A mixed integer programming model and solution method for the operation of an integrated water supply system

The problem addressed in this paper focuses on the reduction of electrical energy costs for the operation of hydraulic pumps, which are used for catchment and distribution of water in water supply systems. As the electricity tariffs are different according to the time of day, it is necessary to decide when and how many pumps must be used to collect and distribute water, in each period of time, and, when there must be water transfer among tanks so that the demand of each consumer center is met without a lack of resources. A mixed integer programming (MIP) model is proposed and a competitive MIP-based heuristic is developed to solve this problem. Numerical tests in a case study carried out with the water supply system of a Brazilian city demonstrate the efficiency of the proposed mathematical model and solution method.