Safna et al. (2026) Intelligent Flood Prediction and Early Warning System Using Machine Learning Models

Identification

• Journal: INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT
• Year: 2026
• Date: 2026-02-06
• Authors: M Safna, MS.Surabhi K.S
• DOI: 10.55041/ijsrem56391

Research Groups

Department of Computer Applications, Nehru College of Management, Coimbatore, Tamil Nadu, India.

Short Summary

This paper develops an intelligent flood prediction and early warning system using various machine learning models to analyze historical and real-time environmental data. The system aims to improve prediction accuracy and provide timely alerts for effective disaster management and mitigation.

Objective

• To develop a robust and scalable flood prediction and early warning system using machine learning algorithms that analyzes historical and real-time environmental data (rainfall, water level, temperature, humidity, wind speed, soil moisture) to improve forecasting accuracy and provide timely alerts for disaster preparedness.

Study Configuration

• Spatial Scale: General, designed to be scalable and adaptable to different geographical regions by retraining with regional datasets.
• Temporal Scale: Utilizes historical and real-time environmental data for continuous monitoring and dynamic prediction updates, enabling early warnings for impending flood conditions.

Methodology and Data

• Models used: Decision Tree, Random Forest, Support Vector Machine (SVM), Artificial Neural Network (ANN).
• Data sources: Historical and real-time environmental data collected from meteorological departments, river monitoring stations, online weather databases, IoT sensors, and government databases. Parameters include rainfall, river water level, temperature, humidity, wind speed, and soil moisture.

Main Results

• The Random Forest algorithm achieved the highest flood prediction accuracy of approximately 94%, outperforming ANN (92%), SVM (89%), and Decision Tree (86%).
• The proposed system demonstrated a strong capability in identifying high-risk flood conditions, with a high true positive rate indicating correct prediction of most flood events.
• Real-time data integration significantly improved system performance by enabling continuous monitoring and dynamic prediction updates.
• The automated alert generation module responded promptly to high-risk conditions, ensuring timely notification to users and authorities.

Contributions

• Development of an intelligent and reliable flood prediction system that integrates efficient machine learning algorithms with real-time data processing for enhanced prediction accuracy and reliability.
• Provides a scalable and adaptable solution for flood forecasting, offering better performance compared to traditional methods.
• Establishes an automated early warning system with real-time data integration and alert mechanisms, significantly improving disaster preparedness and response.
• Contributes to sustainable disaster management and public safety by minimizing flood-related losses and improving early warning capabilities.

Funding

No explicit funding information was provided in the paper.

Citation

@article{Safna2026Intelligent,
  author = {Safna, M and K.S, MS.Surabhi},
  title = {Intelligent Flood Prediction and Early Warning System Using Machine Learning Models},
  journal = {INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT},
  year = {2026},
  doi = {10.55041/ijsrem56391},
  url = {https://doi.org/10.55041/ijsrem56391}
}