Civil Aviation Passenger Throughput Forecasting Model Based on Machine Learning

This article examines a predictive model for estimating civil aviation passenger traffic using the long short-term memory network (LSTM). Analyzing the historical flight data, meteorological data, and holiday information of an international airport between 2010 and 2020 helped create a forecasting model that effectively captures the long-term dependency of time series data. The process of preparing data includes removing data, filling in missing values, and standardizing data, and then combining multi-dimensional data with feature engineering to improve data quality. The outcomes of the study show that the LSTM model performs reliably on the training and validation data, obtaining an MSE of 0.35, RMSE of 0.59, and MAE of 0.48.This suggests that the model possesses strong prediction accuracy and generalization capabilities. The study in this paper confirms the efficiency of the LSTM model in predicting civil aviation passenger throughput, offers dependable decision-making support for airport management and flight scheduling, and enhances the model's performance by employing proper data processing and feature selection strategies.