Managing stormwater in urban areas is a major concern to prevent flooding. Urban drainage system designers are increasingly interested in finding solutions to reduce runoff volume in urban areas through environmentally friendly solutions, such as low-impact development (LID) measures. This chapter investigates the effectiveness of implementing three types of LID, i.e., rain barrel, infiltration trench, and permeable pavement, to reduce flooding in Semnan City, Iran. The study uses three artificial intelligence algorithms to predict flood reduction—Support Vector Machine (SVM), Least Square Support Vector Machine (LSSVM), and Least Square Support Vector Machines-Grasshopper Optimization Algorithm (LSSVM-GOA). For modeling purposes, we allocated the percentage area of different combinations of the LIDs, coupled with the reduced peak flow coefficient, as input data and used the reduced flood volume corresponding to each LID combination as the output data. The models showed high accuracy in both the training and validation stages. The LSSVM-GOA model revealed the highest accuracy in flood prediction (R2=0.9896, MAE=0.0101, and RMSE=0.0185). The combination of permeable pavement with rain barrel provides the best runoff management solutions since it reduces the flood volume and the peak discharge by up to 80% and 90%, respectively, when compared to the current system. Opposing, the infiltration trench provides the lowest effectiveness in flood mitigation, with the potential to reduce the flood volume and the peak discharge by up to 40% and 60%, correspondingly. The use of intelligent algorithms provides an accurate tool for estimating and predicting reduced urban flood volume and supports decision makers in implementing efficient LID for flood mitigation in urban areas.