This research presents a response surface methodology (RSM) with I‐optimal method of DESIGN EXPERT (version 13 Stat‐Ease) for optimization and analysis of the adsorption process of the cyanide from aqueous solution by activated carbon (AC) and composite activated carbon (CuO/AC) produced by pyro carbonic acid microwave using potato peel waste as raw material. Pyrophosphate 60% (wt) was used for impregnation with an impregnation ratio 3:1, impregnation time of 4 h at 25°C, radiant power of 700 W, and activation time of 20 min. Batch experiments were conducted to determine the removal efficiency of cyanide from aqueous solution to evaluate the influences of various experimental parameters such as contact time, cyanide concentration, initial pH, and dosage of adsorbent on cyanide removal efficiency. The statistical analysis showed that the quadratic model was significant. The model very low probability value (p‐value < 0.0001) for activated carbon and composite activated carbon, and the analysis of variance showed a high coefficient of specification values of adjusted R2 (model accuracy with observation) and predicted R2 (model accuracy without observation). The optimum conditions suggested by the model for the process variable were 160 min, pH = 10, 10 mg/L, 0.1 g/50 mL, for time, pH, concentration, activated carbon, and composite activated carbon dose, respectively. The maximum removal efficiency at these conditions was 98.2% for AC and 99.35% for CuO/AC. The equilibrium adsorption data obtained were analyzed by Langmuir and Freundlich isotherm models and results showed that it was better described by the Freundlich model. Likewise, the data of adsorption kinetic was analyzed using two models: pseudo‐first‐order and pseudo‐second‐order. The results showed that adsorption kinetics was well performed by the pseudo‐second‐order model.