On the use of data-mining algorithms to improve FOQA tools for airlines

In this paper, we investigate potential contributions of clustering algorithms for flight safety management within airlines. Based on independent analysis of a set of commercial flights with different methodologies, we identify the strengths of each approach. Then, the paper discusses the design of an analysis process using a flight decomposition relying on crew activities and including discrete parameters that reflect crew actions. The results lay the foundations for a better use of such data-mining algorithms in the next generation of Flight Data Monitoring tools used by airlines.