Data-Driven Urban Resilience: Computational Frameworks for Disaster Management and Infrastructure Planning 137 views

Urban areas worldwide are facing increasing challenges from natural disasters, climate change impacts, and the need for sustainable infrastructure development. Traditional optimization approaches often fall short in capturing the complex and dynamic behaviors inherent in urban systems—from evacuee movement patterns during emergencies to electric vehicle charging behaviors and flood propagation dynamics. This dissertation presents novel computational frameworks that leverage simulation-driven optimization to enhance urban resilience across multiple domains. The research integrates agent-based simulation, game-theoretic modeling, and physics-based simulations as foundational elements for optimization algorithms, enabling effective decision-making tools for urban planners and policymakers. The dissertation contributes multiple methodological advances through comprehensive case studies. First, it develops optimization frameworks for evacuation planning that account for traffic congestion. Second, it incorporates fairness considerations to ensure equitable treatment of evacuee groups. It also introduces game-theoretic approaches that model strategic evacuee behavior to find equilibrium solutions for large-scale evacuation systems. Third, it presents a simulation-optimization framework for efficient and effective expansion of electric vehicle charging infrastructure. Finally, it develops machine learning models trained on physics-based simulations to enable high-resolution coastal flood forecasting from sparse observations. Collectively, these frameworks demonstrate the power of simulation-driven optimization in creating more resilient, sustainable, and equitable urban systems.

PHD (Doctor of Philosophy)

Urban Resilience; Simulation-Optimization; Computational Modeling; Disaster Management; Infrastructure Planning; Spatiotemporal Forecasting; Agent-based Simulation; Mixed Integer Programming; Heuristic Search; Diffusion Models

English

2025-12-11