IMPROVING QUALITY OF PREDICTIVE MAINTENANCE THROUGH MACHINE LEARNING ALGORITHMS IN INDUSTRY 4.0 ENVIRONMENT

Smart manufacturing is the modern form of manufacturing that utilizes Industry 4.0 enablers for decision making and resources planning by taking advantage of the available data. With the advancement of digitalization and industrial machine connectivity, it is now feasible to gather data in real-time from a variety of sensors (e.g. current, acoustic, vibration etc.) while the process is being carried out. The aim of the paper is to propose a framework for predictive maintenance PdM 4.0 and validate the framework by implementing it for a manufacturing process, milling in which a public data set from NASA repository is used to build and test the proposed PdM 4.0 system. The various machine learning classifiers such as: support vector regression SVR, RF, DT, XGBoost and MLP regressor have been used for remaining useful life and tool wear rate prediction. The model evaluation and comparison is based on metrics like (R- square), root mean square error and mean absolute error.