The experiments presented in this thesis provide novel insight into two scarcely studied areas in the field of lateral hip impact biomechanics. The high energy nature of hip impacts requires high sampling rates for accurate study of hip impact dynamics. However, to date only optical motion capture, with relatively lower sampling rates (240-400 Hz), has been used to measure pelvic deflection during hip impact experiments with human participants. As such, the results from the first study compared the differences between two measurement systems (3D optical motion tracking and 2D high speed videography) in measuring common variables of impact biomechanics (peak force, time to peak force, peak deflection, time to peak deflection and energy absorbed). Although significant differences were seen between systems in measuring TFmax and Emax, the magnitude of differences were at or below 5% of the total magnitude of each measured variable. Furthermore, averaging impacts within a subject reduced the differences between systems for Emax. Furthermore, this study showed the effect of sampling rate on measuring hip impact dynamics, and how sampling at lower frequencies affects the aforementioned variables. Tests on the effect of sampling rate found differential effects contingent on the dependent variable measured. Sampling as low at 300 Hz, significantly reduced measures of Fmax and Dmax, but only by on average 0.7 and 0.5 %, respectively. Whereas measures of TFmax and TDmax increased by on average 9.5 and 6.8 %. Sampling Emax at 500 Hz and 300 Hz increased measures of impact absorption by 2.2 and 2.8 % respectively. Sampling at 4500 Hz was the lowest sampling rate that was not significantly different from 9000 Hz across all dependent variables. The second study in this thesis investigates the influence of contact area on load distribution during lateral hip impacts. In summary, this study shows that all three time-varying signals (Ft, FTt and Dt ) were significantly correlated with time-varying contact area (Ct). These results lend support to the possibility of modeling lateral hip impacts with contact models, but provide little support for a Hertzian model adaptation. Analysis on the relationships between body mass and BMI found both anthropometric measures to correlate significantly with peak impact force, but not with peak impact force directed to the greater trochanter. These results bring into question the feasibility of modeling hip fracture risk with body mass or BMI as inputs, without further investigating the distribution of impact force to the greater trochanter. In this study only contact area was significantly correlated with all measures of GT specific loading, and has never before been implemented in predictive modelling of hip fracture risk. Finally, this study found that although effective mass, total body mass and BMI were significantly correlated with the contact area at peak force, they only accounted for 21, 22 and 33% of the variance in CA. Altogether, this study sheds new light on the role that contact area plays in lateral hip impact loading and the importance of understanding load distribution during lateral hip impacts. It also highlights the importance of moving towards predictive models that incorporate more robust estimate of body composition and geometry, with hopes that these will better help estimate the risk of hip fracture. Overall, this thesis provides insight into the expected differences between measuring hip impact dynamics with two, relatively different measurement techniques. In addition, it highlights the need for further study on the relationship between contact geometry and hip fracture risk, something not currently implemented in most hip fracture risk models.

Since the introduction of helmets the incidence of traumatic brain injuries (TBI) in ice hockey has greatly decreased, but the incidence of concussions has essentially remained unchanged. Despite goaltenders in ice hockey being the only players on the ice for the entire game, few have assessed the performance of ice hockey goaltender masks. In ice hockey, goaltenders are exposed to impacts from collisions, falls and projectiles. The objective of this study was to assess the protective capacity of ice hockey goaltender masks for three accident events associated with concussion. A helmeted and unhelmeted medium NOCSAE headform were tested under conditions representing three common accident events in ice hockey. Falls were reconstructed using a monorail drop. A pneumatic linear impactor was used to reconstruct collisions and projectile impacts were reconstructed using a pneumatic puck launcher. Three impact locations and three velocities were selected for each accident event based on video analysis of real world concussive events. Peak resultant linear acceleration, peak resultant rotational acceleration and rotational velocity of the headform were measured. The University College Dublin Brain Trauma Model (UCDBTM) was used to calculate maximum principal strain (MPS) and von Mises stress in the cerebrum. The results demonstrated the importance of assessing the protective capacity of ice hockey goaltenders masks for each accident, as each event created a unique response. A comparison of unhelmeted and helmeted impacts revealed ice hockey goaltender masks are effective at reducing the risk of both concussion and TBI for falls and projectiles, but less so for collisions. Further, the risk of more serious injuries was found to increase for falls and collisions as impact velocity increased. The results highlight the importance of impacting multiple locations when assessing the protective capacity of ice hockey goaltenders masks, as different impact locations result in unique responses. Overall this study demonstrated ice hockey goaltenders masks capacity to reduce the risk of concussion across three accident events.

The recent emergence of electric scooter (e-scooter) ride share companies has greatly increased the use of e-scooters in cities around the world. In this thesis, firstly, e-scooter injuries reported in the current literature as well as an overview of current e-scooter company policies, state laws, and local laws are reviewed. The most injured regions of the body were the head and extremities. These injuries are generally minor to moderate in severity and commonly include fractures and lacerations. A primary cause of e-scooter accidents is front wheel collisions with a vertical surface such as a curb or object, generically referred to as a "stopper." Therefore, various e-scooter-stopper crashes were simulated numerically across different impact speeds, approach angles, and stopper heights to characterize their influence on rider injury risk during falls. A finite element (FE) model of a standing Hybrid III anthropomorphic test device was used as the rider model after being calibrated against certification test data. The angle of approach was found to have the greatest effect on injury risk to the rider, and it was shown to be positively correlated with injury risk. Smaller approach angles were shown to cause the rider to land on their side, while larger approach angles caused the rider to land on their head and chest. Additionally, arm bracing was shown to reduce the risk of serious injury in two thirds of the impact scenarios. The majority of e-scooter rider fatalities (about 80%) are recorded in collisions between a car and an e-scooter. Therefore, crashes between an e-scooter and a sedan (FCR) and a sports utility vehicle (SUV) were simulated using finite element models. The vehicles impacted the e-scooter at a speed of 30 km/hr in a perpendicular collision and at 15 degrees towards the vehicle, to simulate a rider being struck by a turning vehicle. The risks of serious injury to the rider were low for the head, brain, and neck, but femur/tibia fractures were observed in all simulations. The primary cause of head and brain injuries was found to be the head-ground impact if such an impact occurred. / Master of Science / The recent emergence of electric scooter (e-scooter) ride share companies has greatly increased the use of e-scooters in cities around the world. In this thesis, firstly, e-scooter injuries reported in the current literature as well as an overview of current e-scooter company policies, state laws, and local laws are reviewed. The most injured regions of the body were the head and extremities. These injuries are generally minor to moderate in severity and commonly include fractures and lacerations. A primary cause of e-scooter accidents is front wheel collisions with a vertical surface such as a curb or object, generically referred to as a "stopper." Therefore, various e-scooter-stopper crashes were simulated numerically across different impact speeds, approach angles, and stopper heights to characterize their influence on rider injury risk during falls. A finite element (FE) model of a standing Hybrid III anthropomorphic test device was used as the rider model after being calibrated against certification test data. The angle of approach was found to have the greatest effect on injury risk to the rider, and it was shown to be positively correlated with injury risk. Smaller approach angles were shown to cause the rider to land on their side, while larger approach angles caused the rider to land on their head and chest. Additionally, arm bracing was shown to reduce the risk of serious injury in two thirds of the impact scenarios. The majority of e-scooter rider fatalities (about 80%) are recorded in collisions between a car and an e-scooter. Therefore, crashes between an e-scooter and a sedan (FCR) and a sports utility vehicle (SUV) were simulated using finite element models. The vehicles impacted the e-scooter at a speed of 30 km/hr in a perpendicular collision and at 15 degrees towards the vehicle, to simulate a rider being struck by a turning vehicle. The risks of serious injury to the rider were low for the head, brain, and neck, but femur/tibia fractures were observed in all simulations. The primary cause of head and brain injuries was found to be the head-ground impact if such an impact occurred.

Anthropometric test devices (ATDs), commonly referred to as crash test dummies, are tools used to conduct aerospace and spaceflight safety evaluations. Finite element (FE) analysis provides an effective complement to these evaluations. In this work a FE model of the Test Device for Human Occupant Restraint (THOR) dummy was developed, calibrated, and validated for use in aerospace and spaceflight impact analysis. A previously developed THOR FE model was first evaluated under spinal loading. The FE model was then updated to reflect recent updates made to the THOR dummy. A novel calibration methodology was developed to improve both kinematic and kinetic responses of the updated model in various THOR dummy certification tests. The updated THOR FE model was then calibrated and validated under spaceflight loading conditions and used to asses THOR dummy biofidelity. Results demonstrate that the FE model performs well under spinal loading and predicts injury criteria values close to those recorded in testing. Material parameter optimization of the updated model was shown to greatly improve its response. The validated THOR-FE model indicated good dummy biofidelity relative to human volunteer data under spinal loading, but limited biofidelity under frontal loading. The calibration methodology developed in this work is proven as an effective tool for improving dummy model response. Results shown by the dummy model developed in this study recommends its use in future aerospace and spaceflight impact simulations. In addition the biofidelity analysis suggests future improvements to the THOR dummy for spaceflight and aerospace analysis. / Master of Science

The pedestrian is one of the most vulnerable road users. According to the World Health Organization (WHO), traffic accidents cause about 1.34 million fatalities annually across the world. This is the eighth leading cause of death across all age groups. Among these fatalities, pedestrians represent 23% (world), 27% (Europe), 40% (Africa), 34% (Eastern Mediterranean), and 22% (Americas) of total traffic deaths. In the United States, approximately 6,227 pedestrians were killed in road crashes in 2018, the highest number in nearly three decades. To protect pedestrians during Car-to-Pedestrian Collisions (CPC), subsystem impact tests, using impactors corresponding to the pedestrian's head and upper/lower leg were included in regulations. However, these simple impact tests cannot capture the complex vehicle-pedestrian interaction, nor the pedestrian injury mechanisms, which are crucial to understanding pedestrian kinetics/kinematics responses in CPC accidents. Numerous variables influence injury variation during vehicle-pedestrian interactions, but current test procedures only require testing in the limited scenarios that mostly focus on the anthropometry of the 50th percentile male subject. This test procedure cannot be applied to real-world accidents nor the entire pedestrian population due to the incredibly specific nature of the testing. To better understand the injury mechanisms of pedestrians and improve the test protocols, more pre-impact variables should be considered in order to protect pedestrians in various accident scenarios. In this study, simplified finite element (FE) models corresponding to 5th percentile female (F05), 50th percentile male (M50), and 95th percentile male (M95) pedestrians were developed and validated in order to investigate the kinetics and kinematics of pedestrians in a cost-effective study. The model geometries were reconstructed from medical images and exterior scanned data corresponding to a small female, mid-sized male, and tall male volunteers, respectively. These models were validated based on post mortem human surrogate (PMHS) test data under various loading including valgus bending at knee joint, lateral/anterior-lateral impact at shoulder, pelvis, thorax, and abdomen, and lateral impact during CPC. Overall, the kinetic/kinematic responses predicted by the pedestrian FE models showed good agreement against the corresponding PMHS test data. To predict injuries from the tissue level up to the full-body, detailed pedestrian models, including sophisticated musculoskeletal system and internal organs, were developed and validated as well. Similar validations were performed on the detailed pedestrian models and showed high-biofidelic responses against the PMHS test data. After model development and validation, the effect of pre-impact variables, such as anthropometry, pedestrian posture, and vehicle type in CPC impacts were investigated in different impact scenarios. The M50-PS model's posture was modified to replicate pedestrian gait posture. Five models were developed to demonstrate pedestrian posture in 0, 20, 40, 60, and 80 % of the gait cycle. In a sensitivity study, the 50th percentile male pedestrian simplified (M50-PS) model in gait predicted various kinematic responses as well as the injury outcomes in CPC impact with different vehicle type. The pedestrian FE models developed in this work have the capability to reproduce the kinetic/kinematic responses of pedestrians and to predict injury outcomes in various CPC impact scenarios. Therefore, this work could be used to improve the design of new vehicles and current pedestrian test procedures, which eventually may reduce pedestrian fatalities in traffic accidents. / Doctor of Philosophy / The pedestrian is one of the most vulnerable road users. According to the World Health Organization, traffic accidents cause about 1.34 million fatalities annually across the world. This is the eighth leading cause of death across all age groups. Among these fatalities, pedestrians represent 23% (world), 27% (Europe), 40% (Africa), 34% (Eastern Mediterranean), and 22% (Americas) of total traffic deaths. In the United States, approximately 6,227 pedestrians were killed in road crashes in 2018, the highest number in nearly three decades. To protect pedestrians in traffic accidents, subsystem impact tests, using impactors corresponding to the pedestrian’s head and upper/lower leg were included in regulations. However, these simple impact tests cannot capture the complex vehicle-pedestrian interaction, nor the pedestrian injury mechanisms, which are crucial to understanding pedestrian kinetics/kinematics responses in traffic accidents. Numerous variables influence injury variation during vehicle-pedestrian interactions, but current test procedures only require testing in the limited scenarios that mostly focus on the anthropometry of the average male subject. This test procedure cannot be applied to real-world accidents nor the entire pedestrian population due to the incredibly specific nature of the testing. To better understand the injury mechanisms of pedestrians and improve the test protocols, more pre-impact variables should be considered in order to protect pedestrians in various accident scenarios. In this study, simplified pedestrian computational models corresponding to small female, average male, and large male pedestrians were developed and validated in order to investigate the kinetics and kinematics of pedestrians in a cost-effective study. Overall, the kinetic/kinematic responses predicted by the pedestrian models showed good agreement against the corresponding test data. To predict injuries from the tissue level up to the full-body, detailed pedestrian computational models, including sophisticated musculoskeletal system and internal organs, were developed and validated as well. Similar validations were performed on the detailed pedestrian models and showed high-biofidelic responses against the test data. After model development and validation, the pre-impact variables were examined using the average male pedestrian model, which was modified the position to replicate pedestrian gait posture. In a sensitivity study, the average male pedestrian model in gait predicted various kinematic responses as well as the injury outcomes in lateral impact with different vehicle types. The pedestrian models developed in this work have the capability to reproduce the kinetic/kinematic responses of pedestrian and to predict injury outcomes in various pedestrian impact scenarios. Therefore, this work could be used to improve the design of new vehicles and current pedestrian test procedures, which eventually many reduce pedestrian fatalities in traffic accidents.

Car collisions are the third leading cause of unintentional death and injury among children aged 5 to 14. The pedestrian lower-extremity represents the most frequently injured body region in car-to-pedestrian accidents. Several sub-system tests (head, upper and lower legs) were developed for pedestrian protection in Asia and Europe. However, with exception of a child headform impact test, all other subsystem tests are designed for prediction of adult pedestrian injuries. Due to differences in impact location and material properties, existing subsystem tests and dummies designed for adult pedestrian cannot be used for child pedestrian protection by simple scaling. Thus, the development of a computational child pedestrian model could be a better alternative that characterizes the whole-body response of vehicle-pedestrian interactions and assesses the pedestrian injuries. Although several computational models for child pedestrian were developed in MADYMO/LS-DYNA, each has limitations. Children differ structurally from adults in several ways, which are critical to addressing before studying pediatric pedestrian protection. To aid in the development of accurate pediatric models, child pedestrian lower-extremity data presented in literature were first summarized. This review includes common pedestrian injuries, anatomy, anthropometry, structural and mechanical properties. A Finite Element (FE) model corresponding to a six-year-old child pedestrian (GHBMC 6YO-PS) was developed in LS-DYNA. The model was obtained by linear scaling an existing adult model corresponding to 5th percentile female anthropometry to an average six-year-old child's overall anthropometry taken from literature, and then by morphing to the final target geometry. Initially, the material properties of an adult model were assigned to the child model, and then were updated based on pediatric data during the model validation. Since the lower extremity injuries are the most common injuries in pedestrian accidents, the model validation focus on the pelvis and lower extremity regions. Three-point bending test simulations were performed on the femur and tibia and the results were compared to Post-Mortem Human Subject (PMHS) data. The knee model was also simulated under valgus bending, the primary injury mechanism of the knee under lateral loading. Then, the whole pedestrian model was simulated in lateral impact simulation and its response was compared to PMHS data. Finally, the stability of the child model was tested in a series of pediatric Car-to-Pedestrian Collision (CPC) with pre-impact velocities ranging from 20 km/h up to 60 km/h. Overall, the lower extremity and pelvis models showed biofidelity against PMHS data in component simulations. The stiffness and fracture FE responses showed a good match to PMHS data reported in the literature. The knee model predicted common ligament injuries observed in PMHS tests and a lower bending stiffness than adult data. The pelvis impact force predicted by the child model showed a similar trend with PMHS test data as well. The whole pedestrian model was stable during CPC simulations. In addition, the most common injuries observed in pedestrian accidents including fractures of lower limb bones and ruptures of knee ligaments were predicted by the model. The child model was accepted to be used according to Euro-NCAP protocol, so it will be used by safety researchers in the design of front ends of new vehicles in order to increase pedestrian protection of children. / Master of Science / Car collisions are the third leading cause of unintentional death and injury among children aged 5 to 14. The pedestrian lower-extremity represents the most frequently injured body region in car-to-pedestrian accidents. Several tests focusing on head, upper and lower legs were developed for pedestrian protection in Asia and Europe. However, with exception of a child headform impact test, all other subsystem tests are designed for prediction of adult pedestrian injuries. Due to differences in impact location and material properties, existing subsystem tests and dummies designed for adult pedestrian cannot be used for child pedestrian protection by simple scaling. Thus, the development of a computational child pedestrian model could be a better alternative that characterizes the whole-body response of vehicle–pedestrian interactions and assesses the pedestrian injuries. Although several computational models for child pedestrian were developed in MADYMO/LS-DYNA software, each has limitations. Children differ structurally from adults in several ways, which are critical to address before studying pediatric pedestrian protection. To aid in the development of accurate pediatric models, child pedestrian lower-extremity data presented in literature were first summarized. This review includes common pedestrian injuries, anatomy, anthropometry, structural and mechanical properties. A Finite Element (FE) model corresponding to a six-year-old child pedestrian (GHBMC 6YO-PS) was developed in LS-DYNA. The model was obtained by linear scaling an existing adult model corresponding to 5th percentile female anthropometry to an average six-year-old child’s overall anthropometry taken from literature, and then by morphing to the final target geometry. Initially, the material properties of an adult model were assigned to the child model, and then were updated based on pediatric data during the model validation. Since the lower extremity injuries are the most common injuries in pedestrian accidents, the model validation focus on the pelvis and lower extremity regions. Three-point bending test simulations were performed on the femur and tibia and the results were compared to Post-Mortem Human Subject (PMHS) data. The knee model v was also simulated under valgus bending, the primary injury mechanism of the knee under lateral loading. Then, the whole pedestrian model was simulated in lateral impact simulation and its response was compared to PMHS data. Finally, the stability of the child model was tested in a series of pediatric Car-to-Pedestrian Collision (CPC) with pre-impact velocities ranging from 20 km/h up to 60 km/h. Overall, the lower extremity and pelvis models showed biofidelity against PMHS data in component simulations. The stiffness and fracture FE responses showed a good match to PMHS data reported in the literature. The knee model predicted common ligament injuries observed in PMHS tests and a lower bending stiffness than adult data. The pelvis impact force predicted by the child model showed a similar trend with PMHS test data as well. The whole pedestrian model was stable during CPC simulations. In addition, the most common injuries observed in pedestrian accidents including fractures of lower limb bones and ruptures of knee ligaments were predicted by the model. The child model was accepted to be used according to Euro-NCAP protocol, so it will be used by safety researchers in the design of front ends of new vehicles in order to increase pedestrian protection of children.

Fall-related hip fractures are a substantial public health issue. Unfortunately, little is known about whether the effective stiffness of the pelvis, a critical component governing impact force during lateral falls, differs substantially across different segments of the population. The objective of this thesis was to enhance the knowledge base surrounding pelvis impact dynamics by assessing the influence of gender and body mass index (BMI) on the effective stiffness of the pelvis, and on resulting peak loads applied to the hip, during sideways falls. Towards this end I conducted pelvis release trials (in which the pelvis was suspended and suddenly released onto a force plate) with males and females with low (<22) and high (>28) BMIs. One resonance-based (kvibe), and three force-deflection based (k1st, kcombo 300, and kcombo opt) methods of effective pelvic stiffness estimation were examined. The resulting stiffness estimates, and peak forces sustained during the pelvis release experiments, were compared between each BMI and sex group. The optimized force-deflection stiffness estimation method, kcombo opt provided the strongest fit to the experimental data. Strong main effects of BMI (f (1,13) = 10.87, p = 0.003) and sex (f (1,13) = 5.97, p = 0.022) were found for this stiffness estimation method. Additionally, a significant BMI-sex interaction was observed (f (3,6) = 5.31, p = 0.030), with low BMI males having much higher stiffness estimates than any other group. Normalized peak forces were higher in low BMI participants than in high BMI participants (f(1,13)=24.9, p<0.001). Linear regression demonstrated that peak impact force was positively associated with effective pelvic stiffness (β = 0.550, t(25) = 3.110, p=0.005), height (β = 0.326, t(25) = 2.119, p=0.045) and soft tissue thickness (β = 0.785, t(25) = 4.573, p<0.001). This thesis has demonstrated that body habitus and sex have significant effects on the stiffness of the pelvis during lateral falls. These differences are likely related to a combination of soft tissue and pelvic anatomical differences between BMI and sex groups. Pelvic stiffness, along with other easily collected variables, may be helpful in predicting peak forces resulting from lateral falls in the elderly. Differences in pelvic stiffness estimates between BMI and sex groups, and estimation method, necessitate careful consideration. These data will aid in selecting the most appropriate pelvic stiffness parameters when modeling impact dynamics for higher energy falls.

Fall-related hip fractures are a substantial public health issue. Unfortunately, little is known about whether the effective stiffness of the pelvis, a critical component governing impact force during lateral falls, differs substantially across different segments of the population. The objective of this thesis was to enhance the knowledge base surrounding pelvis impact dynamics by assessing the influence of gender and body mass index (BMI) on the effective stiffness of the pelvis, and on resulting peak loads applied to the hip, during sideways falls. Towards this end I conducted pelvis release trials (in which the pelvis was suspended and suddenly released onto a force plate) with males and females with low (<22) and high (>28) BMIs. One resonance-based (kvibe), and three force-deflection based (k1st, kcombo 300, and kcombo opt) methods of effective pelvic stiffness estimation were examined. The resulting stiffness estimates, and peak forces sustained during the pelvis release experiments, were compared between each BMI and sex group. The optimized force-deflection stiffness estimation method, kcombo opt provided the strongest fit to the experimental data. Strong main effects of BMI (f (1,13) = 10.87, p = 0.003) and sex (f (1,13) = 5.97, p = 0.022) were found for this stiffness estimation method. Additionally, a significant BMI-sex interaction was observed (f (3,6) = 5.31, p = 0.030), with low BMI males having much higher stiffness estimates than any other group. Normalized peak forces were higher in low BMI participants than in high BMI participants (f(1,13)=24.9, p<0.001). Linear regression demonstrated that peak impact force was positively associated with effective pelvic stiffness (β = 0.550, t(25) = 3.110, p=0.005), height (β = 0.326, t(25) = 2.119, p=0.045) and soft tissue thickness (β = 0.785, t(25) = 4.573, p<0.001). This thesis has demonstrated that body habitus and sex have significant effects on the stiffness of the pelvis during lateral falls. These differences are likely related to a combination of soft tissue and pelvic anatomical differences between BMI and sex groups. Pelvic stiffness, along with other easily collected variables, may be helpful in predicting peak forces resulting from lateral falls in the elderly. Differences in pelvic stiffness estimates between BMI and sex groups, and estimation method, necessitate careful consideration. These data will aid in selecting the most appropriate pelvic stiffness parameters when modeling impact dynamics for higher energy falls.

Les blessures de l'abdomen représentent une faible proportion (5%) des blessures lors d'accidents de la route mais elle augmente fortement pour les blessures sérieuses à sévères (16%). L'abdomen du mannequin THOR (Test Device for Human Occupant Restraint), destiné aux futures réglementations de choc frontal, nécessite des améliorations de sa biofidélité et un critère de blessure. Le travail présenté est en trois parties :Premièrement, les paramètres principaux de la réponse mécanique de l'abdomen du THOR et de Sujets Humain Post Mortem (SHPM) sous chargements impacteur et ceinture furent identifiés à l'aide d'un modèle mécanique simplifié. La comparaison des paramètres mécaniques du THOR et des SHPM a mis en évidence les changements nécessaires pour l'amélioration de la biofidélité. Il apparaît que la viscosité équivalente du THOR doit être augmentée d'un facteur 5 et que l'interaction avec la pièce bassin doit être modifiée du fait qu'elle augmentait la rigidité d'un facteur 8. Ces changements furent inclus dans le modèle Éléments Finis (EF) d'un abdomen prototype incluant des capteurs de pression APTS (Abdominal Pressure Twin Sensors) pour caractériser le chargement de l'abdomen.Deuxièmement, la réponse mécanique du prototype a été évaluée en simulations d'essais charriot, ce qui a montré que l'abdomen prototype a peu d'influence sur la cinématique globale du mannequin mais que la flexion du tronc peut faire augmenter la pression dans les APTS. Cela a mené à des recommandations supplémentaires au niveau de la conception de l'abdomen.Finalement, en vue de définir un critère de blessure pour l'abdomen, la pression des APTS a été corrélée aux blessures des organes décrites dans les études sur SHPM de la littérature ou prédites par le modèle EF humain THUMS / Abdominal injuries represent a small proportion (5%) of road crash injuries but their proportion increases considerably with regard to serious and severe injuries (16%). The abdomen of the Test device for Human Occupant Restraint (THOR), intended to be used in future frontal impact assessments, needs further developments regarding its biofidelity and injury criterion. The work performed in this thesis project was in three folds: Firstly, the main parameters of the THOR and Post Mortem Human Subjects (PMHS) abdomen responses under impactor and seatbelt loadings were identified using a lumped element model. The comparison between the THOR and the PMHS mechanical parameters highlighted desired changes for THOR abdomen biofidelity improvement. It was found that THOR material viscosity should be increased by 5 and that interaction with the pelvis flesh should be modified as it increased by 8 the abdomen stiffness. These changes were included in the Finite Element (FE) model of an existing abdomen prototype which is equipped with Abdominal Pressure Twin Sensors (APTS) to quantify the abdomen load. Secondly, the response of the prototype was evaluated in sled test simulations which showed that the prototype abdomen had little influence on the dummy overall kinematics but that the torso flexion could increase the pressure in the APTS. This led to additional recommendations regarding the abdomen design. Finally, for the abdominal injury criterion definition, the APTS pressure was correlated with organ injuries as reported in published PMHS tests or as predicted by THUMS human FE model

In the United States pedestrian fatalities in vehicle impacts have increased over the last 40 years and pedestrians who use wheelchairs (seated pedestrians) have higher mortality rates than standing pedestrians in vehicle impacts. Standing pedestrian protection has generated increased attention and regulatory action but seated pedestrian protection has not been investigated or regulated. To investigate standing pedestrian safety researchers use finite element models of the human body and simulate vehicle impacts. Finite element models can be useful but they are limited by their biofidelity, and often simplify the complex anatomy of the human body for the sake of computational expense. If modeling results are to be taken seriously to investigate standing and seated pedestrian protection, then further model development and validation is necessary. In this dissertation a finite element model of a male 50th percentile standing pedestrian was enhanced and validated for use in vehicle impact simulations. The standing pedestrian model lower body was further enhanced and validated to study the importance of stabilizing components of the knee. These updates to the standing pedestrian knee joint were imported into an occupant model and further validated in occupant loading scenarios. The updated standing pedestrian was used to explore the effect of modeling component failure on vehicle impact. Simplified and detailed occupant models were used to model seated pedestrians in vehicle impacts to explore seated pedestrian injury risks. The seated pedestrian head and brain typically reported the highest risks of injury, usually because of head-ground contact. A lap belt, airbag vest, and bicycle helmet were tested on the seated pedestrians. The lap belt and airbag vest typically increased injury risks and the bicycle helmet reduced injury risks. The work presented in this dissertation may inform future modelers, vehicle designers, and safety equipment developers on standing and seated pedestrian safety. / Doctor of Philosophy / In the United States pedestrian fatalities in vehicle impacts have increased over the last 40 years and pedestrians who use wheelchairs (seated pedestrians) have higher death rates than standing pedestrians in vehicle impacts. Research studies have examined how to protect standing pedestrians, but not seated pedestrians. The goal of this work was to begin investigating seated pedestrian safety. To investigate standing pedestrian safety researchers use computer models (finite element models) of the human body and simulate vehicle impacts. These finite element models can be useful but they are limited by how life like they are. If modeling results are to be taken seriously to investigate standing and seated pedestrian protection, then further model improvement is necessary. In this dissertation a finite element model of an average North American male standing pedestrian was improved for use in vehicle impact simulations. The standing pedestrian model lower body was further improved to study the importance of stabilizing components of the knee. These updates to the standing pedestrian knee joint were imported into a seated model with the same anatomy. Simplified and detailed seated models were used to model seated pedestrians in vehicle impacts to explore seated pedestrian injury risks. The seated pedestrian head and brain typically reported the highest risks of injury, usually because of head-ground contact. A lap belt, airbag vest, and bicycle helmet were tested on the seated pedestrians. The lap belt and airbag vest typically increased injury risks and the bicycle helmet reduced injury risks. The work presented in this dissertation may inform future modelers, vehicle designers, and safety equipment developers on standing and seated pedestrian safety.