Hub-and-spoke UAM network design model considering service quality

With the rapid development of the urban low-altitude economy, scientific layout of vertical take-off and landing (VTOL) facilities and the construction of an efficiently operated urban air mobility (UAM) network constitute the key and core of UAM development. Differentiating VTOL facilities by capacity and function into vertiports and vertistops, this study established a non-strict multi-allocation two-level hub-and-spoke UAM network that incorporates capacity constraints. A multi-objective optimization model was established with the objectives of minimizing VTOL facility construction cost and network transportation cost, minimizing total travel time, and minimizing the average service quality penalty score. The model was solved using a hybrid framework based on the non-dominated sorting genetic algorithm Ⅲ, enhanced by an embedded variable neighborhood search to co-optimize the location decisions for VTOL facilities and the network allocation scheme. Ground traffic travel data of Beijing was taken as a case study. UAM travel demand was predicted and a candidate VTOL facility set was generated. Based on this, model validity was verified and parameter sensitivity experiments were conducted. Research results show that, compared to traditional two-level network structures, the two-level hub-and-spoke network reduces total cost by 0.9%, reduces total travel time by 12.1%, shortens the average travel time per unit demand by 4.12 minutes, and delivers superior service quality. The number of VTOL facilities significantly affects objective function values and network congestion. An optimal trade-off among all objectives is achieved when deploying 3 vertiports and 8 vertistops. The proposed model thus enhances both economic benefits and transportation efficiency while improving UAM service quality, offering scientific decision support for the planning of hub-and-spoke UAM networks.