An improved cloud adaptive genetic annealing algorithm is proposed for the multi-objective FJSP with machine deterioration effect [16]. In terms of shortcomings for poor local search ability and premature convergence in GA, we improve fitness calculations, cross-variation, etc. Fitness calculation is combined with local search ability and probability jump property of simulated annealing algorithm to make it jump out of the local optimal solution. The Hamming similarity is inserted in the crossover operation, and the similarity is used to detect whether crossover operation is required, which can accelerate the running efficiency and convergence speed of the algorithm. Then, the cross-operation combines the adaptive crossover probability of the cloud model to enhance the global search capability of the algorithm. At last, we set standard position and cross position to improve cross-operation, which can enhance the global search ability of the algorithm. Through simulation experiments, the effectiveness of the algorithm for the integrated multi-objective shop scheduling algorithm is verified.