Kaprakattu, Arun Raj (2025) AI-based preventive maintenance system for network infrastructure: Implementation and performance analysis. World Journal of Advanced Research and Reviews, 26 (1). pp. 3817-3824. ISSN 2581-9615
![WJARR-2025-1496.pdf [thumbnail of WJARR-2025-1496.pdf]](https://eprint.scholarsrepository.com/style/images/fileicons/text.png)
WJARR-2025-1496.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial Share Alike.
Abstract
This article details an artificial intelligence-powered preventive maintenance system designed specifically for networking devices. As network infrastructure grows increasingly complex, traditional reactive maintenance approaches have proven inadequate for ensuring optimal performance and reliability. The system leverages advanced telemetry collection frameworks, machine learning algorithms, and predictive analytics to detect potential failures before they impact service quality. Through continuous monitoring of core system metrics, interface traffic data, and network-specific parameters, the system can identify anomalous patterns, forecast component degradation, and recommend appropriate remediation actions. The implementation methodology encompasses comprehensive data collection, baseline establishment, model development, and training phases. Alert classification mechanisms prioritize issues based on severity while automated response capabilities translate analytical insights into actionable maintenance strategies. Performance metrics demonstrate significant improvements in network availability, maintenance efficiency, and operational costs compared to traditional approaches, highlighting how AI-driven preventive maintenance is transforming network operations.
Item Type: | Article |
---|---|
Official URL: | https://doi.org/10.30574/wjarr.2025.26.1.1496 |
Uncontrolled Keywords: | Artificial Intelligence; Preventive Maintenance; Network Telemetry; Anomaly Detection; Predictive Analytics |
Depositing User: | Editor WJARR |
Date Deposited: | 27 Jul 2025 15:00 |
Related URLs: | |
URI: | https://eprint.scholarsrepository.com/id/eprint/2312 |