SISCOG participated in the 24th Conference on Artificial Intelligence (EPIA 2025), held from 1–3 October at the University of Algarve, near Faro.
On 3 October, Luís Albino presented a paper co-authored with Filipe Gouveia and Ricardo Saldanha, introducing a novel approach to one of rail’s most complex challenges: creating conflict-free train timetables.
The paper, titled “A MaxSAT Approach for the Train Timetabling Problem with Route Choice and Other Features”, proposes an optimization method capable of handling multiple constraints, including route choices, train cancellations, and alignment with ideal timetables. The approach was validated using real-world data from Swiss Federal Raiload and Washington Metro, involving scenarios with over 2,000 trains.
EPIA, organised under the patronage of the Portuguese Association for Artificial Intelligence (APPIA), is a leading international forum for AI research and innovation.
Interested in how AI and optimisation
can transform railroad timetabling?