A sudden change in height and velocity which demarcates a supercritical flow from a sub-critical flow in a steady free-surface flow is called a standing jump. They have been well-studied for hydraulic flows for years (see [1] and references therein). Some studies investigated such standing jumps in flows of dry granular materials down an incline [2-4]. In a recent study [5], we have proposed a general relation, based on the depth-averaged mass and momentum equations, able to predict the height of standing jumps. This general relation was carefully checked thanks to laboratory data for different boundary conditions (roughness and slope of the bottom, discharge), and two kinds of fluids (water, dry grains). However this general equation needs additional knowledge, for instance on the length of the jump and the effective friction inside the jump, to become fully predictive in any case. A numerical model based on Discrete Element Method (DEM) has been developed, which recreates standing granular jumps in the same conditions as the recent laboratory tests conducted with granular materials [6]. This numerical experiment allows varying many parameters: the roughness and the slope angle of the bottom, and the discharge like in the laboratory tests, but also other variables, like the grain diameter, the inter-particle friction or the restitution coefficient of particles. This also allows measuring precisely variables hard to obtain with laboratory tests, such as the local density within the flow or the friction between the flow and the bottom. DEM simulations are used in order to study specifically the influence of each macro and micro parameter on the geometry of the standing granular jump. It is a way to check the relevance of the general relation for the jump height proposed in [5] for a wider range of boundary conditions and mechanical properties of grains. Moreover, a careful attention is given to the development of the closure relations needed, but unknown a priori, for that general relation. Fed with those new closure relations, the theoretical relation becomes fully predictive and directly usable, for example in the design of avalanche protection dams.