Advanced point-merge system design method integrated with multi-source airspace features and artificial intelligence

In this study, a flight procedure design method that integrates multi-source airspace features and artificial intelligence is proposed under widespread artificial intelligence adoption and the progressive development and application of advanced air traffic management systems. Aiming to alleviate airspace capacity and demand imbalances in flight operations, the study focuses on terminal area flight procedure design, selecting the Point Merge System (PMS) as the subject—an advanced navigation-based procedure that enhances aircraft sequencing and facilitates efficient air traffic management within terminal airspace. The study first analyzes the relationship between terminal area operational environments and PMS structural configurations, and develops a corresponding mathematical model. Next, a flight procedure parameter generation framework based on a Condition Variational Autoencoder (CVAE) is designed to enable the transformation of terminal area layout and service capacity characteristics into structured PMS design outputs. Finally, Xiamen Gaoqi International Airport is used as a case study to demonstrate the proposed method. A designed PMS scheme is developed for the airport, and assessments are conducted in terms of safety and operational efficiency. Simulation results indicate that the proposed method can reduce airspace conflict risks by up to 38.10% and improve flight operational efficiency by 28.83%. These findings validate the effectiveness of the proposed flight procedure design approach. Furthermore, the method exhibits advantages in scalability and efficiency, allowing for rapid integration with existing procedure design processes.