Publication Details
Abstract
Floods remain one of the most destructive natural disasters, causing widespread damage to infrastructure, agriculture, and human life, particularly in flood-prone regions with limited early warning capabilities. This research presents FloodNexis, an intelligent multi-layer flood monitoring system designed to provide accurate, real-time flood prediction and early warning through the integration of environmental sensing, artificial intelligence, and cloud-based analytics. The proposed system combines multiple data sources, including rainfall intensity, river water levels, soil moisture, and weather forecast data, to create a comprehensive monitoring framework. FloodNexis employs machine learning algorithms trained on historical flood datasets from South and Southeast Asia to identify complex patterns and predict flood risks across three categories: river floods, urban flash floods, and coastal flooding. The system architecture consists of IoT-based sensor nodes, a cloud processing layer, AI prediction models, and a user alert interface. Real-time data collected from sensors is transmitted to the cloud, where the AI model processes the information and generates early warnings with risk classifications such as low, moderate, and high flood probability. The proposed solution offers a cost-effective, scalable, and intelligent flood monitoring approach that enhances disaster preparedness, reduces economic losses, and improves community resilience through timely alerts and data-driven decision-making.