Prediction of Crowding Levels at Different Time Periods for Beijing Subway Line 6 Based on a Combined Model of Autoregression and Linear Regression with Exogenous Variables

As a critical traffic artery connecting the city center with the Tongzhou New Town subsidiary administrative center of Beijing , Beijing Subway Line 6 bears significant passenger flow pressure. Issues such as carriage overcrowding and low passenger comfort are prominent, especially in the section between “Jintai Road” and “Shilipu” stations. To address this issue, this paper employs a hybrid modeling approach combining an Autoregressive AR model with an exogenous variable of daily total passenger flow. By integrating statistical data on the real time crowding levels of Beijing Subway and the total passenger flow data from the preceding week, this study aims to achieve accurate predictions of carriage crowding levels for different time periods and directions. The research findings are expected to optimize passenger travel experience and provide theoretical references and practical insights for the intelligent operation of urban rail transit. Qishi Feng | Yuxiang Chen | Xiang Li | Sifan Zhang | Zhe Tan "Prediction of Crowding Levels at Different Time Periods for Beijing Subway Line 6 Based on a Combined Model of Autoregression and Linear Regression with Exogenous Variables" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-9 | Issue-5 , October 2025, URL: https://www.ijtsrd.com/papers/ijtsrd97646.pdf Paper URL: https://www.ijtsrd.com/computer-science/data-processing/97646/prediction-of-crowding-levels-at-different-time-periods-for-beijing-subway-line-6-based-on-a-combined-model-of-autoregression-and-linear-regression-with-exogenous-variables/qishi-feng