A human body model based method for injury risk prediction considering occupant stature and posture

Abstract

This study developed a framework to generate injury prediction functions in order to evaluate the safety of occupants in traffic accidents, using human body model simulations to analyse the influences of crash severity, occupant stature, occupant posture, and restraint stiffness. The THUMS AM50-O Human Body Model (HBM) was used in a sled test setup. Six influencing parameters were included in injury prediction: impact velocity, occupant stature, Body Mass Index (BMI), torso angle, seating position, and seatbelt restraint force. Impact velocity was found as a significant influencing factor on skull injury, while BMI mainly affected chest injury. Sequential sampling method was used for generation of simulation matrix, and high-order interaction terms were identified. As injury prediction functions, quadratic regressions correlate better with the simulation results than other forms of regressions, the combinations of parameters in which also enhanced the robustness of injury prediction when considering uncertainty of input parameters. Using numerical models to generate injury prediction functions can better characterize occupant variations, such as sitting posture, than using traffic accident reconstruction. Such obtained functions can provide reasonable estimation on occupant injury risk based on accident information, and we also hope it may benefit vehicle safety improvement and triage protocol in medical treatments of traffic accident victims.

Keywords:

• Traffic injury prediction
• occupant diversity
• morphing
• finite element simulation
• parametric modelling

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was supported by National Key R&D Program of China (Grant Number 2017YFE0118400).