Stochastic model applied to water demand management in Brazil

Abstract

The increasing scarcity of water sources near urban areas, coupled with urbanization and population growth, necessitates the development of solutions that optimize water demand management, which entails a more accurate estimation of consumption patterns and implementing measures that promote the rational use of water. A stochastic residential water demand model was built based on the superposition of pulses of constant intensity and variable arrival time and duration applied to the microcomponent consumption at a 1-min resolution, reflecting the aggregate system contributions to the total instantaneous demand. The model can support the planning of water supply systems (WSSs) and was calibrated using data from Brazilian institutes and a literature review of statistical data on users and end uses of water in Brazilian dwellings. Innovatively, it proposes the use of water demand forecasting aimed at applying demand management measures evaluating the replacement of conventional devices for water-saving ones by analyzing hydrographs of 50–50,000 households, in addition to detailing the influence of the K2 peak factor. The results demonstrated a new water consumption profile with about a 40% reduction in water demand. A new equation for K2 is proposed, which was shown to be underestimated for small populations.