Data-driven Disaster Response Systems Dependent on Time of Day, Season and Location for Megacities

By 2050, 70% of the world's population will live in urban areas. As cities continue to grow, disaster risk is expected to increase exponentially. This proposed research aims to design a replicable and transferrable information-sharing system for disaster preparedness and response in megacities. Such a system would integrate understanding of natural hazards as well as human behavior in order to inform immediate response in the face of rapid-onset disasters, such as floods, earthquakes, and fires. Furthermore, this research will aid critical decisions regarding long-term urban resilience. Case studies selected for this project are Tokyo, Taipei, and New York City due to their complex urban environments and diverse populations, which present a challenge for disaster management. Traditionally, disaster planning has relied limited analysis regarding possible disaster scenarios. Notably, past planning efforts often do not distinguish between event time of day, workdays vs. weekends, seasons, or urban locations (e.g. indoor, outdoor, underground). Furthermore, traditional approaches have failed to capture the diverse needs of the affected social groups. For example, early warning systems near Tokyo and Taipei Main Stations have not been provided in foreign languages or a friendly manner for the disabled and elderly. More broadly, limited scenario analysis has serious implications for urban resilience. Current infrastructure systems tend not to be designed to withstand mega-disasters. For instance, a capacity-exceeding extreme flood (larger than a 200-year flood) has not been adequately considered in Taipei's flood management plan. We propose to address dynamic disaster scenarios as well as the needs of vulnerable socio-economic groups in urban areas. To do so, we will integrate both scenario-based and data-driven planning, while incorporating sensor data and stakeholder engagement. We will take a mixed-methods approach while synthesizing data from a wide variety of sources, such as from smartphones and in-depth interviews. This research comprises three key components: 1) data collection, analysis, and simulation of hazards and human responses, 2) design of information-sharing systems and emergency response, and 3) policy recommendations for urban risk governance towards resilience by developing a framework to collaborate with stakeholders and analyzing vulnerability across socio-economic groups.