Uncertain Resource Allocation for Coastal Floods Based on the Resilience of Interdependent Water Infrastructures

Floods are known as one of the most frequent natural events that impose many risks and consequences, especially for coastal communities. Utility companies such as wastewater and electricity have to improve the resilience of their assets vulnerable to the occurrence of extreme events. In this study, a framework is developed to allocate financial resources for resilience improvement of interdependent wastewater treatment plants (WWTPs). A probabilistic multicriteria decision-making (MCDM) framework is developed to assess the attribute-based flood resilience of WWTPs and to better understand the uncertainties associated with random attributes. In the second method, the best management practices (BMPs) are designed to evaluate their effects on the inundated areas, and the recovery of damaged facilities is used to determine the performance-based resilience of WWTPs through network modeling. System characteristics and relevant information for modeling and assessing wastewater facilities, and their power supply and pumping dependencies are identified and considered in the framework. The combined attribute and performance-based resiliencies provide a unique metric for resource allocation. The methodology was tested on the southern coast of Brooklyn in New York. The results of the study demonstrated that by considering uncertainties in the MCDM framework, financial resources allocated to each WWTP can vary by as much as $2.7 million (over 10%). This results in different allocation schemes that enable better-informed decisions on the distribution of funds for resilience improvement according to decision makers’ risk-taking attitudes. Additionally, implementing BMPs resulted in a maximum of 10.2% on top of 14.6% resilience improvement of interdependent WWTPs. The Rockaway WWTP, which has no infrastructure dependencies with the other two WWTPs, was not considered in this part of the research. The proposed models can be utilized in similar geographical settings.