Check out the final program!
09:00–09:15 Workshop Opening
09:15–10:15 Keynote Talk - Dynamic Games and Reinforcement Learning in Transportation Systems - Fernando P. Santos
Abstract - Urban transportation implies a highly complex socio-technical system. Planning urban transit requires accounting for interests, often conflicting, of multiple stakeholders and citizens. Likewise, designing fair and efficient transportation systems entails complex optimization problems and calls for new methods to model citizens behaviors when facing commuting decisions. In this talk I will present applications of evolutionary game theory and multi-agent reinforcement learning to model and optimize decisions in transportation systems. I will first present an evolutionary game theoretical model to understand the dynamics underlying the public, private and civic sectors when enacting incentives for electric vehicles adoption. I will then move to public transportation and showcase the application of reinforcement learning to 1) design fairer public transportation networks and 2) model the dynamic process of users’ routing decisions. We will observe that public incentives are key to trigger EV adoption, yet public-civil synergies are essential; multi-objective RL combined with Lorenz dominance can be used to identify network extensions with equitable reward distributions; and network expansions can simultaneously increase efficiency and amplify inequality, depending on individuals’ learning dynamics.
10:15–11:00 Coffee Break
11:00–11:25
Michał Szczygieł, Antoni Zięciak, Michał Pikus, Agata Kozina, Jarosław Wąs
Diffusion-Based Pedestrian Trajectory Prediction as Implicit Multi-Agent Coordination
11:25–11:50
Sudeep Aryan Gaddameedi, Mohit Garg, Eric Gyamfi
GRASP: A Scalable RAG-Enhanced Multi-LLM Architecture for Intention-Aware Autonomous Vehicle Decision-Making
11:50–12:15
Aitor López Sánchez, Marin Lujak, Frederic Semet, Holger Billhardt
Distributed Multi-Agent Routing Architecture for Reconfigurable, Task-Specific Fleets
12:30–14:00 Lunch Break
14:00–14:25
Rohit Aggarwal, Vimal Kumar
RideCopilot: An Agentic, LLM-Based Conversational System for Multi-Stakeholder Taxi Operations
14:25–14:50
Emmanouil Rigas, Georgios Chondropoulos, Nick Bassiliades
Scheduling Parcel Placement in Smart Locker Locations
14:50–15:15
Blerina Spahiu, Corinna Castaldi, Giuseppe Vizzari
Spatial-Temporal Analysis of Urban Cultural Events: A Visual Analytics Framework
15:30–16:15 Coffee Break
16:15–16:40
Filip Kamiński, Janusz Miller, Jarosław Wąs
A Stochastic Agent-Based Approach to the Modelling of Alpine Skiing
16:40–17:40 Discussion, Wrap-Up, and Workshop Closure
19:30 Workshop Dinner (TBD)