Bus Passenger Volume Forecasting Model Based on XGBoost Integrated Learning Algorithm

Bus passenger volume forecasting is the basis of traffic planning and dispatching, but weather and temperature often greatly impact the short-term fluctuation of passenger volume, which makes bus planning difficult. With the popularity of bus smart cards, it is possible to use historical bus smart card data to forecast bus passenger volume. This paper presents a prediction model of integrated learning algorithm based on XGBoost, which uses the data to combine the weather, temperature, and other data affecting trips with Shaoxing bus smart card data to forecast Shaoxing's total passenger volume and commuter line passenger volume. Seven hyperparameters of the model are adjusted by GridSearchCV module. The average absolute error obtained in the verification set is 11.1146 and the accuracy of passenger volume prediction every 3 min reaches 95.6%, which indicates the model can predict passenger volume more accurately.