Fuzzy index for public supply water quality

Abstract

Human population growth and increased industrial activity in recent decades have contributed to a range of environmental problems, including the contamination of groundwater and surface water. In order to help in the management of these resources, water quality indices are used as tools to summarize multiple parameters and express them in the form of a single number. The ability to provide both an integrated assessment of changes in environmental variables, as well as performance tracking, has resulted in such indices being increasingly employed in surface water monitoring programs. The aim of this study was to develop an Index for Public Supply Water Quality (IPS) using a fuzzy inference methodology. Linguistic systems generally provide satisfactory tools for qualitative purposes, enabling the inclusion of descriptive variables with reduced loss of individual information. Validation of the technique was achieved by analysis of measurement data obtained for the Sorocaba River, provided by CETESB. The new procedure proved more rigorous, compared to classical IPS. It could be readily applied in the evaluation of other water bodies, or be adjusted to incorporate additional parameters also considered important for the assessment of water quality. © 2013 IEEE.