The dynamic characteristics of automobile demand are critical for national economic and energy prediction. In this paper, a dynamic automobile demand simulation model is developed utilizing a simultaneous-equations system. The system considers the interaction between supply and demand and motorcycle demand is taken into consideration. Although forecasting automobile demand has been previously investigated, it has not been within such a dynamic simulation framework in Taiwan. Our model includes the current and lagged auto and motorcycle quantity, economic, oil price and management variables. Relevant tests are applied to evaluate the simulation model performance and results show its high capability. Finally, a number of simulation experiments considering various oil price and economic growth scenarios for 1996-2005 are demonstrated.