The optimization of the train frequency of the Airport express line (AEL) is crucial for improving the efficiency of air-rail intermodal transport. It directly influences passenger transfer convenience and overall service quality, thereby bolstering the competitiveness of the transport system This study focuses on the optimization of “AEL and Flight Succession” in the context of air-rail intermodal transport. By analyzing the departure and landing time of airport flights, we assess the demand from various passenger flows and identify key factors that impact the connection between the AEL and flights. Based on these factors, we develop a demand-driven optimization model for AEL frequency, aimed at minimizing total travel time and the number of unserved passengers. A simulated annealing algorithm is employed to solve this model. The Lanzhou-Zhongchuan AEL serves as a case study for validation. The results demonstrate that the optimized schedule reduces total passenger travel time costs by 0.93% and 3.82%, respectively, while accounting for passenger time sensitivity and fairness principles, with a difference of 2.89% between these scenarios. In addition, the optimization scheme decreases the number of unserved passengers by 14.7% and reduces the percentage of flights and trains failing to meet occupancy constraints by 17%. This study illustrates that the schedule optimization strategy not only effectively increases the number of served passengers but also significantly reduces total intermodal and commuter travel time. Such findings provide a solid scientific foundation for AEL operations and management to develop a more efficient and rational train schedule in the context of air-rail intermodal transport.

The purpose of this paper is to optimize the accuracy of airport express passenger flow prediction so as to meet the need for the optimal allocation of traffic resources against the background of accelerated urbanization and the rapid development of airport express services. A fusion intelligence network model (FINM) is proposed, which integrates the advantages of convolutional neural networks, bidirectional long short-term memory networks, and gated recurrent units. Firstly, by using the powerful feature extraction ability of convolutional neural networks, local features and detail information are captured from the input data to improve the data representation ability. Secondly, bidirectional long short-term memory networks are used to process the sequence data, capture the global information and its context relationship, and enhance the model’s understanding of the dependence of time series data. Finally, gated recurrent units are introduced to simplify the computational complexity while maintaining high prediction accuracy and training efficiency. Based on the actual passenger flow data for Tianjin Metro Line 2 on a 30 min time scale, the proposed FINM is verified. The experimental results show that the model achieves an excellent performance, with 0.0160, 0.0947, 0.0160, 0.1255, 18.40, and 0.7788 in key indicators such as loss value (Loss Value), mean absolute error (MAE), mean squared error (MSE), root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R-Squared). Compared with the comparison algorithm, this model shows significant advantages in all indicators, which proves its effectiveness in dealing with complex time series data.