Road flooding has always caused many problems in cities due to inadequate surface runoff collection network. Therefore, it is important to identify flood-prone areas and channels that have been flooded. In this study, using SWMM model, the sufficiency of Surface Runoff Collection Network in Shahrekord was investigated during 2, 5 and 10 year rainfall returns. It should be noted that so far no studies have been conducted on the issue of urban runoff in this area. In this simulation, model calibration was performed for discharge index at 3 and depth index at 2 rainfall events. The error coefficients of NSE, RMSE, and BIAS% were used to compare the simulated model error with the observed values. By sensitivity analysis, equivalent width index was selected as the most sensitive index of the model. The validation was performed on two discharge parameters and runoff depth, each in 2 separate rainfall events, in several random canals and nodes. The acceptable values for the error coefficients showed high accuracy of the simulation. After the validation phase, the model was run for 2, 5, and 10 year rainfall return periods and it was found that flooding occurs in 19. 4%, 20. 68% and 21. 52% of city canals, respectively. The location of the flooded channels indicates that the southwestern part of the city will be flooded. The optimal dimensions of the canals to prevent flooding were determined during the 10-year rainfall return period and the volume of concrete needed to modify the dimensions of each canal was also estimated.