Used Frying Oil Decolorization Study Using Eco‐Friendly Nano‐Clay Adsorbent: Process Optimization by RSM and ANFIS Approaches

ABSTRACT Waste remediation offers a sustainable pathway for the recovery and valorization of waste resources. In this study, the process of decolorization of used frying oil (UFO) was investigated using an eco‐friendly Fuller's earth (FE) adsorbent. The key process parameters, including adsorbent dosage, contact time, and temperature, were optimized and analyzed using Box Behnken Design—response surface methodology (BBD‐RSM) and Adaptive Neural Fuzzy Inference System (ANFIS) models. Under the established optimal condition of FE dosage of 4.5 wt%, contact time of 30 min, and temperature of 90°C, the predicted color removal efficiencies of the RSM and ANFIS were 94.91% and 95.60%, respectively, while the actual removal efficiency achieved was 95.65%. The statistical analyses of the models showed that the ANFIS model exhibited superior predictive capability, with a higher coefficient of determination ( R 2 ) of 0.999 and a lower mean absolute percentage deviation (MAPD) of 0.20%, as compared to the RSM model with R 2 of 0.987 and MAPD of 0.68%. Sensitivity analysis indicated that the bleaching temperature and adsorbent dosage were the most significant variables affecting the process. The physicochemical characteristics of decolorized used frying oil (DUFO) suggest its potential as a useful industrial raw material for the production of value‐added products, such as biodiesel, hydraulic fluids, soaps, and bio‐lubricants.