A Review of Aircraft Predictive Maintenance Techniques Via Machine Learning

The use of industrial predictive maintenance has grown with the increased availability of data from sensors in industrial equipment. In the aircraft industry, predictive maintenance has become crucial for optimizing maintenance schedules, reducing aircraft downtime, and identifying unforeseen problems. However, there are few assessments of predictive maintenance applications dedicated solely to the aircraft operation business. To address this gap, a systematic review of publications between 2018 and 2023 in five digital libraries (Elsevier, IEEE, MDPI, Springer, and Wiley) was conducted. Initially, 8,512 documents were found, with 51 remaining following exclusion criteria. This review paper provides a comprehensive state-of-the-art systematic literature assessment of various data types, applications, techniques, initiatives, and potential for predictive maintenance in the aircraft industry. The study focused on researches related to remaining useful life (RUL), failure detection, and equipment condition assessment. The goal of this study is to identify and highlight the issues and opportunities for future research on turbofans, actuators, auxiliary power units (APUs), and transformer rectifier units (TRUs) in aircraft. This work contributes to the field by providing an in-depth assessment of AI applications in turbofans, actuators, APU, and TRU, as well as discussing the challenges and future potential in this area.