Robotic cell reliability optimization based on the technology of digital twin

Thumbnail Image
Τσουμπού, Σοφία
Journal Title
Journal ISSN
Volume Title
In recent years, the traditional manufacturing industry has been affected by the development of digital technologies and Industry 4.0 wave. Due to the rising digitalization in every step of the manufacturing, manufacturing companies have now the chance to reach a completely new level of productivity. Robotic systems have become a standard tool in modern manufacturing, due to their unique characteristics, such as repeatability, precision, speed, high payload, and ability to operate in dangerous areas. However, robotics manipulators suffer from low reliability. Low reliability increases the probability of disruption in the manufacturing processes, delays are induced in the manufacturing/production systems’ short-term schedule, faults and failures are arising, minimizing in this way the productivity and by extension the profit. This research work presents a methodology for the improvement of reliability in a robotic cell based on the technologies and tools that offered by Industry 4.0 and particularly by the Digital Twin (DT) concept as well as the Machine Learning (ML) techniques under the framework of Predictive Maintenance (PdM). The reliability is highly associated with the maintenance processes. As a result, by implementing maintenance at the optimum time, the reliability of the robotic cell can be improved. Firstly, our aim is focused on the DT concept in order to monitor and control the health status of our system. Secondly the scope is to identify the critical components of a robotic cell, detect and classify the faulty behavior of the critical component and propose the framework of the prediction of its Remaining Useful Life (RUL) in order to improve the reliability of the whole robotic cell. Then, based on the diagnostic and prognostic results, engineers can apply the appropriate maintenance tasks to prevent their robotic cell from serious failures and ensure high performance of their system.
Reliability optimization, Robotic cell, Industry 4.0, Predictive maintenance, Digital twin, Remaining useful life, Machine learning