A Simulation-Based Framework for Leveraging Shared Autonomous Vehicles to Enhance Disaster Evacuations in Rural Regions with a Focus on Vulnerable Populations

2025-01-20

AI Summary

This study addresses accessibility and equity challenges faced by vulnerable rural populations—including persons with disabilities, older adults, and limited-English-proficient individuals—during disaster evacuation. We propose the first two-stage coordinated scheduling framework for shared autonomous vehicles (SAVs) specifically designed for rural contexts. Methodologically, the framework integrates integer linear programming with SUMO-based microscopic traffic simulation and innovatively incorporates dynamic road network resilience modeling—accounting for road closures and capacity degradation—to support both pre-disaster SAV pre-positioning and post-disaster real-time response. Evaluated in Sumter County, Florida, full SAV deployment (100% fleet coverage) reduced peak congestion by 37% and increased average vehicle speed by 22%, while significantly improving evacuation coverage and traffic flow stability. Results demonstrate the framework’s robustness and equity-preserving capabilities under infrastructure degradation, offering a scalable solution for resilient rural emergency mobility.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: AI for Accessibility

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT

Abstract

Rapid advancements in autonomous vehicles (AVs) are poised to revolutionize transportation and communities, including disaster evacuations, particularly through the deployment of Shared Autonomous Vehicles (SAVs). Despite the potential, the use of SAVs in rural disaster evacuations remains an underexplored area. To address this gap, this study proposes a simulation-based framework that integrates both mathematical programming and SUMO traffic simulation to deploy SAVs in pre- and post-disaster evacuations in rural areas. The framework prioritizes the needs of vulnerable groups, including individuals with disabilities, limited English proficiency, and elderly residents. Sumter County, Florida, serves as the case study due to its unique characteristics: a high concentration of vulnerable individuals and limited access to public transportation, making it one of the most transportation-insecure counties in the state. These conditions present significant challenges for evacuation planning in the region. To explore potential solutions, we conducted mass evacuation simulations by incorporating SAVs across seven scenarios. These scenarios represented varying SAV penetration levels, ranging from 20% to 100% of the vulnerable population, and were compared to a baseline scenario using only passenger cars. Additionally, we examined both pre-disaster and post-disaster conditions, accounting for infrastructure failures and road closures. According to the simulation results, higher SAV integration significantly improves traffic distribution and reduces congestion. Scenarios featuring more SAVs exhibited lower congestion peaks and more stable traffic flow. Conversely, mixed traffic environments demonstrate reduced average speeds attributable to interactions between SAVs and passenger cars, while exclusive use of SAVs results in higher speeds and more stable travel patterns.

Problem

• Simulation framework for rural disaster evacuations
• Prioritizing vulnerable populations with SAVs
• Improving traffic flow through SAV integration

Innovation

• Simulation-based SAV deployment framework
• Prioritizes vulnerable populations in evacuations
• Integrates SUMO and mathematical programming