ABSTRACT With the rapid advancement of urbanization in China, the increase in urban population and motor vehicles has caused problems such as traffic congestion, environmental pollution, and traffic accidents. In response, China has been promoting the development of urban rail transit, thereby optimizing the spatial layout and enhancing urban functions. In urban rail transit scheduling, passenger flow prediction is used as a key decision‐making basis and its results are essential to the smooth operation of urban rail transit. This paper takes the inbound passenger flow of the Shanghai subway as the research object and comprehensively considers multi‐source data such as automatic fare collection (AFC) credit card data, external environmental data, air quality data and the nature of the land around the subway station. By fully processing and analysing these data, the potential laws between them and passenger flow are deeply explored. On this basis, a short‐term passenger flow prediction model of urban rail traffic based on ResNet‐Bi‐At‐LSTM is constructed. In the end, through the example analysis of Shanghai urban rail transit, it is verified that the model proposed in this paper can more accurately and comprehensively predict short‐term passenger flow.
Xu et al. (Thu,) studied this question.