Optimal Combination of Measures for Flood Risk Mitigation Within Probabilistic and Uncertain Situations

Investment strategy in flood prevention measures plays an important role in mitigating flood disasters with great uncertainty. In consideration of the cognitive level of decision-makers, this study aims to propose the optimal investment strategy for flood risk mitigation on the basis of minimizing total discounted costs. A dynamic mixed-integer nonlinear optimization model was first developed to identify suitable types and investment time of flood prevention measures in a probabilistic situation. An additional optimization approach in an uncertain situation was deployed for the comparison of investment strategies based on probability theory and uncertainty theory. The optimization model was applied to the case of regular and extreme floods in the city of Nanjing, China. The results denoted that cognitive level, growth rate of flood losses, discount rate, annual budget for investment, and risk preference had varying impacts on the optimal combination of flood prevention measures. Types and investment times of measures for flood risk mitigation vary widely under two cognitive levels in probabilistic and uncertain situations. Risk mitigation effects and economic benefits will be better when the growth rate of flood losses is higher. With an increase of discount rate, measures requiring a higher cost will be delayed for investment, and measures with a lower cost will be advanced for implementation. Financial budget from the government limits the choice of alternative measures for mitigating flood risks. Under the same growth rate, discount rate, budget, or risk preference, there are more types of flood prevention measures with earlier time for investment in probabilistic situation, in comparison to uncertain situation.