Short-term passenger demand forecasting is of great importance to theon-demand ride service platform, which can incentivize vacant cars moving fromover-supply regions to over-demand regions. The spatial dependences, temporaldependences, and exogenous dependences need to be considered simultaneously,however, which makes short-term passenger demand forecasting challenging. Wepropose a novel deep learning (DL) approach, named the fusion convolutionallong short-term memory network (FCL-Net), to address these three dependenceswithin one end-to-end learning architecture. The model is stacked and fused bymultiple convolutional long short-term memory (LSTM) layers, standard LSTMlayers, and convolutional layers. The fusion of convolutional techniques andthe LSTM network enables the proposed DL approach to better capture thespatio-temporal characteristics and correlations of explanatory variables. Atailored spatially aggregated random forest is employed to rank the importanceof the explanatory variables. The ranking is then used for feature selection.The proposed DL approach is applied to the short-term forecasting of passengerdemand under an on-demand ride service platform in Hangzhou, China.Experimental results, validated on real-world data provided by DiDi Chuxing,show that the FCL-Net achieves better predictive performance than traditionalapproaches including both classical time-series prediction models and neuralnetwork based algorithms (e.g., artificial neural network and LSTM). This paperis one of the first DL studies to forecast the short-term passenger demand ofan on-demand ride service platform by examining the spatio-temporalcorrelations.
展开▼
