A multi-scale validation of the capability of the SAFER human body model (v9) to predict the risk for an occupant to sustain two or more rib fractures in vehicle crashes was carried out. The rib fracture risk was evaluated by means of a probabilistic rib fracture prediction method. A variety of loading conditions was evaluated, from published lab tests with post mortem human subjects (PMHS) to detailed accident reconstructions and population-based reconstructions. The PMHS load cases were table-top, impactor and sled tests. The detailed accident reconstructions included 20 occupants involved in real-world crashes. For the population-based reconstructions more than 100 simulations with a generic vehicle interior model were carried out. Parameters regarding both the generic model and the occupant were varied in the population-based simulations. The predicted risk for an occupant to sustain two or more rib fractures was evaluated for the PMHS sled reconstructions as well as for the detailed and population-based reconstructions. The predicted 2 or more rib fracture risk was compared to the actual number of fractured ribs sustained by the PMHS and the occupants. Generally, two or more fractured ribs observed in the PMHS tests, the vehicle crashes and NASS data were successfully predicted with the model