Dominant disaster detection and prediction in coastal areas using neural network system to optimize disaster management in coastal areas

The object of the study is detection and prediction dominant disaster in coastal areas. The problem being addressed is the lack of accurate and efficient early warning systems for these disasters, which can result in significant damage and economic loss. To solve this problem, this study develops an innovative application and website designed to predict the most dominant disasters in coastal areas. This system utilizes real-time data processing to provide early warnings and risk assessments, assisting communities and emergency response teams in preparing for potential threats. Testing results indicate that 89 % of the system’s predictions are effective in disaster management. The research methodology includes observation, data collection, dataset preprocessing, analysis, and the development of a smart detection system (SDS) using Geographic Information System (GIS)-based mapping and clustering techniques. The findings are explained through the hybrid deep neural network (DNN) method, which analyzes various environmental factors, including temperature, wind speed, wave height, weather disturbances, and sea level fluctuations. Additional features, such as daily weather forecasts, enhance the system’s predictive capabilities. This intelligent disaster management system, powered by a neural network, ensures effective disaster prediction and mitigation. The system is designed to be applied in coastal areas with limited technology, thereby improving disaster preparedness. Additionally, the application enables governments to monitor and respond to disasters more efficiently. By integrating artificial intelligence (AI)-based solutions, this research significantly contributes to disaster management, offering innovative strategies to minimize risks and enhance emergency response efforts