One-shot traffic assignment with forward-looking penalization

Traffic assignment (TA) is crucial in optimizing transportation systemsand consists in efficiently assigning routes to a collection oftrips. Existing TA algorithms often do not adequately consider realtimetraffic conditions, resulting in inefficient route assignments.This paper introduces Metis, a coordinated, one-shot TA algorithmthat combines alternative routing with edge penalization and informedroute scoring. We conduct experiments in several cities toevaluate the performance of Metis against state-of-the-art oneshotmethods. Compared to the best baseline, Metis significantlyreduces CO2 emissions by 18% in Milan, 28% in Florence, and 46%in Rome, improving trip distribution considerably while still havinglow computational time. Our study proposes Metis as a promisingsolution for optimizing TA and urban transportation systems.