A Machine Learning–Based Predictive Modeling for Optimizing the Critical Useful Life of Aeroengines: A Field Maintenance Perspective

Establishing critical useful life plays a central role to determine aeroengine health status including aeroengine parameter changes from adverse material conditions or metal fatigue. The useful life assessment serves to support maintenance teams by enabling predictive maintenance followed by part replacement or conditions improvement. The proposed research works to improve the ability of turbofan aeroengine useful life estimation while targeting practical deployment during maintenance operations at field locations. A field maintenance–oriented ensemble bagged regression model for aeroengines represents the proposed method within this research. The present study reaches an error index of 7.06 with 98.95% model fitness when applying it to critical useful life training data. The projected model received its validation through experiments on test and field datasets. Field tests revealed that among 25 machine learning models the proposed model delivered optimal results since its error index was determined at 10.5337 with 97.60% accuracy compared to prior research findings. This study delivers an optimal solution, which enables aviation maintenance crew and techno managers to achieve effective critical useful life evaluation and decision-making for maintenance needs. This research provides essential guidance to industries under maintenance and repair operations for the reassessment of field-based critical parameters identification.