Analysis of Abdominal Injury Mechanisms in Frontal Sled Simulations Using a Human Body Model

Yu-Ki Higuchi (JARI)･Mitsutoshi Masuda (JAMA)･Fusako Sato (JARI)

Abdominal injuries are a critical injury outcome for reducing fatalities and serious injuries in road traffic crashes; however, assessment methods using anthropomorphic test devices (ATDs) have not yet been sufficiently established. Toward the development of abdominal injury assessment procedures, this study analyzed the ITARDA Micro Data, an in-depth accident investigation database in Japan, to identify frontal crash configurations and abdominal injury patterns. The associated injury mechanisms were then investigated through frontal sled simulations using the THUMS AF05 human body model.

A Human Body Model Based Sensitivity Analysis of Belt Routing Conditions in Frontal Impact Simulation

Kaito Nishimura･Takashi Yoshida･Shigeru Kotama･Nozomi Saito (Tokai Rika)

This study compared occupant responses under a standard belt routing condition and multiple alternative belt routing conditions in frontal impact simulation using the THUMS human body model under identical analysis conditions.

Analysis of Pelvic Rotation in Human Body Model During Frontal Impact

Toshiharu Azuma･Yuqing Zhao･Koji Mizuno (Nagoya University)･Kei Nagasaka･Takahiro Suzuki･Idemitsu Masuda (Suzuki Motor)

In frontal impacts, rearward pelvic rotation can cause the lap belt to slip off the pelvis, increasing the risk of submarining. To investigate this phenomenon, finite element simulations of rear-seat sled tests under frontal impact conditions were conducted to analyze the pelvic rotation that occurs under lap belt restraint. The moment time histories around the pelvic center of gravity were compared using human body models (THUMS) of an average male and a small female, clarifying the characteristics of the rotational behavior attributed to pelvic shape.

Prediction of the Occupant Lower-Extremity Injury Criterion (Tibia Index) in Frontal Collisions Using Machine Learning

Seiya Tanamoto (University of Yamanashi)･Kei Nagasaka･Idemitsu Masuda (Suzuki Motor)･Yuta Yokoyama (Diver Technology)･Hirofumi Sugiyama (University of Yamanashi)･Shigenobu Okazawa (University of Yamanashi/Diver Technology)

Lower-extremity injuries sustained by occupants in frontal collisions are difficult to predict because vehicle deceleration and dashboard deformation are intricately correlated. In this study, a machine learning model is developed to estimate the occupant lower-extremity injury criterion (Tibia Index) using dashboard deformation modes and acceleration pulse inputs in sled simulations. In addition, the influence of analysis cases on prediction accuracy is investigated. Furthermore, the validity of the proposed method is evaluated through comparison with numerical simulation results.

Relationship between the shape of the front of a passenger vehaicle and its energy absorption characteristics

Keisuke Fukuyama･Takahiro Isshiki･Koji Mikami (JARI)

In traffic accident analysis, energy absorption distribution diagrams are used to calculate the deformation energy of vehicles. Existing distribution diagrams show room for further improvement in their ability to adapt to the diverse front-end shapes of vehicles in recent years. To address this issue, this study clarifies the relationship between the front-end shape and energy absorption characteristics of recent vehicles.

Effectiveness Evaluation of an Age-Adjusted D-Call Net Algorithm for Fatal and Severe Injury Prediction

Shizue Katsumata･Noboru Tanase･Takahiro Andoh (Toyota Motor)･Mayu Ishii (Institute for Traffic Accident Research and Data Analysis)

In the current operation of D-Call Net, the fatal and severe injury rate is calculated assuming an occupant age of 65 years. In this study, accident data are linked with D-Call Net notification data to investigate the actual age distribution of occupants involved in accidents and its impact on injury severity estimates.