A local time-dependent reliability analysis method of bridge girder based on structural health monitoring (SHM) data is presented in this paper. Firstly, the original monitoring strain data collected by SHM system is preprocessed to eliminate the influence of monitoring error; secondly, considering the influence of the temperature effect on the statistical distribution of the monitoring strain data, in this study, the cluster analysis method is employed to divide the statistical time period of SHM data for the local position of bridge girder based on the annual temperature monitoring data, and the extraction of the statistical characteristics of the monitoring strain data in the statistical time period is realized; then, the time-dependent probability density distribution of structural resistance of bridge girder with considering resistance attenuation is analyzed based on the time-dependent model of structural resistance; finally, the whole period time-dependent reliability of local location for the existing bridges is analyzed reasonably. Through the case study of real bridge, the rationality and effectiveness of the proposed method are verified, which provides the theoretical basis and application method for the local time-dependent reliability analysis of bridge girder.