Improved Ensemble-Learning Algorithm for Predictive Maintenance in the Manufacturing Process

Industrial Internet of Things (IIoT) technologies comprise sensors, devices, networks, and applications from the edge to cloud. Recent advances in data communication application using IIoT have streamlined predictive maintenance (PdM) for equipment quality management manufacturing processes. PdM is useful fields such as device, facility, total management. based on cloud or computing has revolutionized smart To address problems, herein, we develop a new calculation method that improves ensemble-learning algorithms with adaptive learning make boosted decision tree more intelligent. The algorithm predicts main issues, product failure unqualified equipment, advance, thus improving machine-learning performance. Herein, semiconductor blister packing machine are used separately analytics. former help predicting yield process. predict packaging quality. Experimental results indicate proposed accurate, an area under receiver operating characteristic curve exceeding 96%. Thus, provides practical approach PDM processes machines.