OpenGOV: Constraint-Based Optimization of Track Occupation Diagrams
Cosling supports SNCF Réseau in developing OpenGOV, a decision-support tool designed to optimize Track Occupation Diagrams (TOD). This tool helps assigning trains to appropriate platform tracks and allocating circulation routes that minimize operational conflicts. Station capacity optimization relies on a detailed modeling of railway traffic constraints (infrastructure, rolling stock, operational standards, crossings, schedules, etc.) and the ability to quickly generate optimized solutions.
OpenGOV uses advanced constraint programming algorithms developed by Cosling, based on Choco Solver. Routing thousands of trains is executed in just a few seconds, with proven optimality using a multi-objective function. By drastically reducing the number of conflicts, these technologies improve station circulation flow and enhance the robustness of railway operations.
3 Tools
The OpenGOV suite includes three modules to support decision-making throughout the entire process:
- OpenGOV Etudes: provides objective quantitative support for preliminary studies
- OpenGOV Conception: optimizes transport plans from initial design to pre-operational adaptation
- OpenGOV Temps-Réel: assists agents in managing real-time disruptions
+50 Stations
OpenGOV Design has been deployed in more than 50 stations across France. The solution is highly configurable to accommodate the various specificities of different stations (e.g., 2TMV), whether modeled at a macroscopic or microscopic level.
+30,000
Every day, our algorithms optimize over 30,000 train circulations. This leads to better infrastructure utilization, more robust track occupation diagrams, and ultimately, improved punctuality.
Track Occupation Diagram Optimization from Design to Pre-Operational Adaptation
Track Occupation Diagrams are at the core of railway network management. Leveraging the efficiency of Choco Solver, OpenGOV optimizes these diagrams from the design phase and dynamically adjusts them during pre-operational adaptation. The mathematical model integrates the deviation from the previous solution into the multi-objective function, ensuring minimal unnecessary changes and a seamless user experience. These models account for potential disruptions and anticipate conflicts, enabling smooth traffic adjustments in response to real-world uncertainties.
