Uncovering Hidden Patterns in Flight Safety Data Through Statistical Analysis

Abstract

This research aims to identify trends in data that indicate a potential increase in the risk of aviation accidents. By analyzing aviation incident historical data, we look at the relationship between incidents on the ground and those in flight, as well as minor incidents and major incidents within an Army aviation unit. Machine learning algorithms applied to data sets involving crew experience and mishap reports may determine whether indications exist that forecast a higher risk of an aviation incident for an aviation organization within the U.S. Army. Achieving zero preventable mishaps regarding aviation operations requires a proactive approach to hazard identification and risk management, which is explored in the methods of this project. The results from the analyzed data determine if any consistencies exist in the conditions within an aviation unit leading up to recordable mishaps.