A finite element model of a human chest, modified THUMS model, was validated by means of mechanical post mortem human subject table top and indenter tests. The tissue conditions for the table top tests were eviscerated, denuded and intact. The loading configurations were diagonal belt, hub, distributed and criss‐cross belt. The validation was carried out by comparing predicted chest deflections and posterior reaction force by measured deflections and forces. The indenter tests were carried out on the denuded tissue condition. For the indenter tests the validation was carried out by comparing predicted chest deflections at various locations with measured post mortem human subject deflections.

In the table top validation the chest effective stiffness predicted with the model for the various tissue and loading conditions were all within the 95% confidence interval of the mechanical test results. For the model evaluation by means of the indenter tests generally the correct displacement trend was predicted.

The thorax model was used to carry out a force and energy analysis. For the whole thorax the greatest amount of energy for the diagonal belt, distributed and criss‐cross loading conditions was absorbed by the skin and fat. For the hub loading condition the greatest amount of energy was absorbed by the internal tissues of the chest. For the ribs the greatest amount of energy was for the intact thorax loaded by the diagonal belt and the primary energy‐absorbing structures were ribs 8 – 10 and the left clavicle. For the distributed loading condition, the greatest amount of energy was absorbed in left and right ribs 8. For the hub loading condition, the greatest amount of energy was absorbed by left and right ribs 2 – 4. For the criss‐cross loading condition, the greatest amount of energy was absorbed by left and right ribs 8 – 10.