Circular economy transition through physics informed digital twins for sustainable wastewater management in textile industry

The textile industry is a major contributor to industrial wastewater pollution, particularly in developing regions where monitoring infrastructure and real-time data acquisition remain limited. These constraints hinder proactive wastewater management and slow the adoption of circular economy practices in industrial production systems. Conventional wastewater treatment models often rely on extensive data availability or fail to adequately capture coupled biochemical and hydraulic dynamics under operational uncertainty. To address these challenges, this study proposes a data-efficient digital twin framework for sustainable management of textile wastewater treatment systems. By integrating mechanistic process modeling with Physics-Informed Neural Networks, the framework preserves physical consistency while learning system behavior from sparse observations. An inverse learning strategy is employed to estimate key kinetic and hydraulic parameters directly from limited biochemical oxygen demand and chemical oxygen demand data, addressing data constraints in industrial environments. The primary scientific contribution lies in the development of a stable, non-oscillatory modeling framework that supports real-time monitoring, parameter inference, and virtual experimentation within a unified structure. Sensitivity analysis identifies aeration intensity, biological reaction kinetics, influent flow rate, and biomass yield as dominant drivers of pollutant removal efficiency, providing actionable insights for operational optimization. From a sustainability perspective, the proposed approach facilitates circular economy transitions by enhancing treatment efficiency, reducing pollutant discharge, and enabling adaptive, resource-efficient decision-making. The framework is transferable to a wide range of industrial wastewater systems operating under similar data constraints. The proposed digital twin therefore provides a scalable and transparent tool for advancing sustainable industrial water management and circular economy practices.