Abstract—Recently, factory automation has been implemented using sensor networks. In general, the equipment deployed in automated factories is expensive. Due to the huge maintenance expenses associated with manufacturing plant equipment, there is a growing need for technology that can predict the lifespan of equipment consumables. Real-time fault prediction technology is essential because downtime in a process can led to substantial financial losses for a factory. Predictive Maintenance (PdM), which predicts replacement cycles instead of relying on Preventive Maintenance (PM) following equipment failure, can enhance productivity. Hence, this paper developed a predictive maintenance technology based on Industrial Internet of Things (IIoT). The developed platform can predict and verify the state of equipment in real time. To predict faults, we generated virtual voltage and frequency data for the inspection equipment of the Shift-by-wire Control Unit (SCU). We then applied this data to three models: the Recurrent Neural Network (RNN), the Long Short-Term Memory (LSTM), and the Gated Recurrent Unit (GRU), and compared their performance. Among them, the GRU model achieved the highest prediction speed and accuracy, with an R2-score of 0.992. We utilized this platform to develop a real-time AI prediction management system with the goal of improving productivity.
 
Keywords—Predictive Maintenance (PdM), Artificial Intelligence (AI), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), inspection equipment


Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.