The number of catastrophic events such as extreme rainfalls and hurricanes has been growing. These events pose a major threat to the life safety and economic prosperity of urban regions. Flood control networks play a pivotal role in mitigating the risk associated with the stormwater generated by extreme rainfalls and hurricanes. The objective of this study is to propose a framework to examine the vulnerability in flood control infrastructure networks. This framework applies graph theory concepts and tools to define a vulnerability index for flood control network components (e.g., channels and rivers). The topological attributes of flood control networks are used to determine the vulnerability index based on structural attributes of flood control networks. First, a flood control network is modeled as a directed graph and storage facilities are incorporated into the network. Second, co-location exposure, upstream channel susceptibility, and discharge redundancy are characterized as important vulnerability attributes of a channel in flood control network. Then, these three characteristics are formulized based on the topological attributes of the network and characteristics of channels. The vulnerability index is then determined based on the three vulnerability characteristics. The proposed vulnerability index can be used to evaluate the impact of different risk reduction policies on flood control network vulnerability and determine the optimal mitigation strategies aiming at flood risk reduction, such as widening vulnerable channels, placement of storage facilities in the network or increasing the redundancy of the network. The framework is implemented on two watersheds in Harris County (Texas, USA) and the results' implications for decision-making in infrastructure management and hazard mitigation planning are discussed. The results highlight the capability of the proposed graph-based framework to inform flood risk reduction through evaluation of the vulnerability of infrastructure networks.

极端降雨和飓风等灾难性事件的数量一直在增加。这些事件对城市地区的生命安全和经济繁荣构成重大威胁。防洪网络在减轻与极端降雨和飓风产生的雨水相关的风险方面发挥着关键作用。本研究的目的是提出一个框架来检查防洪基础设施网络的脆弱性。该框架应用图论概念和工具来定义防洪网络组件（例如渠道和河流）的脆弱性指数。利用防洪网络的拓扑属性，根据防洪网络的结构属性确定脆弱性指标。第一的，防洪网络被建模为有向图，并且存储设施被整合到网络中。其次，共址暴露、上游通道敏感性和流量冗余被表征为防洪网络中通道的重要脆弱性属性。然后，根据网络的拓扑属性和通道的特性，将这三个特性公式化。然后根据三个脆弱性特征确定脆弱性指数。建议的脆弱性指数可用于评估不同风险降低政策对防洪网络脆弱性的影响，并确定旨在降低洪水风险的最佳缓解策略，例如拓宽脆弱渠道、在网络中放置存储设施或增加网络的冗余。该框架在哈里斯县（美国德克萨斯州）的两个流域实施，并讨论了结果对基础设施管理和减灾规划决策的影响。结果突出了所提出的基于图的框架通过评估基础设施网络的脆弱性来通知减少洪水风险的能力。