Simulations of top-hat thin-walled sections of dual-phase steel DP800 subjected to axial crushing have been performed taking into account process history and measured geometric imperfections, thickness variations and material variations. The simulations were based on experiments performed by Fyllingen et al. [Fyllingen, Ø., Hopperstad, O.S., Langseth, M., 2008. Robustness study on the behaviour of top-hat thin-walled high-strength steel sections subjected to axial crushing. International Journal of Impact Engineering, in press, doi:10.1016/j.ijimpeng.2008.03.005], who investigated the robustness of a top-hat section subjected to axial crushing. The geometry variation and spatial strain hardening variation were mapped onto the model. The fracture parameter and strain-rate sensitivity were based on values obtained from one of the batches. It was emphasised to use an element type, element size, a fracture criterion and a spot-weld model typically used by the industry. Compared to nominal models especially the thickness variations, geometric imperfections and material failure criterion influenced the behaviour. The material batch variation resulted in large differences in the batch means of the mean crushing forces and the variation in the geometric imperfections and thickness resulted in variation in the mean crushing force within each batch. Compared to the experiments the model generally under-predicted the mean crushing force.