Enhanced digital twin development for a conveyor belt system: integrating PLC control, CNN decision-making, Gradio-based HMI, and fuzzy logic

Abstract

This paper introduces an advanced Digital Twin (DT) of a conveyor belt system, designed to enhance industrial automation and decision-making in the context of Industry 4.0. Built using Factory I/O, the DT simulates the physical conveyor's behavior, integrating seamlessly with a Programmable Logic Controller (PLC) for real-time control. A Convolutional Neural Network (CNN) enhances decision-making by analyzing conveyor activity, such as defect detection and sorting optimization. A Fuzzy Logic (FL) system also refines CNN outputs, improving reliability by incorporating factors like confidence scores and operational parameters. A Gradio-based Human-Machine Interface (HMI) offers an intuitive platform for real-time monitoring, interaction, and manual control. The system demonstrates robust synchronization between virtual and physical components, showcasing its ability to improve efficiency, accuracy, and adaptability in industrial processes. This work contributes to smart manufacturing advancements by combining  DT technology, intelligent algorithms, and adaptive control for enhanced industrial supervision.