Flight Delay Prediction Method Based on LSTM

• Published in: 2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE) pp. 1 - 4
• Main Authors: Liu, JiaLin, Zhao, Jiajia, Lu, Yifeng, Wu, Yue
• Format: Conference Proceeding
• Language: English
• Published: IEEE 16.12.2022
• Subjects: flight delay forecast; multi classification prediction; short and long-term memory neural network
• DOI: 10.1109/ICARCE55724.2022.10046519

In recent years, the civil aviation industry is developing more and more quickly. Flight delay has become one of the biggest problems in the civil aviation industry at present. If flight delay can be predicted, the adverse impact of flight delay will be reduced to a greater extent, thus improving the travel efficiency of passengers and the work efficiency of the airport. Aiming at the problem, a multi classification flight delay prediction model based on long short term memory networks (LSTM) is constructed in this paper. This model will combine flight data and weather data, make use of the time series characteristics of flight data, and realize the multi classification prediction of flight delay because the delay of the previous flight plays a decisive role in the timely departure of the subsequent flight. The experiment shows that the prediction accuracy of this model is greatly improved compared with the traditional neural network.