Scholarlink

With the adoption of the concept of Industry 4.0, the machinery used in large industries is getting more intelligent and complex. Therefore, designing an accurate and reliable fault detection and prediction system remains crucial. Poorly maintained machines can have consequences like replacement of components, severe accidents, etc., leading to downtime and business losses. It suggests that a proper maintenance-related decision-making system is crucial for monitoring the health of machines. The paper aims to develop an open-source predictive maintenance toolkit (called PdMTKT) for fault analysis and prediction. The tool follows a modular architecture with separate modules for data analysis, visualization, processing, and training predictive models. All the modules in PdMTKT have user-friendly graphical user interfaces (GUI) and are currently under active development. The toolkit has been evaluated using the engine health data from the CMAPSS database as a test case, where PdMTKT can perform trend analysis, forecasting and prediction of the engines’ remaining useful life (RUL). Upon full development, PdMTKT aims to be a general-purpose, open-source tool for the predictive maintenance activity of different machines.

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

A Machine Learning Based Open-Source Tool for Predictive Maintenance

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Sanket Krushna Patil,
  • Sibasis Sahoo,
  • Deepak Sharma,
  • Ashish Anand

摘要

With the adoption of the concept of Industry 4.0, the machinery used in large industries is getting more intelligent and complex. Therefore, designing an accurate and reliable fault detection and prediction system remains crucial. Poorly maintained machines can have consequences like replacement of components, severe accidents, etc., leading to downtime and business losses. It suggests that a proper maintenance-related decision-making system is crucial for monitoring the health of machines. The paper aims to develop an open-source predictive maintenance toolkit (called PdMTKT) for fault analysis and prediction. The tool follows a modular architecture with separate modules for data analysis, visualization, processing, and training predictive models. All the modules in PdMTKT have user-friendly graphical user interfaces (GUI) and are currently under active development. The toolkit has been evaluated using the engine health data from the CMAPSS database as a test case, where PdMTKT can perform trend analysis, forecasting and prediction of the engines’ remaining useful life (RUL). Upon full development, PdMTKT aims to be a general-purpose, open-source tool for the predictive maintenance activity of different machines.