Urban traffic management challenges smart cities due to limited centralized systems and complicated traffic patterns. To address these difficulties, the study proposes a unique technique for short-term traffic flow forecasting that takes use of edge computing and 5G-enabled vehicles. In contrast to traditional techniques, the proposed system decentralized processing tasks to vehicle edge devices, enabling real-time data analysis and prediction. The system uses machine learning (ML) approaches, such as Long ShortTerm Memory (LSTM) networks, to accurately estimate traffic flow for each area. Results show significant improvements in prediction precision, computational effectiveness, and robustness compared to the existing system. The proposed system has a strong correlation coefficient of 0.85, a decreased mean absolute error (MAE) of 5.2, and a root mean square error (RMSE) of 6.8, indicating potential for better urban traffic management and decision-making. It marks a big step forward in the development of more efficient and sustainable smart city transportation systems.