Author(s): H. R. Tiedmann; L. Sela; K. M. Faust
Keywords: Hydraulic model; Water utility; Smart urban water management; Academia-industry collaborations
Abstract: Water infrastructure systems are designed for the provision of services to communities under a set of conditions. When these circumstances change (e. g., due to hazards, crises, or population dynamics), hydraulic models can enable water utilities to quickly evaluate system performance and respond accordingly. These hydraulic models can provide an efficient and cost-effective way to simulate changing conditions, test different response scenarios, and better plan for the future, thereby improving system resiliency. However, hydraulic model development and upkeep are time-consuming processes requiring expensive software, skilled technical staff, and immense amounts of data. Such barriers lead to further disparities between wealthier and/or urban utilities that can afford modelling programs and smaller, rural, and/or resourceconstrained utilities that cannot. The amount of data and data processing needed during model development poses significant challenges, especially because the water sector trails other industries in the application of data science and analytics . Further, most modelling literature and technical resources assume the modeller already has all data in the appropriate format and focus only on model building and analysis, excluding data acquisition and processing. While other researchers have put forth steps for improving the use of data analytics in the water sector , demonstrated new data management systems and technologies[3,4], or offered data new classification systems in the hydrology space , a classification system for hydraulic modelling data needs does not exist. To address this gap, we developed a hydraulic model of a real-world water distribution system, classified the data needs, documented the data collection and processing stages, and identified key success factors and challenges encountered. While hydraulic models can offer significant benefits to utilities during crisis response, if model development begins at the onset of a crisis, data-related challenges will likely prevent the model from being ready for use before the event is over.