Performance Prediction of Airport Traffic Using LSTM and CNN-LSTM Models

Willy Riyadi,Jasmir Jasmir
DOI: https://doi.org/10.30812/matrik.v22i3.3032
2023-07-28
Abstract:During the COVID-19 pandemic, airports faced a significant drop in passenger numbers, impacting the vital hub of the aircraft transportation industry. This study aimed to evaluate whether Long Short-Term Memory Network (LSTM) and Convolutional Neural Network - Long Short-Term Memory Network (CNN-LSTM) offer more accurate predictions for airport traffic during the COVID-19 pandemic from March to December 2020. The studies involved data filtering, applying min-max scaling, and dividing the dataset into 80% training and 20% testing sets. Parameter adjustment was performed with different optimizers such as RMSProp, Stochastic Gradient Descent (SGD), Adam, Nadam, and Adamax. Performance evaluation uses metrics that include Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and R-squared (R2). The best LSTM model achieved an impressive MAPE score of 0.0932, while the CNN-LSTM model had a slightly higher score of 0.0960. In particular, the inclusion of a balanced data set representing a percentage of the base period for each airport had a significant impact on improving prediction accuracy. This research contributes to providing stakeholders with valuable insights into the effectiveness of predicting airport traffic patterns during these unprecedented times.
What problem does this paper attempt to address?