Clustering Analysis of Airport Traffic Similar Days Affected by Epidemic Based on HI-K-means

Abstract

During the epidemic period, the daily traffic fluctuation of the airport has a strong correlation with the current control policy. The analysis of traffic similar days can provide a reference for the optimization of airport traffic management. Aiming at the analysis of traffic similar days, a clustering model of traffic similar days based on HI-K-means (Hierarchical k-means clustering algorithm) is proposed. This algorithm combines the advantages of Hierarchical clustering and the K-means clustering algorithm and makes up for the defects of the two algorithms. Taking Tianjin Binhai Airport as an example, cluster analysis is carried out. Finally, it is concluded that the three types of traffic similar days can better match the daily traffic under different policies, indicating that the model has strong availability and high accuracy.