Author(s)

S. Mecca & S. Dotto

Abstract

Dynamic models of small municipal water supply systems suffer from lack of residential demand data by end-use. Metering specific appliances for adequate samples of customers is often beyond the financial capacity of municipal budgets. Yet such models offer the promise of informing the decision process and improving the quality of the system. This paper presents a process for normalizing the results of monitoring studies sponsored by the American Water Works Association (AWWA) [1] to water demand extracted from pumping records and utility billing data for the Jamestown Municipal Water System. This water supply system has been extensively studied using a dynamic stochastic model in Stella [2] (Stella is a software product and trademark of High Performance Systems, Inc. Hanover, NH 03755 USA.). The normalization was carried out with a 3-parameter fit covering outdoor usage and a single parameter, which we have called a \“conservation factor”, covering all categories of indoor usage adjusted to minimize the difference between actual annual consumption and AWWA national averages. The normalized results have been incorporated into the Stella model and the structure of the residential demand segment is presented together with a simulated comparison of the actual billing-pumping data and the normalized demand for a model year, 1999. 1 Introduction Municipal water supply systems can benefit from dynamic models that represent the many aspects of their water supply systems. Such models can serve as tools for policy and decision analysis related to problems such as adequacy of supply and how to allocate scarce resources to improving the system. Yet small municipal water

Keywords