Spatio-temporal analysis of unstable approach incidents with QAR Big Data

A stable approach is vital for flight safety, while unstable approach (UA) is one of the main causes of flight risks. This study aims to analyze the spatio-temporal patterns with QAR big data via exploratory data analysis (EDA) technologies. In this paper, the spatio-temporal distribution characteristics of the UA incidents are preliminarily explored via time series thermodynamic diagram, spatial distribution map and spacetime cube. In addition, Pearson correlation coefficient and its geographically weighted version are used to explore the correlations between the UA incidents and factors, including elevation, wind grade and bad weather. Results show that the spatial distribution of UA incidents of different brands is quite different, and there are apparent correlations between the UA incidents and altitude, wind grade and bad weather, while local correlations appear to vary spatially across China.