The improvement of accuracy of short-term passenger flow prediction plays a key role in the efficient and sustainable development of metro operation. The primary objective of this study is to explore the factors that influence prediction accuracy from time granularity and station class. An important aim of the study was also in presenting the proposition of change in a forecasting method. Passenger flow data from 87 Metro stations in Xi’an was collected and analyzed. A framework of short-term passenger flow based on the Empirical Mode Decomposition-Support Vector Regression (EMD-SVR) was proposed to predict passenger flow for different types of stations. Also, the relationship between the generation of passenger flow prediction error and passenger flow data was investigated. First, the metro network was classified into four categories by using eight clustering factors based on the characteristics of inbound passenger flow. Second, Pearson correlation coefficient was utilized to explore the time interval and time granularity for short-term passenger flow prediction. Third, the EMD-SVR was used to predict the passenger flow in the optimal time interval for each station. Results showed that the proposed approach has a significant improvement compared to the traditional passenger flow forecast approach. Lookback Volatility (LVB) was applied to reflect the fluctuation difference of passenger flow data, and the linear fitting of prediction error was conducted. The goodness-of-fit (R²) was found to be 0.768, indicating a good fitting of the data. Furthermore, it revealed that there are obvious differences in the prediction error of the four kinds of stations.

The operation of the subway system necessitates a comprehensive understanding of passenger flow characteristics at station locations, as well as a keen awareness of the average travel distances at these stations. Moreover, the travel distances at the station level bear a direct relationship with the built environment composed of land use characteristics within the station’s catchment area. To this end, we selected the land use features within an 800 m radius of the station (land use area, distribution of points of interest, and the surrounding living environment) as the influencing factors, with the travel distances at peak hours on the subway network in Xi’an as the research subject. An improved SSA-XGBOOST-SHAP interpretable machine learning framework was established. The research findings demonstrate that the proposed enhanced model outperforms traditional machine learning or linear regression methods in terms of R-squared, MAE, and RMSE. Furthermore, the distance from the city center, road network density, the number of public transit routes, and the land use mix have a pronounced influence on travel distances, reflecting the significant impact that mature built environments can have on passenger attraction. Additionally, the analysis reveals a notable nonlinear relationship and threshold effect between the built environment variables comprising land use and the travel distances during peak hours. The research results provide data-driven support for operational strategy management and line capacity optimization, as well as theoretical underpinnings for enhancing the efficiency and sustainability of the entire subway system.