Predictive Maintenance for Aircraft Engine Using Machine Learning: Trends and Challenges

This article aims to prove that Machine Learning (ML) methods are effective for Predictive Maintenance (PdM) and to obtain other developing methods that suitable applied on PdM, especially for aircraft engine, and potential method that can apply on future research, and also compared between articles in International and Indonesia institution. Maintenance factors are important to prognostic the states of a machine. PdM is one of the factor strategies based on realtime data to diagnosis a failure of the machine through forecasting remaining useful life (RUL), especially on aircraft machine where the safety is priority due to enormous cost and human life. ML is the technique that accurately prediction through the data. Applied ML on PdM is the huge contribution for saving cost and human life guarantee of safety. This work provides the literature survey for recent research which trends and challenges on PdM of aircraft engine using ML that compared the research from international and Indonesia from 2016 to 2021. Result of this work shows that ML method, Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) are the best method to calculate PdM with more than 99% on rate accuracy, and low level of Indonesia institution research which focused on PdM on aircraft engine using ML