The effective properties of metal matrix composites (MMCs) depend on matrix material and reinforcement property specifications as well as bonding at interphase. The use of numerical methods such as finite element (FE) and mean field homogenization (MFH) can assist in predicting MMC properties thus reducing time and cost of optimizing composite properties through experiments. In the present work, a multiscale representative volume element (RVE) of the microstructure of reduced graphene oxide (rGO) reinforced Aluminium (Al) matrix composite (rGO/Al) is created in MSC DigiMat and analysed using Abaqus software. The effect of porosity and rGO reinforcement on thermal conductivity and strength of the rGO/Al composites is studied. The variation in thermal conductivity between FE-RVE and experimental data is a maximum of 2.2% and a minimum of 0.07% for rGO reinforcement of 1 wt.% and 3 wt.% respectively. The results show good agreement between FE-RVE simulation, MFH and experimental data. This approach can provide an efficient technique for selecting matrix and reinforcement phase properties for MMC fabrication. Keywords: Al/rGO composite, Multiscale finite element-representative volume, Thermal and mechanical properties