Association Rules Mining with QAR Data: an Analysis on Unstable Approaches
Jiayi Xie, Huabo Sun, Yang Jiao, Binbin Lu · 2019 IEEE 1st International Conference on Civil Aviation Safety and Information Technology (ICCASIT) · 2019
Stable approach is vital for flight safety, and more attention should be paid in analyzing approaching phase of aircraft stages. This study aims to detect unstable approaches (UA) with the quick access recorder (QAR) data, and analyze association rules between parameters in approaching phase via exploratory data analysis (EDA) and association rule mining techniques. Results show that wind speed should be noticed when aircraft landing at Kunming Changshui airport in January, and high wind speed tend to cause roll, air speed and heading overrun. Moreover, pilots should pay close attention to roll when it exceeds warning value, and make adjustment in time to avoid pitch undergoing the warning value. These findings could be useful in preventing UA incidents and improving flight safety.