2026-09-20T21:27:51Zhttps://api.repository.cam.ac.uk/server/oai/request

oai:www.repository.cam.ac.uk:1810/3693082024-06-13T00:43:51Zcom_1810_219477com_1810_256064col_1810_219486

Methods to Preserve Social Networks at Emergency Shelters https://doi.org/10.17863/CAM.109171 Douglass, Nathaniel Bithell, Michael During large-scale disasters, social structures are disrupted. The current focus of emergency management is to ensure that people evacuate safely. Lost in the process is a mechanism for keeping existing social structures intact, despite substantial evidence that maintaining social structures during and after disasters is beneficial. Instead of displaced people self-assembling at emergency shelters, where the social composition may appear somewhat random, this research introduces algorithms incorporating graph and hypergraph partitioning to reunite displaced people with their friends, family, and communities within shelters without unduly prejudicing the speed and efficiency required during evacuations. The research further investigates how to achieve this result dynamically, by assessing real-time information flows and readjusting recommended shelter locations mid-evacuation, all the while maintaining social connections tailored for each individual. University of Cambridge 2023-08-31 Thesis Doctoral Doctor of Philosophy (PhD) eng https://www.repository.cam.ac.uk/handle/1810/369308 https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/f5a29a63-5fd4-4cd1-9bc5-f0bafbdccc17/download 7269e01c7ee32a17102308a6a36f20cb https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/4126039c-b99c-4df1-bf72-96f0391d3abe/download 87eda9de84448d1f82354d60eee3eb5f https://www.rioxx.net/licenses/all-rights-reserved/ agent-based model centrality disaster displaced people emergency management emergency planning evacuation evacuee graph graph theory hypergraph mobility network science partitioner partitioning Red Cross refugee reunification shelter simulation social network social network analysis