In the context of global climate change and rapid urbanization, floods are increasingly frequent and intense, and the scientific allocation of pre-disaster flood prevention funds is essential for enhancing cities' preparedness for sustained flood protection. However, existing studies often overlook urban social vulnerability in allocating funds for pre-disaster flood prevention and quantitatively analyze the benefits of the allocated flood prevention funds. To solve this problem, relevant indicators like socioeconomic factors, infrastructure, and environmental conditions were used to assess social vulnerability (SoVi). Based on the assessment results, a Disaster Prevention Funding Allocation Model based on Social Vulnerability Assessment (SO-FAOM) was constructed, which aims to reduce the overall social vulnerability index and the number of extremely vulnerable areas. In this model, Geodetector was used to establish a hierarchical quantification mechanism for benefit coefficients across different levels of priority support, and NSGA-II and TOPSIS were employed to solve for the optimal allocation. Applied it to Wenzhou City to obtain the optimal allocation of funds and quantitatively analyze the benefits generated by the funds. Additionally, with the same total investment, the benefits of the Flash Flood Intensity Scheme (FFIS) and SO-FAOM allocation results were compared. FFIS reduced the total SoVi by 4.2 % and the number of extreme SoVi zones by 35.5 %, while SO-FAOM achieved reductions of 5.8 % and 57.0 %, respectively. Results showed that the scientific allocation of flood prevention funds by incorporating socioeconomic factors pre-disaster could help maximize the effectiveness of the use of funds and improve the city's ability to sustain flood protection.