CLASSIFICATION OF COMMERCIAL, INDUSTRIAL, AND INSTITUTIONAL WATER USERS THROUGH CLUSTERING OF PROPERTY APPRAISER AND WATER BILLING DATA

摘要

The US EPA and the Water Research Foundation have identified as a major research need, the development of a standardized methodology to classify commercial, industrial, and institutional (CII) water users. Schemes to classify CII customers differ greatly and often lack adequate disaggregation, making evaluations of CII water use across sectors difficult. The primary challenge in evaluating CII water use and classifying its customers is that water use patterns vary widely because of the diversity within the CII sectors such as retail stores which range from small convenience stores to regional shopping malls. This large water use variability is driven by differences in water use processes and technologies, varying employee and visitor occupancy patterns, and seasonal effects driven by fluctuations in populations and climatic factors. In order to overcome these complex challenges, this methodology proposes the use of utility water billing data spatially linked to property appraiser data to arrive at a detailed understanding of how CII customers use water. In order to arrive at a standardized approach that can be applied across the United States, this methodology uses clustering techniques to arrive at homogeneous CII categories. For this analysis, actual billing and property appraiser data for nearly 7,000 CII customers from Florida-and Texas will be used. This will provide a large enough sample by which to cluster on the intensity of water use (liters per square meter of building area per day), and property appraiser attributes such a land use classification and heated building area. A standardized classification methodology to separate CII into homogeneous groups would allow a utility to better evaluate their CII sectors for water conservation, as well as project future water demands with greater certainty.