The NSF-JST SHIELD research examines the physical and human parameters that influence disaster outcomes in the US and Japan, intending to enhance disaster resilience, particularly for vulnerable populations such as older adults. Currently developing ML-driven mobility modeling approaches that integrate geospatial, temporal, and behavioral data to enhance decision-making and operational planning in disaster scenarios.
Supervisor: Prof. Nalini Venkatasubramanian
References
2026
-
PARA-FLOW: An Event-Based Framework for Modeling Accessible Multi-Modal Urban Mobility
Fernanda Ventorim; Andrew Chio; Ruben Vescovo; Erick Mas; Shunichi Koshimura; Nalini Venkatasubramanian
Proceedings of the 34th ACM International Conference on Advances in Geographic Information Systems (SIGSPATIAL 2026), 2026
Urban mobility models support transportation planning, accessibility assessment, and emergency preparedness, yet many assume homogeneous agents and unconstrained routing. We present PARAFLOW, an event-based framework for accessible multi-modal urban mobility under routine and disrupted conditions. PARA-FLOW separates behavioral intent from trajectory generation through two stages: Behavioral Event Generation (BEG), which generates daily mobility schedules for heterogeneous agents based on demographic attributes, disability profiles, and contextual conditions, and Network Trajectory Generation (NTG), which realizes accessibility- and topology-constrained trajectories. Evaluation using census-derived data for approximately 307,000 people with disabilities in Orange County, California, shows that PARA-FLOW generates topologically feasible routes consistent with agents’ transport preferences and disability profiles. Under infrastructure degradation, it quantifies accessibility losses, with average shelter travel times increasing from 25.1 to 92.8 minutes and 157 individuals losing shelter access. Together, these results demonstrate scalable, accessibility-constrained urban mobility modeling.
2025
-
RADAR: Resource Allocation for Disaster Resilience in Senior Health Care
Modeste Mefenya Kenne; Fernanda Ventorim; Chad Cossey; ZhengHui Hu; Julie Rousseau; Nalini Venkatasubramanian
Proceedings of the 33rd ACM International Conference on Advances in Geographic Information Systems (SIGSPATIAL 2025), 2025
During disasters, ensuring that critical response resources are efficiently allocated to the most appropriate locations is crucial for minimizing adverse impacts and saving lives. To this end, we present RADAR, a data-driven platform that integrates multisource GIS feeds (e.g. USGS earthquake alerts, Cal Fire wildfire perimeters) with facility and transportation data to support proactive planning and real-time recommendations that can be used by Emergency Operations Centers to guide populations to safety. RADAR uses policy-driven stable matching to optimize routing and resource assignment for evacuation planning and resource delivery. The aggregate model allocates populations in impacted facilities to alternate short-term facilities (e.g., hospitals), and a fine-grained extension for long-term senior-care facilities personalizes allocation using resident preferences, medical profiles, and social constraints. RADAR adapts as conditions evolve by utilizing historical data, live traffic, and changing facility status. We validated RADAR’s efficacy in several disaster settings, including real events such as the Palisades wildfire and tabletop drills (earthquake and water-contamination scenarios) involving first responders.