Evacuation planning for homeland defense
(pdf file of powerpoint presentation 1049 kb)

LEAD PRESENTER: Prof. Shashi Shekhar

Computer Science Department, University of Minnesota

200 Union Street SE #4 192, Minneapolis, MN 55455

(612) 624-8307 :fax: (612) 625-0572 :email: shekhar@cs.umn.edu

SHORT DESCRIPTION:

Evacuation route-schedule planning identifies paths and schedules to move at-risk population out to safe areas in the event of terrorist attacks, catastrophes, or natural disasters. Its goal is to identify’ near-optimal evacuation routes and schedules to minimize evacuation time despite limited transportation network capacity and the possibly large at- risk population. Finding the optimal solution is computationally exorbitant due to the extremely large size of the transportation networks (million nodes and edges) and the limited capacities. We propose novel geo-spatial algorithms to determine competent evacuation plans. Evaluation of our methods for evacuation planning for a disaster at the Monticello nuclear power plant near Minneapolis/St. Paul Twin Cities metropolitan area shows that the new methods lowered evacuation time relative to existing plans by providing higher capacities near the destination and by choosing shorter routes. (We have a set of PowerPoint slides including a few with maps of evacuation routes for evacuating population near Monticello’ power plant.)

FUNDING SOURCES:

US Army Research Laboratory (federal agency) is sponsoring the work on use of high performance computing techniques to reduce computation time to produce evacuation plans quickly. Federal Highway Authority (federal agency) will sponsor follow-on work to determine contra-flow configurations of the transportation networks to increase outbound capacities and reduce total evacuation time. Collaborators include Mr. QingSong Lu, Mr. Sangho Kim, Prof. Eil Kwon (Minnesota StateUniversity), and Mr. Mike Sobolesky (MnIDOT).