Urban spatial data-driven planning and operations for eVTOL city logistics under airspace, energy and charging constraints

Urban Air Mobility (UAM) logistics planning is still often carried out on abstract Euclidean instances, in which demand generation, low-altitude airspace restrictions, payload-dependent energy use, and en-route charging are modeled separately. This paper proposes a city-data-driven planning and operations framework for eVTOL city logistics that connects these elements within a single workflow. Building heights, points of interest, vertiport locations, and airport-related no-fly-zone boundaries are used to construct proxy delivery-demand scenarios, feasible operating areas, and candidate charging resources for comparative network planning, rather than to estimate observed delivery demand. A payload-sensitive energy model and a hybrid routing heuristic then jointly account for airspace feasibility, payload variation, route planning, and en-route charging decisions. The computational behavior of the heuristic is first examined on synthetic instances with 20, 50, and 100 customers, after which the framework is applied to a Chengdu case study built from real urban spatial inputs and simulated 50-customer task batches. The heuristic solves all instances with up to 50 customers at lower cost than ant-colony and particle-swarm benchmarks; in the Chengdu case, paired comparisons on 30 shared task batches show that the proposed charging layout attains costs statistically indistinguishable from six demand-weighted facility-location baselines under Holm-corrected Wilcoxon tests, with paired differences within about ± 1.2 % , while explicitly incorporating the airspace, endurance, and charging feasibility that such benchmarks do not model. The framework is therefore intended as a comparative planning and network-screening tool under limited demand data, supporting early deployment decisions for eVTOL city logistics.