A stochastic approach to resilience assessment of urban water systems from source to tap

The design of urban water systems faces long-term uncertainties in a multitude of parameters, from the hydroclimatic and socioeconomic realms, such as population growth, climate change and shifting demand patterns. To analyze such systems in a holistic way, many models for sub-systems are typically involved, while the performance of different designs is generally measured against a variety of metrics and in different times scales for each sub-system. In this work, we present a framework for stress-testing urban water systems based on the novel metric of a system's resilience, i.e., the degree to which a water system continues to perform under progressively increasing disturbance. The framework covers the entire water cycle, by coupling a water resources management model to a hydraulic water distribution model thus covering the water system from source to tap. The framework is underpinned by a stochastic simulation module supporting the representation and capturing of uncertainty throughout the water cycle. To assess the system's resilience under uncertainty, we "stress-test" it with an ensemble of scenarios whose parameters are stochastically changing within a design horizon. The approach is showcased through a synthesized case study.