In order to develop a tire that can run safely even on snowy and icy road surface, it is necessary to elucidate the complex friction phenomenon with prediction technologies. This paper introduces prediction technologies for winter tire development. Finite Element Method and Finite Volume Method are used to model tire and snow, respectively. Snow deformation is calculated by the Eulerian formulation to solve the complex interaction between snow and tire tread pattern. In order to represent the shear characteristics of snow, the Mohr-Coulomb plasticity model is used. This prediction technology is used to improve snow traction performance through pattern design. In addition, a crushable foam material and Discrete Element Method are applied to model the mechanism of snow deformation. Thermal fluid-structure interaction analysis considering phase change using particle-based simulation is applied to elucidate the friction characteristics of rubber on ice. Ice and water are modelled using Lagrangian particle-based method, and a phase change model is introduced to consider ice melting. The method predicts that the coefficient of friction decreases due to low drainage performance when the number of sipes is small.