SZU-YUN LINEl-Tawil S.Aguirre B.2026-03-162026-03-162021https://www.scopus.com/record/display.uri?eid=2-s2.0-105027911680&origin=resultslisthttps://scholars.lib.ntu.edu.tw/handle/123456789/736380Emergency financial assistance provided for disaster relief creates opportunities for fraudulent behavior. History has shown that the amount of recovery funds lost to improper and potentially fraudulent payments could research millions to billions of dollars per event. Reducing such cost and investing available resources more appropriately could help communities recover more rapidly from the impact of severe disasters. Unfortunately, research on the effects of fraud activity has been overlooked in resilience research. In this study, a conceptual model is formulated based on the observed phenomena and criminology theory. Then, an agent-based computational model, which includes a simulation environment of a community facing an earthquake disaster, fraudsters, and application inspectors is created to model benefit fraud behavior. The simulation considers the effect of both micro-level disaster demands due to building damages and meso-level social variables on benefit fraud, and estimates the cost to communities associated with these post-disaster crimes. Statistical data from the U.S. government reports in the aftermath of Hurricane Katrina and Rita is used for calibration. The results of simulation runs demonstrate that strengthening application review and lessening target vulnerability can help lessen the loss caused by benefit fraud, but it is important to find the balance between the loss of fraudulent payments and the speed of aid distribution in order to improve the overall resilience of communities.falsebenefit fraudcommunity resiliencefinancial assistance policypost-disaster crimessocial impactconference paper2-s2.0-105027911680