Modeling of ever increasingly complex industrial processes requires insatiable computing power. To keep reasonable calculation times, one way is to decompose the domain into smaller ones and solve each subdomain on independent processors. The message passing interface (MPI) protocol and the ParMetis library are two helpful tools to achieve such a task. The paper presents an algorithm that uses both tools to parallelize ICE3D, a finite volume scheme for Ice accretion on aircraft, a module of the well-known code FENSAP-ICE. The parallelized code is used to calculate ice accretion on large-scale problems such as wings and complete aircraft.