A multiple regression model approach was developed to estimate buffering indices, as well as biogas and methane productions in an upflow anaerobic sludge blanket (UASB) reactor treating coffee wet wastewater. Five input variables measured (pH, alkalinity, outlet VFA concentration, and total and soluble COD removal) were selected to develop the best models to identify their importance on methanation. Optimal regression models were selected based on four statistical performance criteria, viz. Mallow’s Cp statistic (Cp), Akaike information criterion ( AIC), Hannan– Quinn criterion ( HQC), and Schwarz–Bayesian information criterion ( SBIC). The performance of the models selected were assessed through several descriptive statistics such as measure of goodness-of-fit test (coefficient of multiple determination, R2; adjusted coefficient of multiple determination, Adj-R2; standard error of estimation, SEE; and Durbin–Watson statistic, DWS), and statistics on the prediction errors (mean squared error, MSE; mean absolute error, MAE; mean absolute percentage error, MAPE; mean error, ME and mean percentage error, MPE). The estimated model reveals that buffering indices are strongly influenced by three variables (volatile fatty acids (VFA) concentration, soluble COD removal, and alkalinity); while, pH, VFA concentration and total COD removal were the most significant independent variables in biogas and methane production. The developed equation models obtained in this study, could be a powerful tool to predict the functionability and stability for the UASB system.