AbstractShort-term passenger flow prediction in urban rail transit plays an important role because it in-forms decision-making on operation scheduling. However, passenger flow prediction is affected by many factors. This study uses the seasonal autoregressive integrated moving average model (SARIMA) and support vector machines (SVM) to establish a traffic flow prediction model. The model is built using intelligent data provided by a large-scale urban traffic flow warning system, such as accurate passenger flow data, collected using the Internet of things and sensor networks. The model proposed in this paper can adapt to the complexity, nonlinearity, and periodicity of passenger flow in urban rail transit. Test results on a Beijing traffic dataset show that the SARI-MA–SVM model can improve accuracy and reduce errors in traffic prediction. The obtained pre-diction fits well with the measured data. Therefore, the SARIMA–SVM model can fully charac-terize traffic variations and is suitable for passenger flow prediction.
Deep Learning Architecture for Short-Term Passenger Flow Forecasting in Urban Rail Transit
