Based on the general method of life cycle analysis of concrete structures exposed to harsh environment, this method takes the time-varying seismic capability of each failure state as the functional index, and connects the residual function and recovery process after the disaster with the time of the earthquake event. By sampling the time-varying bridge fragility model to obtain the bridge earthquake failure samples, combined with the time-varying function index, the genetic algorithm is used to solve the resource-constrained project scheduling problem, the specific recovery process of the bridge after the earthquake is given, and finally the seismic resilience of the bridge during service is obtained. The results show that when the time-varying function is not considered, the calculated seismic resilience of the bridge is significantly greater than that of the bridge considering the time-varying function, which will overestimate the ability of the bridge to resist and recover from the earthquake, which is not conducive to the development of the post-earthquake restoration work. The selected control time ( t h - t 0 ) should be reasonable. If the control time ( t h - t 0 ) is too small, the calculated seismic resilience of the bridge is zero universally, and the seismic resilience of the bridge cannot be well express.