Synergistic Skies: Distributed Optimization for eVTOL Dispatching and Charging in Advanced Air Mobility

Advanced Air Mobility (AAM) promises on-demand aerial services using electric vertical take-off and landing (eVTOL) aircraft across urban and regional networks, but scaling operations hinges on how vertiports allocate scarce pads and chargers while routing aircraft under stochastic demand and tight energy constraints. This paper develops a vertiport-centric, distributed optimization framework for Urban Air Mobility (UAM)—a metropolitan subset of AAM—in which each vertiport manages its local eVTOL fleet, charging resources, and queues, and coordinates with neighbors through lightweight signals. To ensure scalability and resilience to backhaul limitations, we employ the Alternating Direction Method of Multipliers (ADMM) with a consensus split that separates local subproblems solved in parallel at vertiports from global consistency projections. The method supports rolling-horizon execution, adaptive penalty tuning, and simple projections for capped-simplex charging constraints. In simulation scenarios representative of multi-vertiport UAM corridors, the proposed approach achieves higher service rates, lower passenger waiting times, more balanced pad utilization, and markedly reduced wall-clock solution times. The results indicate that distributed, infrastructure-aware coordination is a practical path to scalable, regulation-conscious AAM operations. We compare the proposed ADMM-based approach against a distributed mixed-integer linear programming (MILP) formulation. Results show that ADMM achieves near-equivalent service performance to distributed MILP while significantly reducing wall-clock solution time, improving scalability, and maintaining robust energy safety margins.