Parametric study of booster seat design characteristics

Forman, Jason / Miller, Matthew / Perez-Rapela, Daniel / Gepner, Bronislaw / Edwards, Marcy A. / Jermakian, Jessica S.
Insurance Institute for Highway Safety
February 2022

Objective: Belt-positioning booster seats seek to improve the fit of a seat belt around a child’s pelvis and shoulder. There are a wide variety of booster designs on the market, and it is unclear what characteristics of booster design influence the interaction between a child occupant and belt restraint system during a collision. The objective of this study was to use large-scale, automated, parametric simulations with the PIPER 6-year-old model to study the effects of various booster design characteristics (along with posture and belt anchor geometry) on child occupant response and restraint interaction during simulated collisions.
Methods: In this study, we used the PIPER finite element model of a 6-year-old human to study restraint interactions with genericized booster seat models capturing the fundamental geometries and stiffnesses currently available on the market. Other study parameters included the posture of the occupant (upright, moderately slouched, and slouched) and the lap-belt anchor geometry. The key outcome of interest was the phenomenon of submarining, where the pelvis slides under the lap belt and the belt loads directly into the abdomen. We used an automated iterative simulation sampling approach, combined with neural network metamodeling, to continually sample and simulate the study space until our predicted precision converged to a steady state. Just over 700 simulations were performed to quantify a study space comprised of 17 variables describing booster geometry and stiffness, along with occupant posture and belt anchor geometry.
Results: According to our simulations, the most influential parameters in predicting submarining risk were the stiffness of the booster and the posture of the child. Simulations with boosters of traditional high-stiffness construction (similar to stiff plastic) and with the child model seated in an upright posture showed a reduced risk of submarining. Simulations with a low-stiffness booster (similar to an inflatable booster) almost always resulted in submarining, regardless of the other aspects of the booster design. A slouched posture also increased the risk of submarining, for both low-stiffness and high-stiffness boosters.
Conclusions: These results suggest that it is nearly impossible to design a low-stiffness booster that will provide robust protection against submarining. In these simulations, low-stiffness boosters almost always resulted in the pelvis sliding below the lap belt, with the lap belt loading into the abdomen. In addition, these results suggest that posture is also critical to kinematics and restraint interaction, with slouched postures resulting in an increased risk of submarining. Thus, these results suggest that boosters that are of a high stiffness and promote an upright posture may provide protection against submarining.