Passenger flow forecasting approaches for urban rail transit: a survey

Passenger flow forecast is the prerequisite and foundation for urban rail transit planning and operation. With the continuous expansion of rail network scale and the surge of passenger flow, the passenger flow prediction task becomes increasingly important and arduous. This paper presents an overview of the current research on passenger flow forecast in the field of urban rail transit, which mainly incorporates short-term passenger flow forecast, passenger flow forecast under emergency, typical days' passenger flow forecast and long-term passenger flow forecast for new opening line and extended line. The prediction characteristics in each subfield are discussed and the state-of-the-art forecasting approaches are reviewed. A multitude of existing studies shows that the forecast under different scenarios went through an imbalanced development. There are special prediction procedures and various applicable models in each scenario. Finally, we propose some future research prospects and discuss the potential applications of passenger flow forecast.