ABSTRACT This study reports the development of an augmented microbial consortium for the efficient bioremediation of laundry and kitchen greywater. An indigenous consortium isolated from kitchen sludge was enhanced with Micrococcus luteus , Rhodococcus equi , and Aspergillus niger , resulting in significantly improved pollutant removal. Process optimization using Response Surface Methodology (RSM) identified optimal conditions at 33.2°C, pH 8.0, an inoculum size of 198 µL, and a C/N ratio of 1.9. Under these conditions, maximum removal efficiency of 83.5% (COD), 81.5% (oil and grease), and 87.8% (sulphate) were achieved within 96 hrs. The Artificial Neural Network (ANN) model demonstrated high predictive performance across training (R 2 = 0.992), validation (R 2 = 0.893), and testing (R 2 = 0.816) phases, with an overall R 2 of 0.964. The RSM model provided robust individual response predictions (R 2 for COD = 0.966, oil and grease = 0.997, and sulphate = 0.984). These results indicate that ANN captured the nonlinear relationships among operating variables with acceptable predictive capability under the limited dataset conditions, while RSM effectively described individual parameter interactions. Growth kinetic analysis indicated substrate inhibition at higher concentrations, with the Haldane model providing the best fit (R 2 = 0.977). The use of coconut coir as a support matrix provides a promising foundation for future pilot‐scale investigations into decentralized treatment systems.
Rajpal et al. (Thu,) studied this question.