Nicole Anne Vera Cruz@ncoveracruz

Imagine a hospital in a remote area of the Philippines. The router they rely on for internet access has been running for years and is starting to overheat. Normally, no one would notice until it stops working. In a country like the Philippines, frequent network outages, connectivity issues are common, often affected by high packet loss, latency, and poor signal strength due to inclement weather and high temperatures. These disruptions impact schools, businesses, and healthcare services. With 86.98 million internet users and an internet penetration rate of 73.6% as of January 2024 in the Philippines, the significance of reliable connectivity cannot be overstated (DataReportal, 2024). With CatBoost AI integration, our solution categorizes and analyzes environmental factors such as temperature, humidity, and precipitations’ effects on measures of connectivity like signal strength, packet loss, and latency, which is trained to predict the probability of an outage. Whilst existing sensors can detect potential issues, notifying the AI, which informs the facility administrator days in advance. This enables timely maintenance or replacement to prevent any disruption in critical network access. Business Value: OutageAI minimizes the risk of network downtime while optimizing operational and service efficiency. Predicting failures will help mitigate costs associated with outages since networks are an essential feature for industries heavily reliant on network connectivity for optimized distribution of information – aligned with UN SDG 9: Build resilient infrastructure, promote inclusive and sustainable industrialization, and foster innovation. From education to healthcare and businesses, where uninterrupted access is critical to support network connectivity, OutageAI provides a scalable solution through a Flutter web application triggered via Flask API calls of the risk outage model.