Electrochemical machining (ECM) is one of the non-conventional machining processes which is mostly used to machine difficult-to-machine materials such as super alloys, Ti-alloys, stainless steel, alloy steel etc. The major requirement of the process is that work piece should be electrically conducting in nature. A large number of parameters influence material removal rate (MRR) and surface roughness (SR) of parts produced by ECM. Usually, tool makers use thumb rules and machine manuals to set optimal parameters for the process. In this work, response surface methodology is adopted to study the effect of four important parameters such as current, voltage, flow rate of electrolyte and inter-electrode gap on MRR and SR. Statistically validated regression equation are developed relating response like MRR and SR with input parameters. Finally, a non-dominated sorted genetic algorithm is used to find out the optimal process parameters that simultaneously maximize MRR and minimize SR. The set of Pareto solutions provide flexibility to the tool makers to choose the best setting depending on applications. © 2012 Published by Elsevier Ltd.