Telecommunications
AI-Driven Predictive Maintenance for Telecom Infrastructure
Created a centralized system to monitor 10,000+ network nodes, predicting hardware failure with high accuracy to reduce unplanned downtime.
Completed: 2/28/2025
🎯 Project Scope
Development of data ingestion pipelines (sensor data), time-series database design, and a web dashboard for field engineer alerting and dispatch.
⛰️ The Challenge
Reactive maintenance was costly and often resulted in service blackouts, impacting millions of customers and violating SLAs.
🛠️ The Hanva Solution
Implemented a time-series forecasting model (LSTM) on streaming sensor data, triggering automated alerts to maintenance teams hours before actual component failure.
Key Technologies & Architecture
- • InfluxDB
- • Grafana
- • Python
- • Azure IoT Hub
- • Scala