In view of the dynamic characteristics of flexible job shop scheduling under machine fault disturbance and the ambiguity of job delivery time, this paper adopts a rolling window rescheduling strategy driven by event and cycle, and uses the method of linear weighted sum to establish a multiobjective flexible job shop dynamic scheduling model with the objective of minimizing the maximum completion time, minimizing energy consumption and maximizing customer satisfaction. A GASA algorithm combining genetic algorithm with simulated annealing algorithm is designed. The effectiveness of the algorithm is verified by comparing the simulation results with those obtained by genetic algorithm.