The aim of this work was to improve the ability of electro-Fenton technique for the remediation of wastewater contaminated with synthetic dyes using a model azo dye such as Azure B. Batch experiments were conducted to study the effects of main parameters, such as dye concentration, electrode surface area, treatment time, and voltage. In this study, central composite face-centered experimental design matrix and response surface methodology were applied to design the experiments and evaluate the interactive effects of the four studied parameters. A total of 30 experimental runs were set, and the kinetic data were analyzed using first- and second-order models. The experimental data fitted to the empirical second-order model of a suitable degree for the maximum decolorization of Azure B by electro-Fenton treatment. ANOVA analysis showed high coefficient of determination value (R(2) = 0.9835) and reasonable second-order regression prediction. Pareto analysis suggests that the variables, time, and voltage produce the largest effect on the decolorization rate. Optimum conditions suggested by the second-order polynomial regression model for attaining maximum decolorization were dye concentration 4.83 mg/L, electrode surface area 15 cm(2), voltage 14.19 V, and treatment time of 34.58 min.