Predicting Subway Passenger Flows under Incident Situation with Causality
Xiannan Huang, Shuhan Qiu, Quan Yuan, Chao Yang
TL;DR
The paper tackles predicting subway passenger flows during incidents by decoupling normal-condition OD flow prediction from incident-induced causal effects. It introduces a two-stage framework: first, a conventional OD predictor trained on normal data, and second, a causal-effect predictor calibrated via synthetic-control-based identification and placebo tests to forecast incident impacts; final predictions are obtained by integrating both components. Empirical results on Shanghai subway data show consistent accuracy gains, with particularly large improvements for OD pairs significantly affected by incidents (up to ~30%). The approach enhances interpretability by isolating incident mechanisms and offers practical tools for managers to anticipate congestion and plan mitigations, while remaining adaptable to future methodological refinements and richer data sources.
Abstract
In the context of rail transit operations, real-time passenger flow prediction is essential; however, most models primarily focus on normal conditions, with limited research addressing incident situations. There are several intrinsic challenges associated with prediction during incidents, such as a lack of interpretability and data scarcity. To address these challenges, we propose a two-stage method that separates predictions under normal conditions and the causal effects of incidents. First, a normal prediction model is trained using data from normal situations. Next, the synthetic control method is employed to identify the causal effects of incidents, combined with placebo tests to determine significant levels of these effects. The significant effects are then utilized to train a causal effect prediction model, which can forecast the impact of incidents based on features of the incidents and passenger flows. During the prediction phase, the results from both the normal situation model and the causal effect prediction model are integrated to generate final passenger flow predictions during incidents. Our approach is validated using real-world data, demonstrating improved accuracy. Furthermore, the two-stage methodology enhances interpretability. By analyzing the causal effect prediction model, we can identify key influencing factors related to the effects of incidents and gain insights into their underlying mechanisms. Our work can assist subway system managers in estimating passenger flow affected by incidents and enable them to take proactive measures. Additionally, it can deepen researchers' understanding of the impact of incidents on subway passenger flows.