Despite all the progress in renewable energy research to date, combustion-based energy conversion is still the dominant technology, particularly in the transportation sector; a dominance that is expected to continue in the foreseeable future. It is thus crucial to further improve the conventional combustion engine technologies to meet the global challenge of producing clean and efficient energy. It is now widely recognized that development of improved combustion devices rely strongly on fundamental understanding of turbulent reacting flow processes. Simulation tools with higher fidelity are therefore instrumental to provide further insight into details of such processes. Among various challenges in predictive simulation of combustion phenomena, handling of chemical kinetics is particularly daunting due to complexity of realistic chemical mechanisms and disparity of the scales involved. Moreover, incorporation of chemical kinetics, involving large number of species and reaction steps, is still computationally prohibitive. Additional methods are therefore required to enable affordable utilization of complex kinetics. The objective of our research is to develop advanced methodologies for accurate prediction of turbulent combustion. In this study, a novel dimension reduction method is employed for efficient representation of detailed kinetics. æAn important advantage of this method over global mechanisms is that the total number of rate equations is smaller than the number of species included in the model and the concentration of all the species are predicted. This approach has thus far shown a lot of promise for affordable implementation of detailed chemistry in turbulent combustion simulations.