Scholarlink

This research develops an IoT-driven predictive maintenance model using the XGBoost algorithm for accurate failure classification. The model incorporates Stratified K-Fold cross-validation, data preprocessing with RobustScaler, and fine-tuned parameters to enhance performance. Achieving 97.32% accuracy, the model outperforms traditional methods like SVM, KNN, and Decision Tree. The results highlight the effectiveness of integrating IoT and machine learning in predictive maintenance, offering a reliable solution to reduce downtime, optimize maintenance schedules, and improve industrial operations.

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

An IoT-Driven Machine Learning Model for Predictive Maintenance Classification in Industrial Systems

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • U. Srilakshmi,
  • J. Manikandan,
  • Dinesh Valluru,
  • Amerendra Reddy Panyala,
  • Baddepaka Prasad,
  • Mireyala Nagavamsi

摘要

This research develops an IoT-driven predictive maintenance model using the XGBoost algorithm for accurate failure classification. The model incorporates Stratified K-Fold cross-validation, data preprocessing with RobustScaler, and fine-tuned parameters to enhance performance. Achieving 97.32% accuracy, the model outperforms traditional methods like SVM, KNN, and Decision Tree. The results highlight the effectiveness of integrating IoT and machine learning in predictive maintenance, offering a reliable solution to reduce downtime, optimize maintenance schedules, and improve industrial operations.