For improving automobile ride comfort, safety and decreasing tire dynamic load to road, the four degree of freedom suspension model was established. MOGA (multi-object genetic algorithm) and multi-criteria decision method were applied to multi-objective optimization of suspension parameters, with root-mean-square values of vertical acceleration of body, front and rear tire load at 50, 80 and 100?km/h on A,B and C level road as objective functions. The results show that on average, root-mean-square value of vertical acceleration of body reduces by 40%, root-mean-square value of dynamic load of front and rear wheel reduces by 30%, and the performance gets better. This method, making decision after searching the Pareto solutions, can help the designer get the best answer under different demand. Compared with the traditional multi-objective optimization method, it is more applicable and effective.