Multi-Step Short-Term Traffic Flow Prediction Based on a Novel Hybrid ARIMA-LSTM Neural Network

Accurate and real-time traffic flow prediction is the foundation for intelligent transportation systems (ITSs). Since traffic flow time series contains both linear and nonlinear patterns, both theoretical and empirical findings have indicated that a combination of different models outperforms individual models. We propose a novel hybrid autoregressive integrated moving average (ARIMA) and long short-term memory (LSTM) neural network model (ARIMA-LSTM) for multi-step short-term traffic flow prediction. Firstly, we use the ARIMA model to extract linear parts. Then we formulate a novel neural network containing LSTM layers, a concatenation layer, and current linear components and a multi-output layer for multi-step prediction. Finally, the neural network is optimized on a global scale. To test the performance of proposed model, we use the freeway traffic volume data and employed individual ARIMA, LSTM models and the hybrid ARIMA-ANN model for comparison. The test results indicate the proposed hybrid ARIMA-LSTM model is a reliable model.