A Novel Graph-Based Vulnerability Metric in Urban Network Infrastructures: The Case of Water Distribution Networks

The key contribution of this paper is to embed the analysis network in a framework based on mapping from input space whose elements are nodes graph or entire into an information probability distributions associated objects space. Specifically, node distribution its node-to-node distances and whole aggregation these distributions. In two proposed for analysis: Jensen-Shannon Wasserstein, respectively theory optimal transport theory. This representation allows compute distance between original one obtained by removal edges use as index increase vulnerability induced removal. way new characterization obtained. has been tested real-world water networks. results discussed along those which relate loss efficiency given spectra adjacency Laplacian matrices network. models algorithms considered have integrated analytics can also support other networked infrastructures among power grids, gas distribution, transit networks included.