Conflicts Resolution in Flood Management through an Interactive Two-Level Stochastic Approach

Flood is becoming more frequent and widespread due to extreme weather and rapid urbanization. Flood could lead to a series of disastrous socio-economic and environmental consequences such as fatalities, damages to property and financial losses. Diversion of flood to pre-planned regions is helpful to mitigate the significant impacts of flooding under extreme weather. In flood diversion management, many socio-economic and environmental factors need to be considered and multiple decision makers with conflictive management objectives exist simultaneously. Considering the priorities of decision makers in the decision-making process, the management objectives need to be sequentially satisfied. The conflicts among these objectives make it difficult to make efficient policies for flood management. Resolving the conflicts is critical for meeting the needs of different stakeholders in face of a flood event. In order to effectively address priority of various decision makers and reflect their tradeoffs, a two-level two-stage stochastic model is proposed and applied to a representative case study for flood management. Compromises among decision makers in a hierarchical decision-making process are evaluated to seek the optimal strategies to meet all the management objectives. The proposed model is also capable of tackling the probabilistic and non-probabilistic uncertainties. The pre-regulated flood management policies are corrected in a two-step process for improving the robustness of decisions under uncertainty.