Digital prototypes are increasingly used for early prediction of vehicle functional design to ensure compliance with regulatory, consumer and internal requirements. Future assessments must take relevant snow load cases into account. Snow is becoming more and more of a focus throughout the automobile industry, especially for electric vehicles. Among others, there are massive snow loads in the front-end and underbody areas, due to the missing waste heat from the combustion engine. Snow also plays an important role in autonomous driving, as it can impair sensors and thus affect driving safety. It is a challenge to model and predict the material behavior of snow, because there exist numerous different types of snow, each of them changing its complex behavior in response to even the smallest change of conditions. The goal of this presentation is to describe this considerable number of snow types using meaningful material parameters and to create a thermodynamically consistent 3D continuum model that describes the complex behavior of snow. In this contribution, an approach is presented that simulates snow as a visco-hypoelasto-plastic continuum. For this purpose, an existing complex material model for snow is integrated into a collocation method: generalised finite differences (GFD). The approach of using a collocation method has several advantages over classical mesh-based and particle-based methods of snow simulation: There is no need of project data from the grid to particles and vice versa, no need of remeshing, boundary conditions can naturally be integrated, and refinement as well as discrete consistency is straightforward. In addition to the theory, applications of different snow loadings are also presented.