Integrate human dynamics in disaster risk management: An agent-based approach to evaluate policy effectiveness

Abstract

Although DRR (disaster risk reduction) policies have been proved effective in some regions, policy implementation lacks a sufficient evidence-based evaluation process. This study aimed to evaluate the effectiveness of DRR policies, including setting up early warning systems, constructing disaster shelters, and incentive mechanisms. By using the agent-based modeling (ABM) approach, a policy evaluation model was developed to integrate human individual differences during disaster events. The model was calibrated to simulate the DRR policy implementation in a debris flow event at Longchi town, China. The main findings are: 1) setting up an early warning system was the most effective measures and fundamental of community-based disaster risk management as the system had contributed to a 30.06% casualties reduction in the case of Longchi town; 2) individual perception on DRR policies was at large variance which influenced the policy effectiveness; 3) marginal benefits of policies to raise public willingness might decrease quickly. Therefore, individual perception and behaviors have a significant impact on the effectiveness of DRR. This study provided an evidence-based approach to the policy-makers to formulate the most cost-effective DRR policies.