Scholarlink

This study focuses on enhancing predictive maintenance (PdM) in aviation through deep learning models with an emphasis on machine learning explainability. By analyzing sensor data, the research develops models to predict engine failures and optimize maintenance schedules. The tuned Long Short-Term Memory (LSTM) model is identified as the best performer, offering superior accuracy and scalability for handling sequential data. In further model developing work, the researcher conducts comparative analysis of different deep learning algorithms for failure prediction of aircraft turbofan engines. Explainability techniques such as SHAP are integrated to provide insights into model predictions, enabling transparent decision-making. The results demonstrate the potential for deep learning to revolutionize aviation PdM by reducing downtime, enhancing safety, and improving operational efficiency.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning Models for Predictive Maintenance in Aviation with Explainable AI Using SHAP

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Justin Gilbert,
  • Chin Min Hwei

摘要

This study focuses on enhancing predictive maintenance (PdM) in aviation through deep learning models with an emphasis on machine learning explainability. By analyzing sensor data, the research develops models to predict engine failures and optimize maintenance schedules. The tuned Long Short-Term Memory (LSTM) model is identified as the best performer, offering superior accuracy and scalability for handling sequential data. In further model developing work, the researcher conducts comparative analysis of different deep learning algorithms for failure prediction of aircraft turbofan engines. Explainability techniques such as SHAP are integrated to provide insights into model predictions, enabling transparent decision-making. The results demonstrate the potential for deep learning to revolutionize aviation PdM by reducing downtime, enhancing safety, and improving operational efficiency.