Microscopic Computer-Aided Tools for Automated Railway Traffic Planning
Nikola Bešinović, Egidio Quaglietta, Rob M.P. Goverde
Abstract
Nikola Bešinović, Egidio Quaglietta, Rob M.P. Goverde
Abstract
To satisfy the growing demand in railway transportation, infrastructure managers have to design more effective timetables. To this aim it is required to rely on automatic timetabling tools that can provide feasible timetables with improved performances. In this paper the authors present a microscopic model of railway traffic to be used in the timetabling process for computing reliable train running, blocking times and minimum headway times, as well as for analysing new or existing timetables. Such a model can be plugged-in in any framework for timetable design and interfaced with typical macroscopic timetabling tools and/or models for scheduling energy-efficient train trajectories. This builds on the developments of the European project ON-TIME which defined an innovative timetabling framework to produce large-scale railway schedules that are operationally feasible, stable against delays, robust versus stochastic perturbations, and energy-efficient. The microscopic model introduced here focuses on guaranteeing feasibility and stability of railway timetables. Feasibility is ensured by a conflict detection module that re-computes train running, blocking, and minimum headway times in case the timetable has track conflicts at the level of block sections. Stability is guaranteed by verifying that the network capacity relating the scheduled services satisfies technical thresholds established by the UIC Code 406 using the timetable compression method. The application to a large part of the Dutch railway network shows how the microscopic tool is able to produce a feasible and stable timetable through an iterative process with a macroscopic timetabling model. Practitioners can now use this tool as an important component of the timetabling process to improve the general quality of timetables.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
To satisfy the growing demand in railway transportation, infrastructure managers have to design more effective timetables. To this aim it is required to rely on automatic timetabling tools that can provide feasible timetables with improved performances. In this paper the authors present a microscopic model of railway traffic to be used in the timetabling process for computing reliable train running, blocking times and minimum headway times, as well as for analysing new or existing timetables. Such a model can be plugged-in in any framework for timetable design and interfaced with typical macroscopic timetabling tools and/or models for scheduling energy-efficient train trajectories. This builds on the developments of the European project ON-TIME which defined an innovative timetabling framework to produce large-scale railway schedules that are operationally feasible, stable against delays, robust versus stochastic perturbations, and energy-efficient. The microscopic model introduced here focuses on guaranteeing feasibility and stability of railway timetables. Feasibility is ensured by a conflict detection module that re-computes train running, blocking, and minimum headway times in case the timetable has track conflicts at the level of block sections. Stability is guaranteed by verifying that the network capacity relating the scheduled services satisfies technical thresholds established by the UIC Code 406 using the timetable compression method. The application to a large part of the Dutch railway network shows how the microscopic tool is able to produce a feasible and stable timetable through an iterative process with a macroscopic timetabling model. Practitioners can now use this tool as an important component of the timetabling process to improve the general quality of timetables.
Key concepts: Headway, Computer science, Scheduling (production processes), Process (computing), Block (permutation group theory), Taxis, Iterative and incremental development, Stability (learning theory)