Long-Term Passenger Flow Forecasting for Rail Transit Based on Complex Networks and Informer

Area of Science:

• Urban planning and transportation science
• Data science and artificial intelligence
• Network theory and complex systems

Background:

• Urbanization drives increasing passenger flow in rail transit, necessitating accurate long-term forecasting for operational efficiency and service quality.
• Forecasting passenger flow is complex due to intricate network structures and external factors like seasonality.
• Existing models struggle with large-scale, complex transit data and network dependencies.

Purpose of the Study:

• To develop an optimized forecasting model for urban rail transit passenger flow.
• To enhance long-term forecasting accuracy by incorporating inter-station influences using complex network theory.
• To provide a robust decision support tool for urban rail transit management.

Main Methods:

• An optimized Informer model was developed for long-term passenger flow forecasting.
• Complex network theory was integrated to account for the influences of interconnected stations.
• The model's performance was compared against ARIMA, LSTM, and Transformer models.

Main Results:

• The optimized Informer model demonstrated superior performance in processing large-scale, complex transit data.
• The model achieved higher accuracy in long-term passenger flow forecasting compared to traditional methods.
• The integration of complex network theory effectively captured network dependencies.

Conclusions:

• The proposed forecasting approach significantly improves the accuracy and efficiency of long-term passenger flow predictions.
• This method offers robust decision support for urban rail transit planning and management.
• The optimized Informer model represents a significant advancement in intelligent transportation systems.