Anomaly detection in commercial aircraft landing at SSK II airport using clustering method

This study focuses on the critical importance of compliant landing procedures for commercial aircraft to mitigate the risk of accidents, incidents, and financial losses. The research aims to raise awareness of these procedures and reduce their associated risks. This study uses K-means, the Gaussian Mixture Model (GMM), and Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) to find the best algorithm for finding unusual landing procedures based on vertical speed and elevation angle. It also uses clustering techniques to look at the data that was collected. The evaluation using the silhouette score, Davies-Bouldin (DB) index, and the Calinski-Harabasz (CH) index reveals that the GMM is the most stable method for forming clusters. Analysis of the anomalies revealed that the vertical speed rule identified 100 % of the data as anomalies, whereas elevation-based anomalies accounted for only 0.8 % of the total data. Limitations of the study include a limited number of features and a brief two-month data collection period. Future research should incorporate additional features, extend the data collection period, and explore more algorithms to enhance the accuracy and robustness of the analysis.