Passenger Flow Scale Prediction of Urban Rail Transit Stations Based on Multilayer Perceptron (MLP)

Accurately predicting passenger flow at rail stations is an effective way to reduce operation and maintenance costs, improve the quality of travel while meeting future demand. The improvement data acquisition capability allows fine-grained large-scale built environment be extracted. Therefore, this paper focuses on investigating relationship between around station discusses whether can applied prediction. Firstly, evaluation system influencing factors based multisource data. inner investigated using Pearson correlation analysis. Based this, a multilayer perceptron (MLP)-based prediction model was developed predict key stations. study results show that impact flow, MLP has better accuracy applicability. scale without historical thus are also applicable new