The current domestic system for bridge maintenance in China primarily relies on bridge inspection data to prioritize the state of bridges and allocate a limited budget accordingly. This strategy focuses on repairing bridges with low performance, concentrating on restorative maintenance without much attention to preventive maintenance. Considering the total lifecycle cost, this traditional maintenance method may require a lower initial investment but leads to significant resource consumption in the mid to late stages. Building on the study of existing highway bridge maintenance systems, this paper develops an intelligent maintenance system based on domestic standards and bridge inspection data. Firstly, it reasonably predicts the long-term performance of bridges based on extensive monitoring data and statistical methods, using dynamic Markov and Weibull distributions model. Then, it establishes a mathematical model that incorporates preventive maintenance concepts, optimized using swarm intelligence algorithms. Finally, as a case study, the model is tested using actual inspection data from a prestressed concrete box girder bridge on the Shenzhen Banyin passage. The results show that, compared to traditional maintenance methods, the intelligent maintenance system can save up to 10% in maintenance costs.