From analysing data on pilot and flight attributes, as well as weather conditions, a DSTA team identified factors to create a suite of algorithms with the aim of generating insights to boost flight safety.
The team integrated datasets across two systems and bridged different data formats. The Safety Information System records flight incidents and their probable causes, while the Flight Administration System records more comprehensive flight data, including important flight and pilot attributes. A combination of statistical tests was then employed to study the data.
A suite of machine learning algorithms was then adapted and combined into a predictive model. This has the potential to be developed further into a pre-flight screening tool for the RSAF to make flying safer.