Arsenic is one of the most serious inorganic contaminants in drinking water on a worldwide scale. To comply with the MCL (maximum contaminant level, 10 mu g/l arsenic in drinking water) established by the World Health Organization, numerous techniques have been studied, such as ion exchange, coagulation and flocculation, precipitation, adsorption and membrane technologies. Among the available technologies applicable to water treatment, membrane filtration has been identified as a promising technology to remove arsenic from water. The goal of this study is to demonstrate the technical and economic viability of removing arsenic (V) using an optimized reverse osmosis process, with minimization of the total cost as the objective of the optimization strategy. The optimization results showed that the total costs of a two-stage membrane cascade used for the removal of arsenic (V) from drinking water for a population of 20,000 inhabitants were 1041 $/d and 0.52 $/m(3) of drinking water produced. Energy consumption was the most relevant cost, corresponding to 35% of the total cost. Sensitivity analysis was performed to determine the total costs of the installation for different scenarios in terms of drinking water production: (i) 0.44-0.56 $/m(3) for electricity prices of 0.05-0.10 $/KW h; (ii) 0.88-0.45 $/m(3) for populations ranging from 5000 to 50,000 inhabitants; and (iii) 0.52-0.61 $/m(3) when the membrane lifetime was reduced from 3 to 1.5 years. The multiobjective optimization solutions, which consider the best compromises among the quality and cost objectives, indicated that the concentration of As (V) in the permeate water can be reduced to 0.5 mu g/I at a feasible cost. (C) 2015 Published by Elsevier B.V.