WATER DISTRIBUTION NETWORKS REHABILITATION BASED ON A SMALL INFORMATION DATASET

An efficient management policy of water distribution networks should be based on the monitoring of networks conditions and operation hydraulic parameters, as well as on records of past maintenance and rehabilitation works. Unfortunately utilities often do not have a sufficiently detailed and up-to-date knowledge of the systems. The decision-making process on investments to undertake cannot however leave aside a minimal level of knowledge which must be acquired in any case, the main problem being the availability of economic resources and time. The paper presents a possible approach to networks rehabilitation in condition of data scarcity, resorting to head-driven hydraulic simulation and a multi-objectives optimization tool, aimed at determining the best intervention options on the base of data-parsimonious performance indices (PIs). The PIs adopted can be seen as good selection criteria of rehabilitation alternatives, providing insight into customers’ demands satisfaction, water losses, network’s physical condition, reliability and robustness, while still requiring a relatively small amount of data