A Rapid Compression-Expansion Machine (RCEM) is a facility widely used in autoignition studies, since fully controlled initial and boundary conditions are guaranteed in it. Thus, the experimental results from RCEMs can be used to validate and analyze both detailed and reduced chemical kinetic mechanisms under transient thermodynamic conditions by replicating the in-cylinder pressure and temperature evolution in the machine, while solving the chemical mechanism and obtaining a simulated ignition delay that can be directly compared to the experimental one. Homogeneous mixtures are usually tested to this aim, since the complex mixing phenomena are removed and the most relevant chemical paths can be more easily identified, allowing the use of 0-D modelling, which is a much simpler modelling approach compared to a CFD (3D) one. Besides of this, if detailed chemistry is needed to be solved, like it is the case here, a model with low computational cost (e.g. 0-D model) is mandatory, which discards CFD simulations. However, 0-D models can lead to unrealistic results, since wall effects cannot be taken into account, which results in faster simulated combustion velocities and much higher pressure rise rates compared to experiments. In this work, a five-zone model has been applied to replicate the in-cylinder conditions evolution of a RCEM in order to improve the simulation results. To do so, CFD simulations under motoring conditions have been performed in order to identify the proper number of zones and their relative volume, walls surface and temperature. Furthermore, experiments have been carried out in a RCEM with different fuels under homogeneous conditions to obtain a database of ignition delays and in-cylinder pressure and temperature evolution. Such experiments have been replicated in CHEMKIN by imposing the heat losses and volume profiles of the experimental facility using a 0-D one-zone model. Then, the five-zone model has been analogously solved and both results have been compared to the experimental ones. The thermodynamical conditions can be much better replicated using the five-zone model. Moreover, the ignition delay predictive accuracy has been also improved in the five-zone model, since slower, more realistic pressure rise rates are reached.