An improved nondominate sorting genetic algorithmII (NSGAII) was used to optimize the intake and exhaust system of a single cylinder gasoline engine, whose torque at middle range speed was decreased noticeably after a catalyst was added. First, a full GTpower model was set up and verified based on experiment data. Then the intake and exhaust system was selected as the optimization variables, and the effects of the diameters and lengths of intake/exhause pipe on the engine performance were investigated individually. Finally an improved nondominate sorting genetic algorithmII (NSGAII) was used to solve the multiobjective optimization problem. The results have shown that the diameters and lengths of the intake/exhause pipe have different ranges and degrees of influence on the engine torque. Different multiobjective optimization simultaneity can be satisfied by adjusting those variables separately. The improved NSGAII, with an elitist preserve strategy and duplicates removing algorithm added, can get the Pareto solution set effectively and satisfy the multiobjective optimization.