A multi‐step airport delay prediction model based on spatial‐temporal correlation and auxiliary features

Abstract

Abstract Airport delay prediction system is fundamental to intelligent air traffic management. However, the prediction of airport delay is affected strongly by spatial‐temporal dependencies and other exogenous dependencies, which would bring serious challenges in prediction. In this paper, the APR‐LSTM model is proposed to address these challenges. The model is driven by the real‐world flight operation data. In the proposed prediction model, the dynamic air traffic delay networks are fed into PageRank algorithm to capture the dynamic spatial dependencies. The input sequences of airport delay vector is weighted by spatial dependencies before sending into the long short‐term memory network (LSTM) and sequence to sequence model, which can realise the multi‐step prediction and the joint mining of spatial‐temporal dependencies. Subsequently, the temporal attention mechanism and auxiliary features are introduced to obtain the long‐term temporal dependencies by exploring the relevance of different time steps and improve the accuracy of prediction model. Furthermore, the results of comparative experiments indicated that the proposed model achieve the best performance compared with other benchmark models. Finally, the spatial‐temporal correlation of different airports, the delay characterisations of networks and the application of prediction results are demonstrated in detail, enabling the interpretability of APR‐LSTM model.

Keywords

• Data handling techniques
• Aerospace engineering computing
• Search engines
• Data mining
• Neural nets