Considering the changeover time in the process of assembly production，a mathematical model with an objective function to minimize the cycle time of robotic assembly line was developed.On the basis of the model，an adaptive strategy was combined into the frame of simulated annealing genetic algorithm，and an Adaptive Simulated Annealing Genetic Algorithm (ASAGA) was then proposed.Finally，an example of the welding assembly for the inner assembly of the left front door in the Body Shop was provided to illustrate greater advantages in the quality of the solution and the algorithm convergence，compared with the SGA (Standard Genetic Algorithm) and SAGA (Simulated Annealing Genetic Algorithm) in solving this problem.