Strategic Urban Air Mobility Network Design with Disaggregated Demand and Scalable Hybrid Optimization

Abstract Strategic urban air mobility (UAM) network design must preserve trip-specific accessibility while supporting repeated optimization and subsequent operational planning. This study develops a six-stage framework that (1) constructs census-block-resolution potential trips and candidate stations, (2) generates complete ground--air--ground paths with multiple endpoint station alternatives, (3) screens paths by absolute and relative no-overhead time advantages and constructs retained-demand scenarios, (4) optimizes hub-location problems with minimum-utilization and functional-substitution requirements, (5) selects a representative network through infrastructure-scale and cross-scenario comparisons, and (6) derives a nested deployment sequence and operational allowances. Two objectives maximize potential total passenger travel-time saving or lexicographically prioritize assigned demand before travel-time saving. A hybrid local-branching and fix-and-optimize (LBFO) method enables repeated optimization across four retained-demand scenarios and 12 station budgets. At representative budgets, runtime summed across both objectives within each scenario decreases by 83.5--99.5\% relative to matched complete-domain Gurobi benchmarks, with primary-objective certified-bound shortfalls below 1\% in all eight tasks. In Greater Los Angeles, the selected network has 40 stations and 160 physical corridors, assigns 30,634 potential trips, and yields 1.89 million passenger-minutes of no-overhead travel-time saving. The demand-weighted median operational allowance is 25.7~min, and 95\% of assigned demand has an allowance of at least 19.0~min while retaining a 40\% relative-saving requirement. These outputs connect strategic network selection to phased deployment and subsequent scheduling, fleet-sizing, and vertiport-capacity analysis.