Edge Computing-Enabled Predictive Maintenance System with Multi-Protocol RTU Integration for Industry 4.0 Applications
JOIV : International Journal on Informatics Visualization · 30.11.2025 · 9 (6) · 2758
DOI: 10.62527/joiv.9.6.4976 ↗Kurz gefasst
Studienprotokoll in der Zeitschrift JOIV : International Journal on Informatics Visualization, online erschienen am 30.11.2025. Verfasst von Hyunwook Kim und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
This paper presents a novel modular Remote Terminal Unit (RTU)-based edge computing architecture for industrial predictive maintenance, addressing critical limitations of cloud-centric approaches in Industry 4.0 environments. The proposed Edge-Fog-Cloud three-tier architecture integrates heterogeneous industrial protocols (RS-485, Modbus RTU/TCP, OPC UA, MQTT) within a unified firmware framework while enabling real-time processing. The system implements a progressive AI model development strategy across three stages: unsupervised anomaly detection using Isolation Forest, pattern recognition with 1D-CNNs, and temporal dependency modelling with bidirectional LSTM networks and attention mechanisms. Experimental validation in industrial testbeds demonstrated significant performance improvements: 78% reduction in response latency (from 312ms to 67ms), 46% decrease in network traffic, and optimal F1-scores of 0.974 for anomaly detection. The system successfully predicted 83% of equipment failures with 30–90-minute advance warning, achieving 31.5% reduction in Mean Time To Repair (MTTR) and 8.3% increase in Overall Equipment Effectiveness (OEE). Economic analysis revealed 192% Return on Investment within the first operational year, with annual cost savings exceeding $214 million through preventive maintenance optimization and downtime reduction. Third-party certification validated 99.43% AI model accuracy and 98.6% communication reliability under industrial conditions. The modular architecture enables incremental deployment, resulting in 35-40% lower implementation costs than commercial solutions, providing a practical framework for Industry 4.0 transformation while maintaining operational continuity and cost-effectiveness.
Zitieren
APA
Kim, H., & Sim, H. (2025). Edge Computing-Enabled Predictive Maintenance System with Multi-Protocol RTU Integration for Industry 4.0 Applications. JOIV : International Journal on Informatics Visualization, 9(6), 2758. https://doi.org/10.62527/joiv.9.6.4976
Vancouver
Kim H, Sim H. Edge Computing-Enabled Predictive Maintenance System with Multi-Protocol RTU Integration for Industry 4.0 Applications. JOIV : International Journal on Informatics Visualization. 2025;9(6):2758. doi:10.62527/joiv.9.6.4976
BibTeX
@article{Kim2025,
author = {Kim, Hyunwook and Sim, Hyun},
title = {Edge Computing-Enabled Predictive Maintenance System with Multi-Protocol RTU Integration for Industry 4.0 Applications},
journal = {JOIV : International Journal on Informatics Visualization},
year = {2025},
volume = {9},
number = {6},
pages = {2758},
doi = {10.62527/joiv.9.6.4976}
}
EUVIMED