Urban Rail Transit Demand Analysis and Prediction: A Review of Recent Studies

Zhiyan Fang, Qixiu Cheng, Ruo Jia, Zhiyuan Liu

Abstract: Urban rail transit demand analysis and forecasting is an essential prerequisite for daily operations and management. This paper categorizes the proposed demand forecasting methods, and focuses on traditional models, statistical models and machine learning approaches, according to their features and fields. Especially, influential and widely-used methods including the four-stage model, land use models, time series methods, Logit regression, Artificial Neural Networks (ANNs) and other referring methods are all taken into discussion.