Urban development leads to replacement of impervious surfaces with pervious surfaces, vegetation and production of various environmental pollutants. In addition, this phenomenon has increased the devastating effects of urban stormwater and has exacerbated the water quality of many freshwater resources (surface and underground) throughout the world. To alleviate this condition, new models of urban stormwater systems, including Water Sensitive Urban Design (WSUD) as well as its software package, MUSIC, have been developed. Regarding the water sensitive urban design, some urban stormwater treatment devices have been proposed and implemented to alleviate the effects of land use changes and improve the quality of downstream water. Although this new models provide cost-effective urban stormwater management methods, it is very challenging to determine an optimal mixture of various treatment devices and minimize the cost of urban stormwater control systems in this regard. Traditionally, the optimum and cost-effective urban stormwater treatment systems have been developed using simulation models such as MUSIC on the basis of trial and error. It should be noted that the latter models require considerable time to get close to an optimal solution. In this study, a simulation-optimization model of urban stormwater management system has been developed using a combination of MUSIC simulation model and Genetic Algorithm thereof. The simulation-optimization model (MUSIC-GA) greatly reduces the computation time and facilitates direct access to absolute optimal solution. The results of a case study on the application of the developed model in Australia have also shown that there have been some desirable convergences between the results of the simulation-optimization model and absolute optimal solution over various implementations of the model.