In metallic alloys, the formation of macrostructure involves different length and time scales. The macrostructure strongly influences the mechanical properties of the materials. Modeling of macrostructure provides valuable insights for tailoring of the macrostructure. In the present study simulation of macrostructure in Al-22Sn has been carried out. Expression for nucleation rate and growth rate were tuned to get a match with the experimentally observed macrostructure. The study incorporates the motion of grains due to Brownian motion. Incorporation of Brownian motion resulted in more uniform distribution locally, which has been experimentally observed. Both simulation and macrograph showed few columnar grains and absence of sharp columnar to equiaxed transition.