Despite the wide range of applications, the traditional computational fluid dynamics-discrete element method (CFD-DEM) simulations have been running into bottlenecks when dealing with large scale systems. To overcome this issue, an augmented coarse-grained CFD-DEM approach which combines the reduced particle stiffness model and the coarse-grained CFD-DEM is developed to multi-dimensionally reduce the computation load. In spatial scale, several original particles are lumped into a computational parcel based on the coarse-grained ratio while the time step of solid phase is enlarged in temporal scale according to the reduced particle stiffness ratio. The accuracy and efficiency of the augmented coarse-grained CFD-DEM are quantitatively evaluated in fluidized beds, and different gas-solid characteristics are obtained via comparing with experimental measurements and traditional CFD-DEM. The results show that the augmented method contributes a significant improvement in computational efficiency while a little loss in numerical accuracy. Generally, the computational efficiency of the augmented coarse-grained CFD-DEM increases with the increase of coarse-grained ratio while it enhances with the decrease of the reduced particle stiffness ratio. This augmented coarse-grained CFD-DEM is expected to be a promising tool to optimize gas-solid flow dynamics in large-scale dense particulate systems. (C) 2020 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.