A Multi-Objective model for shared-ride automated services: Reducing the price of anarchy through centralized matching

Abstract

The rise of Automated Mobility-on-Demand (AMoD) services, such as Waymo and Zoox, has transformed urban mobility. However, the growing demand and expanding robotaxi fleets have the potential to exacerbate congestion. This study introduces a novel centralized ride-matching procedure to enhance the scheduling efficiency of Shared-Ride Automated Mobility-on-Demand Services (SRAMODS). The proposed multi-objective adaptive model addresses the perspectives of on-site, in-vehicle riders, and robotaxi operators by minimizing on-site waiting time, in-vehicle travel duration, and detour distance at each pairing epoch. The proposed model dynamically matches riders within each epoch, with each robotaxi accommodating up to 4 riders simultaneously. A case study using Chicago demand data demonstrates that different weighting distributions lead to distinct matching outcomes, and a balanced weighting across all objectives minimizes the total time spent. Compared to traditional decentralized ride-matching, SRAMODS reduces the price of anarchy—measured as distance traveled per rider—by up to 15%, highlighting the benefits of centralized control. These findings provide policy insights to encourage shared robotaxi adoption through centralized coordination, improving urban mobility while reducing operational inefficiencies and congestion.