Airport Passenger Throughput Prediction Based on Linear Regression

As one of the important industries of in the national economy and social development, civil aviation plays an important role in transportation. Accurate passenger throughput forecasting is also crucial for optimizing airport operations. This study aims to predict the passenger throughput of ZhengDing Airport in Shijiazhuang using the econometric method. Based on the analysis of the influences of GDP per capita, total population and number of flights on passenger throughput, a multiple linear regression model was established. Using the data from 2005 to 2023 as a sample, the significance of each factor to passenger throughput was determined through univariate and multivariate analysis. Considering the problems of multicollinearity and heteroscedasticity in the model, the weighted least square method (WLS) is used to modify the model, and finally, the optimal passenger throughput prediction model is obtained. The model results show that in the 2014 and 2015 forecasts, the error is only -0.10% and 2.09%, respectively, indicating that the model has a high accuracy in short-term forecasts. This study aids in the planning of airport resource allocation and offers a trustworthy forecasting tool for airport management.