Improved rail performance with better traffic management

Delayed trains cost time and money not just for passengers but also for railway operators and national economies. Enhancing the performance of rail networks and improving the exploitation of existing infrastructure is the COMBINE 2 project.

Since the IST programme-funded project ended early last year, the COMBINE 2 algorithms have gone on to be used commercially in the Milan metro and its system has undergone extensive field trials with real trains in The Netherlands, proving its potential to optimise the use of existing infrastructure by allowing real time rescheduling and rerouting of trains.

“COMBINE 2 answers a growing demand for optimisation from railway operators which are under pressure from passengers demanding better services and from the need to reduce costs,” explains Alessandro Mascis, the project’s scientific coordinator at Bombardier Transportation Italy. “In Europe big investments in new infrastructures are not foreseen in the short to mid-term for several reasons, making it essential for railway operators to be able to exploit existing railway networks to the maximum.”

COMBINE 2, a follow-on from the previous EC-funded project COMBINE, developed models and algorithms to efficiently manage the scheduling and routing of trains circulating in either fixed block or moving block signalling areas. Fixed block uses stationary signals, while in essence the newer moving block system uses signals linked to the train in front on the same line.

By improving coordination with detailed real-time information on train positioning in both fixed and moving block areas, the COMBINE 2 system enhances the performance of the whole rail network through optimising scheduling in each local sub-network.

“What is new is that COMBINE 2 proved a theory that defines the coordination rules to ensure that real-time re-planning in each sub-network results in a globally feasible solution for the whole network,” Mascis explains. “For example, the system is able to minimise the time and energy losses that occur when a freight train has to stop in front of a red signal by simply suggesting the train go slower in order meet a green signal.”

When such performance enhancing features are applied globally to the Traffic Management Systems used to govern rail networks the result is significant time and cost savings for rail companies and improved service for passengers. In addition, real-time management of the network permits more trains to operate, increasing services without the need to lay costly new infrastructure.

The COMBINE 2 algorithms are currently being used to automate and optimise management of Line 1 of the Milan metro which carries around half a million passengers a day, with the project partners planning more commercial applications in the future.