AI Based Bootstrapping Method for Uncertainty Management in RUL Estimation of Aircraft Turbofan Engines

To minimize downtime, decrease the expenses related to unplanned equipment breakdowns, and optimize maintenance schedules, Predictive maintenance is increasingly adopted in modern industries to reduce equipment failure and maintenance costs. A key challenge in this domain involves predicting the remaining lifespan of complex systems using incomplete and noisy sensor information. Conventional methods often depend on single-value predictions from regression techniques, which typically fail to represent the uncertainty inherent in forecasts. Being unable to quantify uncertainty may result in overconfident choices that risk operational safety and effectiveness. To address these problems, our study proposes an integrated framework that combines bootstrapping methods with several regression models with several regression techniques, such as Random Forest, Linear Regression, Neural Networks, and Support Vector Regression. Our method computes confidence intervals on error measures like MAE and RMSE providing a more transparent view of forecast dependability. This methodology facilitates accurate model comparisons and enhances decision making by clearly measuring hazards. We aim to bridge the gap between the requirement for risk analysis in predictive maintenance and traditional RUL estimation techniques. The proposed strategy improves decision-making by providing reliable evaluation metrics and contributes to ongoing progress in the field of health management and prognostics for essential systems.