Improving the estimation of volumetric water sales for billing

摘要

This article considers the managerial problem of estimating volumetric water sales for the purpose of billing when consumption data are deficient and when a water utility operates in a heterogeneous urban environment characterized by diverse socioeconomic features. A dummy variable regression model was developed and used to assess how the housing sector affects residential water consumption when the former is moderated by resident population and number of households for a customer account. Results show that spatial variation in residential water consumption exists and that both resident population and number of households are significant determinants. This implies that commercial managers of water utilities should consider service location when estimating "average" volumetric water sales for billing purposes.