Development of a Sensor-Integrated Fuzzy Logic-Based Decision Support System for Real-Time Classification and Management of Textile Wastewater

Abstract

A prototype wastewater treatment system integrated with a fuzzy logic-based decision support model was developed to characterize, classify, and assess the reuse potential of textile wastewater. Unlike conventional approaches that treat all effluent streams uniformly, this study presents a selective management strategy based on real-time pollution load classification to enable cost-effective and sustainable reuse. A total of 132 samples from various textile processes were analyzed for color, total suspended solids (TSS), pH, conductivity, and temperature. The results indicated that approximately 58% of the wastewater was suitable for membrane treatment and reuse. Sensor data from a jet dyeing machine fed a fuzzy logic inference system that determined wastewater treatability in real-time, automatically routing streams to either the drain or the treatment system. The prototype, consisting of pretreatment units and a reverse osmosis (RO) module, achieved 30–60% treatment efficiencies depending on color intensity and salinity. In dyeing trials with the recovered water, ΔE values (0.16–0.76) remained below the threshold (ΔE < 1), confirming reuse viability. In addition, content analyses of the treated water were performed, and the results confirmed that the recovered water met the necessary quality criteria for reuse in textile dyeing processes. This study demonstrates that data-driven classification significantly optimizes the volume of water requiring treatment compared to non-selective strategies, offering a practical pathway for circular water management in the textile industry.