Research on subway passenger flow prediction based on bp neural network

The swift progress of urban rail transit necessitates precise passenger flow forecasting, which serves as the cornerstone for rational train scheduling. It augments the efficiency of train formation capabilities and minimizes the time passengers spend reaching platforms and waiting for trains. This not only diminishes the operational expenses for the operating entity but also curtails travel time and economic costs for passengers. This article utilizes passenger flow data from Kuanzhai Alley Station on Chengdu Metro Line 4 as the research foundation. The gathered data is partitioned into training and validation sets, and a BP neural network-based time series passenger flow forecasting model is established using historical passenger flow data. By comparing the predicted results with actual outcomes, the experimental analysis reveals that the prediction method demonstrates high accuracy.