To alleviate the huge computational cost in supersonic combustor modeling and to improve the accuracy of traditional unsteady flamelet model, a zone flamelet is proposed. The main idea of zone flamelet is to divide the whole turbulent combustion field into a finite number of control zones and the chemical status in each zone is represented by a single flamelet. With proper zone division, the scattering of variables over the mixture fraction space is in controllable small, thus the representative flamelet approaches the real scalar distribution. The flamelets exchange information through flux-conserved convection when across the zone boundary, thus the flamelet variables can be transport from upstream to downstream in a flow manner. Although one additional mixture fraction space is resolved, great computational cost is still saved because the zone division in physical space is much coarser than the flow simulation mesh. A simple historical statistics approach is proposed to estimate the representative temperature, in order to further alleviate the computational cost in solving the flamelet temperature equation usually with numerous sub-models for non-adiabatic terms, e.g. radiation and wall heat loss. The zone flamelet model is then applied to model a scramjet combustor operated at a flight Mach number of 6.5 and a fuel equivalence ratio of 0.8. The performance of zone flamelet model in highly non-equilibrium supersonic combustion is compared with the traditional PaSR model. I. Introduction xperimental measurement of scramjet combustors are difficult, due mainly to the severe thermal load for instruments and the drastic distortion of supersonic flow field by probing disturbance. Especially the time-variance and spatial inhomogeneity of high-Re (Reynolds number) supersonic combustion fields requires transient and three-dimensional information to understand the flow physics in supersonic combustors. Due to the incapability of physical sensors for supersonic combustor measurements, the measurement data are generally scare and can be generally classified as two types. The first type of measurement data is used to analyze the combustion characteristics, including those field information of pressure, temperature, velocity and species concentrations etc. The other type is used to evaluate the combustor performance, including those combustion efficiency, total pressure loss, internal friction drag, wall heat flux and net thrust etc. The second type of data are