This paper aims at the methodology for the optimum design of an industrial solvent purification process, where pervaporation was involved as an energetically and environmental sustainable technology. Using a simplified mathematical model to describe the separation process, the optimization problem focused on the minimization of the membrane area (main capital cost of this technology) needed to reach the desired purification level. Formulated as an NLP to minimize the total cost of the process, subjected to design specifications, the optimization problem was solved by GAMS modeling equation system. The use of nonconvex equations together with the influence of hydrodynamics over the membrane performance, required a different approach to the optimal solution of the proposed objective function. The method was illustrated with examples of the determination of the minimum area needed to reach a final water content in the solvent, proving to be a useful tool in the scale-up process.