An Agent-Based Model of Vertical Tsunami Evacuation Behavior and Shelter Locations: A Multi-Criteria Decision-Making Problem

Vertical evacuation is an alternative protective action to horizontal movement. The effectiveness of vertical evacuation shelters to save lives depends critically on whether people know where they are, believe they will be effective in protecting themselves and their families, and therefore choose to evacuate to them. Compared to the traditional vehicle-based horizontal evacuation, the authors know little about mobilizing for vertical evacuation. The objective of this paper is to present an agent-based modeling (ABM) framework to evaluate vertical evacuation behavior and shelter locations for a near-field tsunami hazard. Three criteria were identified to assess the effectiveness of the vertical evacuation including vertical evacuation behavior, tsunami inundation force, and the expected mortality rate. The results reveal that (1) the choice of vertical shelter locations will directly impact the proportion of the people who evacuate vertically; (2) lower percentage of people evacuate to the shelter if it is far from the population center; (3) the location of shelter significantly impacts the total and categorical mortality rates (e.g., horizontally or vertically evacuated); (4) improvements in evacuees' mobility, such as faster walking speed or shorter milling time, will significantly reduce mortality rate and expand the area of choices for vertical evacuation shelter locations; (5) when more people chose to evacuate vertically, the total mortality rate reduces; and (6) vertical evacuation shelter placement is a multi-criteria decision making problem, often with competing goals. The results of this research provide an evidence-driven vertical evacuation modeling framework to guide decision makers at city, state, and federal level to understand the dynamics of vertical evacuation behavior and choice of vertical evacuation shelter locations for a community.

• Supplemental Notes:
  • This paper was sponsored by TRB committee ABR30 Standing Committee on Emergency Evacuations.
• Authors:
  • Mostafizi, Alireza
  • Wang, Haizhong
  • Dong, Shangjia
  • Cox, Dan
• Conference:
  • Transportation Research Board 97th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2018-1-7 to 2018-1-11
• Date: 2018

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References;
• Pagination: 8p

Subject/Index Terms

• TRT Terms: Evacuation; Intelligent agents; Location; Mathematical models; Multiple criteria decision making; Shelters; Tsunamis
• Uncontrolled Terms: Agent based models
• Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting; Security and Emergencies;

Filing Info

• Accession Number: 01662702
• Record Type: Publication
• Report/Paper Numbers: 18-06293
• Files: TRIS, TRB, ATRI
• Created Date: Mar 16 2018 9:50AM