A data-driven GIS-integrated platform-agnostic system improved the latency of computational imaging of displaced persons during disasters

Abstract

Disaster-induced displacement remains a significant global challenge, requiring real-time tracking, efficient resource allocation, and coordinated emergency response. Traditional disaster management systems often suffer from fragmented information, high latency, and platform dependency, limiting their effectiveness in large-scale crises. This study introduces a GIS-integrated, platform-agnostic system designed for latency-optimized computational imaging of displaced persons, enhancing response efficiency and data accuracy. By leveraging geospatial analytics, artificial intelligence, and high-frequency data processing, the system enables real-time visualization and tracking of affected populations across multiple devices and infrastructures. Unlike existing disaster platforms, the proposed system emphasizes low-latency computational imaging and bidirectional interaction between displaced persons and LGUs. A mobile-based reporting tool allows evacuees and responders to transmit location-based data, while a web-based dashboard provides government agencies with dynamic mapping, predictive analytics, and real-time situational awareness. System evaluation based on ISO/IEC 25010 standards yielded the following scores: Functionality (81.6% for displaced persons, 97.3% for local government), Reliability (80.8%, 93.3%), Suitability (81.4%, 96.0%), and Performance Efficiency (79.9%, 95.0%), with an overall average of 80.9% for displaced persons and 95.4% for local government users. In conclusion, the proposed system establishes a scalable and adaptive framework for disaster response and humanitarian operations, addressing challenges in volcanic eruptions, earthquakes, hurricanes, and other large-scale crises. The results demonstrate high system usability and effectiveness, ensuring seamless cross-platform interoperability and improving emergency response efforts through real-time, data-driven decision-making.