AI-powered optimization of reactive red 195 dye decolorization: Evaluating the impact of operational factors

Machine learning modelling and optimization of the degradation of synthetic dyes are essential aspects that need the attention of researchers when cleaning wastewater. This study pioneers a synergistic approach combining phyto-assisted green synthesis of magnesium oxide nanoparticles (MgO-NPs) with advanced machine learning techniques to degrade Reactive Red 195 dye from textile effluent effectively. Utilizing the eco-friendly and sustainable properties of Azadirachta indica (Neem) leaf extract for MgO-NPs synthesis, we further integrate a sophisticated gradient-boosting regressor model to analyze and predict the decolorization efficiency under varied conditions. The machine learning model, achieving an accuracy (R² = 0.7), not only enhances our predictive capabilities regarding dye decolorization from industrial wastewater but also underscores the critical influence of parameters such as pH, temperature, and time on the treatment process. We found the optimal decolorization to be 79.53 % under the following conditions: concentration = 0.0278, time = 56.67, MON = 5, pH = 4, and T = 40 °C. This method avoids using additional chemicals, offering a more eco-conscious solution for dye decolorizing industrial wastewater and incorporating machine learning into environmental nanotechnology research results in a significant step forward, enabling the predictive optimization of treatment methods and facilitating the development of more efficient, data-driven solutions for small and medium enterprises addressing sustainable approaches for optimizing industrial waste treatment.