Optimizing urban transport: an optimization framework for urban air mobility site selection and route operation under real-world travel demand

With the increasing commuting and logistics transportation distances in urban areas, transportation efficiency is constrained, necessitating the development of innovative mobility solutions. The development of urban air mobility (UAM) system with electric vertical takeoff and landing aircrafts (eVTOL) could alleviate congestion by leveraging underused low-altitude airspace. However, realizing this vision requires proper infrastructure design and route operation planning. This paper focuses on the optimization of UAM vertiport site selection and transport network operation based on traffic big data. The k-means++ clustering algorithm is used to generate the potential vertiport locations for travel demand points. A mixed integer nonlinearprogramming model (MINLP) is formulated, with site selection and eVTOL dispatch as the main decision variables and the goal of minimizing the costs of both operator and users to the greatest extent. The model determines the best choice of eVTOL site locations and route operations under a certain service penetration rate. The model solution provides key infrastructure decisions for implementing an urban eVTOL network. This paper takes the highway traffic data of the Guangdong-Hong Kong-Macao Greater Bay Area as an example, and the sensitivity analysis provides insights into critical aviation parameters and service considerations. The research contributes a novel, data-driven optimization framework for developing suitable UAM networks to advance the adoption of urban air mobility.