Magnesium is known as one of the lowest density metals. With the increasing importance of fuel economy and the need to reduce weight, magnesium has proven to be a very important structural material used in transportation industry. However, the use of magnesium alloys have been limited by its tendency to corrode, creep at high temperature, and higher cost compare to aluminium alloys and steels. Polycrystal plasticity models such as VPSC and EVPSC were used to study deformation mechanisms of magnesium alloys. However, current polycrystal plasticity models with slip and twinning involve a large number of material parameters, which may not be uniquely determined. Furthermore, determining material parameters using traditional trial-and-error approach is very time consuming. Therefore, a genetic algorithm approach is developed in this thesis to optimize these material parameters. The genetic algorithm approach is evaluated by studying large strain behavior of magnesium alloys under different deformation processes. The material parameters for those models are determined by material numerical simulations based on the polycrystal model to the corresponding experimental data. Then the material parameters are used to make prediction of other deformation behaviours (stress strain curves, R values, texture evolution and lattice strain), and the performance is judged by how well the prediction match the actual experimental data. The results show that the genetic algorithm approach works well on determining parameters, it can get reliable results within a relatively short period of time. / Thesis / Master of Applied Science (MASc)