A parametric study, based on design optimization using distributed genetic algorithm, is conducted to develop a reliable design procedure for steel portal frames. The efforts are made to modify distributed genetic algorithm in order to enhance the quality of performance and accelerate the convergence to possible optimum solutions. The modification includes defining a state-of-art mutation scheme and a reproduction procedure. Software called DO-DGA has been developed to handle the design optimization.The study can help structural designers to work out the member forces required to design the elements of steel portal frames without referring to the complicated structural analysis. Implementing the results of structural optimization by DO-DGA, graphs and tables are developed from which the designers can determine the member forces. The main variables in the parametric study of steel portal frame are the slope of pitched rafter, the applied load to the rafter, the length and height of the haunch and the span of the frame. The procedure is promising since it can bring the design optimization into a daily-use tool in design offices.