Protection of the missile and flying systems and space structures against thermal shocks resulted from abrupt heating actions, is one of the most important issues in the space and aerospace industry that highly ensures us a safe mission to every corners of the earth and also distant cosmic points. A successful re-entry of aerospace vehicles highly depend on the thermal protection of their structures against aerodynamic heating. passive thermal insulators especially ablative composites are more effective and economic than others. In order to protect aerospace vehicles against thermal shocks of re-entry, ablative composites usually are used. These thermal insulators must have special thermal and mechanical properties and also their manufacture process must be routine. In the current work a simple model of thermal shield is presented, and the effects of different parameters such as surface temperature, flow velocity, and ambient pressure on the oxidation behavior of the graphite are investigated using computational fluid dynamics. A 2D model of a rectangular chamber containing a graphite bullet is considered that flow of oxygen is blown over it. Geometry and meshing of the chamber is defined by Gambit and simulation is carried out using Fluent software. The CFD results show that graphite weight loss rate increases by increasing temperature (up to 1800 ° C) but at 2000 ° C this trend changes and weight loss rate decreases upon further increase of temperature. By increasing the pressure, flow velocity, and also changing other effective parameters such as graphite geometry and oxygen purity, the numerical results had a good coincidence with the experimental data. It was also tried to determine the effect of the composite material on the ablation rate and length of the sample recession. Generally using the current model, one can predict the graphite weight loss rate, ablation behavior of graphite, and the effect of the various parameters. Keywords : Computational Fluid Dynamics, Ablation, graphite Bullet, Weight loss rate, Recession length