Development of a Global Classification Index for Drinking Water Quality Monitoring

Area of Science:

• Environmental Science
• Water Resource Management
• Public Health

Background:

• Assessing drinking water quality is crucial for public health and effective resource management.
• Existing methods for water quality assessment can be complex and time-consuming.
• Standardized, objective tools are needed for rapid and precise evaluation of distributed water.

Purpose of the Study:

• To develop and validate an automated classification tool for assessing drinking water quality in supply networks.
• To establish a global quality index using multicriteria decision-making methods.
• To identify critical parameters influencing drinking water quality.

Main Methods:

• Development of a global quality index using the Analytic Hierarchy Process (AHP) for objective weighting of 23 physico-chemical and bacteriological parameters.
• Implementation of a fully automated classification system using a Python algorithm, categorizing water into five quality classes.
• Application of the tool to a dataset of 1718 water samples from Bejaia, Algeria.
• Sensitivity analysis using the Sobol method to determine the impact of individual parameters on overall water quality.

Main Results:

• The automated tool successfully classified drinking water quality based on 23 parameters.
• Application to 1718 samples showed 97.56% of cases falling into the 'good' and 'very good' quality classes.
• Sensitivity analysis identified total coliforms, manganese, calcium, and conductivity as highly influential parameters.

Conclusions:

• The developed automated tool provides a rapid, precise, and objective method for assessing drinking water quality.
• The findings confirm the high quality of distributed drinking water in Bejaia, Algeria.
• The tool enables efficient identification of at-risk situations and targeted interventions for intelligent water quality management.