The numerical simulation of combustion processes is an important tool used for design, analysis and optimization of turbines, combustion engines and furnaces, among others. However, one of the major drawbacks that limit the faithful description of reality of combustion models is the computational effort required for the transport equations solution of reactive flow properties such as chemical species mass fractions, which include a nonlinear source term associated to the Arrhenius law. The stiffness and the computational burden related to the determination of such term, largely dominate the cost of simulations that employ detailed models of chemical kinetics combustion. This dissertation describes a study whose objective is to reduce these costs by using a of automatic tabulation technique of the mixture’s thermochemical evolution. Hence, this work presents a state of the art discussion of the technique named in situ adaptive tabulation – ISAT, which shows considerable performance in terms of computational time for the determination of the chemical sources terms, and proposes a modification in the algorithm by delaying the table start to avoid the storage of compositions that exist only in transient state, and which are not statistically representative of the stationary regime. A systematic study of the results in a partially stirred reactor with CO/O2, shows more than 95 per cent gains at the binary tree height used for tabulation, the gains are also optimistic in the storage demand and the results accuracy. A computational time analysis characterizes situations in which the new strategy tabulation could reduce it, when compared to the original algorithm. The strategy performance is confirmed by the study of the chemical system CH4/ar.