Advanced Predictive Maintenance with Machine Learning Failure Estimation in Industrial Packaging Robots

In production systems, the repeated breakdowns of the operation have to be taken into account with great importance. The continuation of long malfunctioning states as well as the temporary interventions involve excessive time and money costs. Industry 4.0 technologies extensively use real-time Big Data collected from the machinery, and this enables potential problems to be addressed and resolved before they become an avalanche for the company. Permanent solutions can be produced, and thereby production efficiency can be established. In this paper, utilizing the Mean Time to Failure (MTTF) values and the past breakdown history of the robot system of the production line an Artificial Neural Network (ANN) model is established for system failure prediction. The proposed model successfully manages predictive maintenance of the machinery without the use of Internet of Things (IoT) technology.