Enhancing urban water resilience through stormwater reuse for toilet flushing

• New version of the analytic-probabilistic model for optimal management of rainwater. • Use of artificial intelligence for statistics and prediction of user's consumptions. • Considering water demand randomness and user habits enhances system effectiveness. • Model validation in an Italian case study confirmed the probabilistic approach's reliability. Droughts can sometimes make water supply critical, particularly in urban areas, which are particularly vulnerable to water scarcity due to the growing demand caused by urbanization. To improve the resilience of water networks, the use of alternative water resources, such as rainwater, can support traditional networks. The present study proposes a new version of an analytic-probabilistic approach to evaluate the probability of stormwater reuse for toilet flushing. It considers water demand as a random variable in the modelling, simplifies the contribution of previous rainfalls to rainwater availability, and makes use of a cloud-streaming platform for the statistical analysis and predictions of users’ consumptions. The equations were tested in a case study in a residential district of the city of Palermo (Italy), where a field campaign collected measurements for water demand from toilet flushing for different users and roof surfaces. The validation of the new formulas confirms the reliability of the model, which allows for expeditious verification of the efficiency of rainwater harvesting systems under different weather and demand conditions. It serves as a supporting tool in both the design and performance estimation of such systems, contributing to the sustainable management of water resources.