flexible timetable requests and can more accurately model train separation near stations.

2. **Evaluation of lazy - constraint strategies**:
- Multiple lazy - constraint selection strategies are implemented and experimentally evaluated.
- The results show that the lazy - constraint method performs better when multiple constraints are added simultaneously, but the differences between different strategies are not significant.

3. **Open - source implementation**:
- A flexible open - source implementation is provided, and users can select different solution strategies as needed.
- The open - source toolkit (Munich Train Control Toolkit, MTCT) is available on [GitHub](https://github.com/cda - tum/mtct).

Through these improvements and evaluations, this paper provides new methods and tools for train path optimization in railway transportation, which helps to improve the efficiency and safety of railway systems.