Among the whole population, small, obese, and/or older occupants are at increased risk of death and serious injury in motor-vehicle crashes compared with mid-size young men. Current adult finite element (FE) human body models (HBM) have been developed in a few body sizes (large male, midsize male, and small female) with reference body dimensions similar to those of the available physical anthropomorphic test devices (ATDs). The limited number of body sizes available has resulted in part because the time needed to develop an FE HBM using typical methods is measured in months or even years. The objective of the current study was to apply a recently developed FE HBM morphing method to generate hundreds of FE human models for occupants with a wide range of stature and body shape and using the diverse human models for impact simulations.

The midsize male THUMS and GHBMC models were used as the baseline models to be morphed into occupants with different combinations of stature and body shape. The target geometries were predicted using statistical geometry models of external body shape and the skeleton (ribcage, pelvis, femur and tibia) developed previously based on 3D body scan and CT data from a total of more than 500 subjects. A landmark-based radial basis function (RBF) interpolator was used to morph the baseline models into target geometries. Anthropometric targets for 112 men were sampled based on US population statistics for age, stature and body mass index (BMI). Using these targets, 100 HBMs were developed by morphing THUMS and 12 by morphing the GHBMC model. Pendulum thorax impact conditions were applied to 36 morphed THUMS models and 12 morphed GHBMC models to investigate effects of occupant characteristics on chest impact responses.

The morphed models were all automatically generated without any manual adjustment, and their mesh quality was reasonable and suitable for impact simulations. The mesh morphing process required about 10-30 minutes per model on a contemporary PC. Peak impact forces and chest deflections in the chest pendulum impact simulations varied substantially with different models, confirming the need to consider population variation in evaluating the occupant responses. The age, stature, BMI, and weight effects on chest impact responses were found to be complex but consistent between the morphed THUMS and GHBMC models. The method developed in this study can help future safety designs for occupants with a wide range of stature and body shape.