In this open access paper, the AIRMES partner proposes a new application of data analysis using aircraft fault record data. It also deals with two types of data analysis techniques for extracting useful information contained in aircraft fault record data: pattern analysis and correlation analysis.

In the pattern analysis, the fault frequency of the aircraft is modeled by a probability density function (PMF) for each fault type, and then the similarity of each PMF is compared to identify the fault types with abnormal pattern. The correlation analysis is performed through a similarity comparison of time series data of occurrence patterns for each fault type.

The obtained results can be used to detect potential problems existing in the aircraft maintenance operation or potential problems existing inside the aircraft. Also, it can be applied as a high-level fault prognosis. Finally, in order to show the validity of the proposed method, illustrative examples of data analysis using real fault record data are provided.