Risk-based decision-making for infrastructure systems under extreme events

Civil infrastructure is facing severe challenges under the risk of natural disasters. In previous explorations, Expected Value and Minimum Expected Life-cycle Cost criterion has been widely used in the analysis of disaster losses. Expected Value cannot estimate the extreme loss consequences with small probability. Minimum Expected Life-cycle Cost criterion fails to consider a multitude of uncertainties in disaster risk. Utility Theory is also commonly used in this field. There are limitations in using Utility Theory alone to assess disaster mitigation options, especially for multiple outcome-oriented options. For this reason, this paper provides a detailed analysis and a comprehensive overview of the application scenarios of these decision-making models within the realm of civil engineering. An experiment using Range Value at Risk to assess the extreme risk loss of typhoon during the cycle was carried out. Utility Theory, Prospect Theory and Cumulative Prospect Theory were respectively applied in decision-making experiments for bridge disaster mitigation options in typhoon risk. The experimental results demonstrate that, given a risk-averse preference: (i) Range Value at Risk can serve as a quantitative measure for extreme tail risk, (ii) Utility Theory is suitable for decision-making scenarios involving a specific number of options and definite outcomes, and (iii) Cumulative Prospect Theory presents a more comprehensive and reliable decision model.