Citation

Meng L, Sasha Dong Z, Liu D. Comput. Ind. Eng. 2023; 184: e109565.

Copyright

(Copyright © 2023, Elsevier Publishing)

DOI

10.1016/j.cie.2023.109565

PMID

unavailable

Abstract

Evacuation models based on public transportation have been shown to increase network capacity for transportation systems while improving community and societal disaster preparedness. However, due to a lack of private vehicles and inconvenient mobility, it is often difficult for vulnerable groups to arrive at shelters on time. To address this issue, this research proposes an evacuation strategy that incorporates the ridesharing concept, allowing individuals with vehicles to provide rides for carless groups. Three mixed-integer programming models are developed based on assumptions such as different capacities and pick-up principles, with the goal of maximizing the number of evacuees transported to assembly points or shelters in a limited amount of time. To evaluate the effectiveness of the proposed ridesharing evacuation models, a real-world case study in Houston is conducted. Numerical analyses are performed with five factors: evacuation scales, data generation, evacuation models, evacuation clearance times, and the number of vehicles involved in the evacuation process. The findings demonstrate that ridesharing evacuation models can provide viable alternative evacuation options to carless and public transit-dependent populations. Furthermore, the study reveals that increasing the number of vehicles to assist vulnerable groups may not be necessary in cities with high population density due to excessive traffic volume, which can hamper disaster response implementation. Additionally, the number of evacuees arriving at shelters or assembly points is unbalanced due to space-time constraints. To address this issue, relief supplies should be distributed on demand in these areas to reduce waste. The findings of this study can inform the development of more effective and efficient evacuation strategies that can better serve communities and vulnerable populations during times of crisis.

Language: en

Keywords

Disaster response; Evacuation planning; Humanitarian relief; Mixed-integer programming; Ridesharing