<![CDATA[Flooding is a frequent natural disaster in Indonesia and has a significant impact on community mobility and daily activities. One of the main challenges in post-flood management is the limited availability of accurate and real-time information regarding water level conditions and flood recession time. This issue is particularly evident in flood-prone areas of Medan City. This study aims to design and develop an intelligent system based on the Internet of Things (IoT) and Artificial Intelligence (AI), implemented in a mobile application to support flood disaster mitigation. The proposed system utilizes an IoT-based barometric pressure sensor to acquire real-time water level data. The collected data are processed using a Long Short-Term Memory (LSTM) model to predict flood recession time. In addition, the application provides safe alternative route recommendations through Google Maps integration. The results indicate that the system is capable of delivering real-time water level information, producing flood recession time predictions with a coefficient of determination (R2) of 0,766, which demonstrates a good level of accuracy, and recommending safe routes to support community mobility during flood events. This system demonstrates strong potential for further development as a digital decision-support tool for flood disaster mitigation.]]>
Copyrights © 2026
