A Stochastic ANP-GCE Approach for Vulnerability Assessment in the Water Supply System With Uncertainties

Water supply systems around the world are becoming increasingly vulnerable as a consequence of rising climate instability and the systems' intrinsic complexity. Assessing the vulnerability of such systems is the primary and fundamental step for mitigating the negative impacts of environmental incidents. There are two main challenges in this context. First, since water supply systems are coupled with human-environment systems, the corresponding vulnerability assessment requires consideration of a host of multidimensional criteria in terms of the structures and management. Second, due to the uncertain and dynamic nature of environmental incidents, available information is not sufficient for experts to make accurate real-time judgments for the weights of criteria. To the best of our knowledge, few weight calculation approaches are able to deal with both uncertainties and subjectivities of experts' judgments under varied scenarios of the incident. These problems are solved by developing new assessment approaches consisting of four steps: 1) developing an analytical network process (ANP) criteria framework; 2) proposing a weight calculation approach by integrating game cross evaluation with stochastic ANP based on the proposed criteria framework; 3) calculating the weighted vulnerability values, rating and visualizing the vulnerability grades accordingly; and 4) performing assessment simulations and sensitivity analysis for possible scenarios. A case study of the Shanghai water supply system demonstrates the effectiveness of the proposed approaches, which provides guides for improvement of system performances based on the trends of variations in the vulnerability grades and the dominating criterion identified for each component.