In order to optimize the delivery route of chilled fishery products vehicles, a corresponding mathematical analysis model was constructed. On the basis of traditional genetic algorithm (GA) and simulated annealing algorithm (SA), GA-SA algorithm is proposed by integrating the two algorithms. The initial solution of the model was obtained by the methods of random generation. According to crossover mechanisms of GA algorithm, the information in different chromosomes were exchanged. The mutation operation was performed by the SA algorithm. The GA-SA algorithm can avoid the drawback of GA algorithm being prone to getting stuck in local optima, which helps to obtain the global optimal solution. The combination of the two algorithms can take advantages of the two algorithms and approach the optimal solution more effectively. The effectiveness of the algorithm was verified by simulation experiments.