Discussion of "Use of Flow Meters for Managing Water Supply Networks"
There is general agreement that water nodal demands, including their temporal pattern, are a major source of uncertainty when calibrating hydraulic network models (Walski et al. 2003). For that reason, a number of methods have been proposed in recent years for their adequate characterization, taking into account their stochastic nature (Kiya and Murakawa 1989; Buchberger et al. 2003).The main goal of the authors of the paper under discussion (in the discussers' opinion) is to extend the use of these models to a new, relevant application. The authors present a model that allows random generation of water demands from measurements in domestic appliances and statistical analysis of the network flows.This information is then used to determine the locations of flow meters and a suitable time step for monitoring pipe flow rates. The proposed methodology is of great interest because of growing concern about the efficient management of water-distribution networks and the reduction of leaks by subdividing the network in district metered areas (DMAs). In the discussers’ opinion, the procedure is well constructed. However, a few comments can be made regarding formal issues, the assumed hypotheses, and model verification.