Las Piñas Flood Monitoring System with Alternate Route Using Bayesian Network via Mobile Application

Abstract

Flood inundation time has increased significantly due to climate change, resulting in temporary loss of mobility by the public and during disaster management. The Las Piñas Disaster Risk Reduction and Management Office (LPDRRMO) operation relies on real-time situation updates during calamities. Barangay responders provide updates on the flood situation in the city and are at risk of hazards since manual measurements are being used. According to the National Disaster Risk Reduction and Management Plan, technological advancement in disaster mitigation should be developed and incorporated during disaster operations. This work applied the internet-of-things (IoT) to create timely updates for weather parameters across the city. Float switches and ultrasonic sensors determine the flood heights, while temperature and humidity sensors measure the atmospheric conditions. Microcontrollers in flood stations process the data, transmit and receive data via short message services (SMS) with an average of 6 seconds refresh rate in a mobile application, depending on the signal strength of sites. The operations are analyzed through a Jetson Nano server. A Bayesian network analysis classifier trained and tested data from historical data provided by the LPDRRMO generating an algorithm with 94.87% accuracy. Then, Dijkstra's shortest path process is employed to reroute the traffic incorporating the "Friendship Route" – an interconnected road network of Las Piñas City across various villages and subdivisions to ease the traffic along the major thoroughfares. The mobile app is made available in cross-platform for Android and iOS operating systems using React.js and React Native, and is named Electronic Flood Warning and Alternative Route System (e-WAS).