AWWA JAW56653
Journal AWWA - Short-Term Water Demand Forecast Modeling Techniques - Conventional Methods Versus AI
Journal Article by American Water Works Association, 07/01/2002
Jain, Ashu; Ormsbee, Lindell E.
Full Description
A water utility's primary purpose is to provide safe drinkingwater to its customers, but it must also plan for its customers'future water needs. A critical aspect of this planningis predicting short-term (peak) water demands and optimizingthe water supply system to meet these demands. Jain and Ormsbeeevaluated eight models (four conventional and four artificialintelligence [AI]) to forecast short-term water demand. The modelswere developed and tested using daily water demand, dailymaximum air temperature, and daily total rainfall.AI models outperformed the conventional models and predictedshort-term water demands during both "normal" and drought conditionsrelatively accurately. These AI models can be run using commercialsoftware, if the system operator is experienced in usingmodels, or simple rule-based AI models can be easily developedand applied using historical demand, weather data, and a standardspreadsheet program. The authors encourage utility operators ormanagers to incorporate a short-term water-demand forecastingmodel into their predictions of future water needs. Includes 25 references, tables, figures.