Evaluation of occupant neck injury response to varied impact conditions using a finite element-mathematical surrogate modeling approach

This study presents car crash-induced neck injury trends in response to variations in three impact variables: velocity (10 - 45 mph; 16.1 - 72.4 km/h), location (front, rear, near side, and far side), and angle (-45 degrees to 45 degrees). By employing a combined finite element (FE)-mathematical surrogate modeling approach, the number of necessary FE crash simulations was significantly reduced. Each motor vehicle collision (MVC) case was simulated with LS-DYNA software, and the extracted neck injury metrics (Nij, Nkm, and Lateral Nij) were used to train Kriging surrogate models. These models produced clear response trends indicating that increased impact velocity and locational proximity to the driver often resulted in the greatest risk of inducing neck injury. Further, the impact angle variable typically produced the greatest risk under direct or slightly oblique angles regardless of the impact location or whether the impact was directed toward the occupant. This lack of interaction between the angle and location variables was likely caused by limitations within the injury metrics, themselves, which do not account for multi-axial or oblique loading, highlighting the need for improved metrics. Finally, distinct pairings of sub-metric component responses were noted in most impact scenarios. These trends may be a signifier for the primary modality of injury and should be assessed in a future study.