Sustainable Water Quality Evaluation Based on Cohesive Mamdani and Sugeno Fuzzy Inference System in Tivoli (Italy)

Clean water is vital for a sustainable environment, human wellness, and welfare, supporting life and contributing to a healthier environment. Fuzzy-logic-based techniques are quite effective at dealing with uncertainty about environmental issues. This study proposes two methodologies for assessing water quality based on Mamdani and Sugeno fuzzy systems, focusing on water’s physiochemical attributes, as these provide essential indicators of water’s chemical composition and potential health impacts. The goal is to evaluate water quality using a single numerical value which indicates total water quality at a specific location and time. This study utilizes data from the Acea Group and employs the Mamdani fuzzy inference system combined with various defuzzification techniques as well as the Sugeno fuzzy system with the weighted average defuzzification technique. The suggested model comprises three fuzzy middle models along with one ultimate fuzzy model. Each model has three input variables and 27 fuzzy rules, using a dataset of nine key factors to rate water quality for drinking purposes. This methodology is a suitable and alternative tool for effective water-management plans. Results show a final water quality score of 85.4% with Mamdani (centroid defuzzification) and 83.5% with Sugeno (weighted average defuzzification), indicating excellent drinking water quality in Tivoli, Italy. Water quality evaluation is vital for sustainability, ensuring clean resources, protecting biodiversity, and promoting long-term environmental health. Intermediate model evaluations for the Mamdani approach with centroid defuzzification showed amounts of 72.4%, 83.4%, and 92.5% for the first, second, and third fuzzy models, respectively. For the Sugeno method, the corresponding amounts were 76.2%, 83.5%, and 92.5%. These results show the precision of both fuzzy systems in capturing nuanced water quality variations. This study aims to develop fuzzy logic methodologies for evaluating drinking water quality using a single numerical index, ensuring a comprehensive and scalable tool for water management.