Flood Resilience of a Water Distribution System

Extreme weather events such as heavy rains and floods are becoming more frequent and severe due to global warming, therefore leading to an increasing interest in methods to evaluate environmental consequences and mitigation strategies. Water supply systems (WSS) represent a class of safety-critical infrastructure prone to damage, with direct impact on public health. They can be cast in the class of cyber-physical systems, since their operation is governed by their physical behaviour—related to topology, fluid-dynamics and technology—which in turn is steered by operation policies and user behaviour—pump and valve management, demand–response mechanisms, etc. In this context, we propose an approach to estimate resilience in the indirect damage caused by a flood on a Water Supply System (WSS). To this end, we combine analysis of an inundation model, which computes the floodwater depth over time on the studied territory, and evaluation of a hydraulic network model by a Pressure-Driven Demand (PDD) approach, which also allows for demand–response mechanisms. Flood damage is assessed in terms of both lack of service experienced by inhabitants and length of pipeworks contaminated by floodwater. The approach is experimented on the WSS of Florence, Italy, which serves about 380,000 users and lies in a flood-prone territory. A sensitivity analysis is with respect to demand–response efficiency, speed, and start time.