How Intercity Expressway Passenger Flow Varies? A View from Month, City Characteristics, and Weather at the Origin and Destination

Abstract This study aims to investigate the impact factors on intercity expressway passenger flow in China. A large-scale data set that integrated multiple data sources, including intercity passenger flow, city characteristics, and weather data, were constructed. To effectively handle and identify hidden patterns in the large-scale data, we employed Explainable Artificial Intelligence (XAI) models to analyze the associations between the impact factors and the intercity expressway passenger flow. The results show that among the XAI models used, the Explainable Extra Tree model that utilized SHAP (SHapley Additive exPlanations) values to explain the contributions of impact factors outperforms all others. In terms of the impact factors, we find that the intercity expressway passenger flow increased remarkably during certain months (such as February and September). In addition to city characteristics and weather variables that have been investigated by previous studies (e.g., GDP and temperature), the male-female ratio, the city tiers, and the temperature at the origin and destination cities were also found to be important contributors to expressway passenger flow. This study contributes to a deeper understanding of intercity expressway passenger flow patterns, offering valuable insights to policymakers for more effective management of intercity expressway passenger transportation.