An applicable graph theory approach for evaluating structural robustness of urban drainage networks

A structural robustness analysis of an urban drainage network can determine how the network behaves under catastrophic failure before collapsing. In this study, a new analytical approach based on graph theory is employed to investigate the performance of urban drainage networks when exposed to a wide range of structural failure scenarios resulting from a random and four targeted failure scenarios. The structural robustness is determined by computing two fragmentation indicators at each node failure level. To validate the graph theory approach, a hydrodynamic simulation is applied, and a robustness indicator based on total flood volume is defined, to quantify the structural robustness of the networks at each considered conduit failure level. Three existing urban drainage networks are used to demonstrate and characterise the proposed approach. The results using the graph theory perfectly matched the result of the hydrodynamic simulation in the random attacks. In targeted attacks, however, it depends on the type of attack scenarios. Overall, the results show that the graph theory approach is a capable methodology in the structural robustness analysis of urban drainage networks and can help to understand and predict their behaviour in facing a wide range of structural failures.