Thermal methods such as in-situ combustion are used to improve the mobility of heavy crude oils. The in-situ combustion process usually starts by injecting an oxygen-bearing gas and artificially igniting the crude oil. As gas injection is continued, burning of crude oil develops, and the self-sustained combustion front drives the crude oil toward the production wells. Several investigators have developed various types of in-situ combustion simulators. A comprehensive historical development of in-situ combustion simulators was given by Farouq-Ali (1977). Among early investigations, perhaps the most informative and rigorous paper on insitu combustion simulation is that of Crookston et al. (1979). Their two-dimensional model simulated three-phase (water-oil-gas) flow in a reservoir where heat was assumed to transfer by conduction and convection, and mass flows by convection. The simulator included gravity and capillary effects, heat loss to the overburden and vaporization-condensation of water and oil. The main feature of this simulator is its rather sophisticated kinetic model. The authors accounted for four different reactions: light oil combustion, heavy oil combustion, heavy oil cracking, and coke combustion. This kinetic model is partly based on experimental results of Dabbous and Fulton (1974) and Boussaid and Ramey (1968). In 1980, Coats (1980) developed a fully implicit three-dimensional in-situ combustion simulator which allowed any number of components and any number of reactions to occur simultaneously. Later, Rubin and Buchanan (1985) extended the work of Coats to include options for Cartesian, non-Cartesian, radial or specific curvilinear grids. To date, all in-situ combustion simulators assumed that a finite number of chemical reactions, coupled with reaction rates that did not account for all the physico-chemical properties of the reservoir, occurred. The kinetic models required the knowledge of activation energies and frequency factors. However, several investigators (Zsako, 1968; Boussaid and Ramey, 1968; Burger and Sahuquet, 1972; Vossoughi et al., 1982a; E1-Shoubary, 1986) have shown the capability of such methods as thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) to predict the kinetics of the in-situ combustion process. This study describes the development and application of a fully implicit one-dimensional in-situ combustion simulator using TGA/DTG experimental results, which requires the knowledge of neither the reactions activation