Urban air traffic has gradually attracted attention in recent years, which will bring endless vitality to future urban development. An objective and accurate method of assessing vertiport capacity is the basis for the air traffic flow management of UAVs, which plays an important role in improving the efficiency of urban airspace resources. First of all, this paper establishes a theoretical capacity calculation model for ground facilities such as takeoff and landing platforms, taxiways and aprons, respectively. Next, this paper analyzes the service characteristics of each ground facility and establishes different types of UAV queuing systems to obtain UAV delay curves based on a UAV Poisson flow arrival model. Subsequently, a suitable acceptable delay level is selected to obtain the corresponding UAV flow, which means the actual capacity of UAV operations. Eventually, the validity of the model is verified through actual drone data arithmetic examples. The calculation results show that the combination of “1 landing + 2 takeoffs” can achieve better capacity results and that the landing platform is more prone to congestion than the takeoff platform. Change in average service time has the greatest impact on the apron capacity, and the takeoff platform is the most sensitive to changes in the acceptable delay level.

City air traffic as a new transportation mode has gradually attracted attention in recent years which will bring endless vitality to future urban development. An objective and accurate assessment of the vertiport capacity for UAVs (Unmanned Aerial Vehicles) is the basis for implementing air traffic flow management for UAVs, which is also a prerequisite for improving the efficiency of urban airspace resources used. Firstly, new topology designs are proposed and named as connected and compact topology designs based on the existing central airport topology design. Subsequently, three modes of operation are summarized for vertiports with multiple TLOF pads: independent operation, dependent operation, and segregated operation. In the next place, the overall traffic flow of the vertiport model is established based on AnyLogic while analyzing the logic of UAV operation in three modes as mentioned above. Eventually, according to the simulation results, the vertiport operation capacity, the UAVs delay, and surface area utilization under different operation modes and topology designs are analyzed. The simulation result shows that the overall average delay time of UAVs for independent operation mode is about 100 s less than that of segregated operation and it also shows that the utilization rate of independent operation mode under central design is as high as 54.42% while the utilization rate of TLOF pads of other design is less than 50%, and its vertiport capacity is the largest, so the independent operational mode under central configuration is the optimal combination.

The terminal area is a convergence point for inbound and outbound traffic, and it is characterized by a complex airspace structure and high traffic density. It is an area that frequently experiences flight congestion and ground delays. A system capable of the intelligent, reliable, timely, and accurate identification of air traffic congestion for air–ground coupled flight flow constitutes a key technology with respect to unlocking the potential capacity of the terminal area, mitigating traffic congestion, and assisting air-traffic-control-related decision making. Therefore, this article aims to extract and analyze the multi-scale and multi-dimensional evaluation indicators of air–ground coupled flight flow congestion, use the TSNE-FCM algorithm to classify congestion levels, and, based on this work, construct a real-time and fast congestion identification model using the LightGBM algorithm. The case study analyzed China Baiyun Airport (CAN), and the experimental results indicate the following: (1) The congestion level classification achieved using the TSNE-FCM algorithm is superior to that achieved using the FCM algorithm. Furthermore, flight delays predominantly occur in slightly congested and congested states. (2) The congestion identification model based on LightGBM outperforms the XGBoost, RandomForest, and ExtraTree models. The macro-average and micro-average AUC curve areas for the LightGBM model were 0.96 and 0.96, respectively. The LightGBM model demonstrates excellent performance and is suitable for identifying congestion levels in practical engineering applications.