The aim of this work was to study the influence of the pre-treatment step, influent chemical oxygen demand (COD), and hydraulic retention time (HRT) on the decolourization and COD removal efficiency of the upflow anaerobic sludge blanket (UASB) reactors for treating textile wastewater. Statistical models were formulated based on these three variables to optimize the decolourization and COD removal efficiency in the UASB reactor using a full factorial central composite design. The high correlation coefficients (R-2=0.99) and the low p-values (0.0001) reveal that the models and model terms are significant, which can be used to optimize the operational variables in an adequate way for the prediction of response variables. The COD removal efficiency of 70% and decolourization efficiency of 81% were observed for real textile wastewater treatment by UASB reactor without pre-treatment. Whereas for pre-treated real textile wastewater, these were 95% and 100%, respectively. The pre-treatment using a pre-investigated composite coagulant (MC+ACH) was vital in the overall treatment efficiency of the UASB reactor. Validation of model predictions for the treatment of synthetic and real textile wastewaters reveals the efficacy of these models for enhancing the decolourization and COD removal efficiency.