The risk of aircraft runway excursion, dependent on multiple factors, is related to operating conditions. The purpose of this paper is to identify the correspondence between features belonging to different aspects that occur in runway excursion events, distinguishing between take-off and landing phases.
To define the correspondence between the characteristic features of runway excursions, this study has applied multiple correspondence analysis (MCA). MCA is used to represent and model data sets as “clouds” of points in a multi-dimensional Euclidean space. There are five variables used in MCA: geographical region, potential cause, aircraft class, flight nature and aircraft damages. For the purpose of this research, the database contains only runway excursion accidents that took place between 2006 and 2016 among all categories of aircraft in all world regions. The events contained in the database were analyzed by separating those that occurred during take-off and those that occurred during landing.
With this method, this study identified a few particularly interesting variable combinations. Generally, the consequence of an aircraft runway excursion is substantial aircraft damage. Also, the most common cause of runway excursion during take-off is aircraft system faults, while during landing, it is weather conditions. Furthermore, the destruction of an aircraft is a result of a runway excursion due to bad weather conditions, both during take-off and landing.
The results of this study can be used by a broad range of civil aviation organizations for runway risk assessment and to select the most effective safety countermeasures for runway excursions.
The authors believe this study is original, especially for the statistical analysis method used.
