Research on Civil Aviation Transportation Scheduling Optimisation System Based on Artificial Intelligence

The rapid development of civil aviation industry makes it difficult for traditional flight scheduling methods to cope with the increasingly complex air transport demand. In this study, an AI-based civil aviation transportation scheduling optimisation system is designed, integrating a novel deep reinforcement learning framework with a validated multimodal fusion algorithm (MMFA) to address spatiotemporal dependencies in aviation data to construct the core architecture of the system. Measurement results show that the system effectively reduces the average flight delay time by 58.1%, improves the slot utilisation rate by 21.3%, increases the flight punctuality rate to 93.7%, and shortens the response time to emergencies by 62.5%. The high performance and significant economic benefits demonstrated by the system in the real environment provide a feasible solution for the intelligent upgrading of civil aviation transport.