Optimal design of macroscopic water networks under parametric uncertainty

• Optimization of macroscopic water systems under parametric uncertainty is presented.
• Multi annual scheduling of water storage and distribution in a city is considered.
• Parametric uncertainty for the precipitation is considered.
• Results of a case study of Mexico show that the solutions are robust to uncertainty.
• Solutions improve the sustainability by reducing the pressure over natural bodies.

The efficient use of water worldwide is of overriding importance due to its vital role in life. Recently, several countries have suffered water scarcity mainly due to population increase and problems associated to climate change such as the change in the precipitation patterns in the world. In this project, a mathematical programming model for the efficient and sustainable use of water under parametric uncertainty is proposed. The model considers rainwater harvesting (which includes catchment, storage and distribution) as alternative water source; it also considers sustainability aspects from the economic and environmental points of view, maximizing the revenue from the sales of water minus the cost of production and treatment, while maintaining desirable levels of water in the natural reservoirs. The uncertainty is a result of the change in the precipitation patterns. The proposed model is applied to a case study for the city of Morelia, Michoacán in Mexico, considering a time horizon of 5 years. Results show the optimal schedule for water storage and distribution to different sectors of the society (public, agricultural and industrial users). It was found that the use of alternative water sources such as harvested rainwater, along with an appropriate planning schedule of storage and distribution might help reduce the pressure over natural reservoirs even under conditions of uncertainty in the precipitation, while satisfying the water demands in a city.