Exploration of The Parameter Space in Frontal Vehicle/Pedestrian Collision and Effects on The Human Body

Exploration of The Parameter Space in Frontal Vehicle/Pedestrian Collision and Effects on The Human Body. Biomed Sci Abstract The purpose of this study is to investigate effect of the pedestrian’s height and weight, and its position during a traffic accident. An initial literature review was conducted to gather accident data in the UK involving pedestrians and a thorough research on pedestrian’s kinematics and injury severity that may arise during an accident. A 2010 model Toyota Yaris was the vehicle of choice for this study which was initially meshed using the 5mm element

The purpose of this study is to investigate effect of the pedestrian's height and weight, and its position during a traffic accident. An initial literature review was conducted to gather accident data in the UK involving pedestrians and a thorough research on pedestrian's kinematics and injury severity that may arise during an accident. A 2010 model Toyota Yaris was the vehicle of choice for this study which was initially meshed using the 5mm element crash criteria found in the Altair Hypermesh library. Additionally, hybrid III dummy was utilised which was scaled to heights of 170cm, 185cm, and 200cm with respective weights of 75kg, 82kg, and 90kg to evaluate the effects of the pedestrian's height and weight to the severity of the injuries. LS Dyna, along with Hypergraph 2D was used in order to conduct an explicit analysis which is suitable to computing rigid bodies and large displacements which were present within the crash simulation.
The simulation process was carried out to give results on 3 different dummies in a stationary and walking stance impacting the front end of the vehicle at 25mph in 3 different positions (total of 18 simulations). The 3 positions the dummy was placed along the front of the vehicle are Position 1 (P1) which is at the centreline of the vehicle, Position 2 (P2) is 466mm on the left-hand side of the vehicle, and Position 3 (P3) is 728mm on the left-hand corner of the vehicle. The data acquired for this thesis are the pedestrian head acceleration in order to evaluate head injuries and lower extremity loads to find out the severity of the lower leg injuries sustained during an accident. From the results, it can be clearly seen that the height, position and the stance of the pedestrian played an important role to the severity level of injuries during an accident. At a stationary position, the pedestrian at all 3 heights shown to sustain greater lower extremity injuries in comparison to when the pedestrian is at a walking stance. Additionally, when the collision occurs at the corner of the vehicle, the data shown that taller pedestrians (185cm and 200cm) while walking were observed to have a lower probability of sustaining severe head injuries as the HIC values were within the acceptable threshold of 1000.

Accident Data
The Department for Transport publishes annual and quarterly data on reported road accident casualties and the data from the report was acquired using the STATS19 reporting system where the information was given by the police that was on the scene of the accidents and the reports from the members of the public. However, the data/statistics obtained using this system do not represent the full accident/casualties occurring as people does not often report all personal injury accidents to the police. Road accidents in 2017, 24,831 in which have serious injuries, 144,369 with minor injuries, and 1,793 reported road deaths. Figure 1 shows the data of casualties in Great Britain in 2017 due to traffic accidents. In 2017, pedestrians accounted for 26% of road deaths with a total of 470 deaths which was 5% more than the recorded pedestrian deaths in 2016.

Pedestrian Injury
In most cases, during a road traffic accident, the injury to the pedestrian are caused by being impacted by the vehicle and the pedestrian hitting the ground. Figure 2 shows the distribution of injuries to the pedestrian that are caused by various vehicle regions. The head, the pelvic area, and the legs of the pedestrians are the body regions that are most likely to be injured with the head trauma being the main cause of most serious injuries and even death to the pedestrians. The initial contact between the pedestrian and the vehicle occurs at the bumper which can lead damage to the knee ligaments bending at the joint and fractures to the shin/tibia. The bumper is the part of the vehicle where its kinetic energy is initially transferred to the pedestrian during the crash which causes the injuries, and this is the reason that the bumper is an important safety feature in vehicles. The bumper, if designed properly, is capable absorbing some of the kinetic energy through deformation which reduces the damage to the pedestrian.
Looking at the sources of injuries (right), Figure 2 indicates that the bumper and the windscreen are the two main areas of the vehicle that are responsible for pedestrian injuries, however, 31.9% of injuries are due to the secondary impact to the ground In order to classify and investigate the severity of injuries, the AIS system (Abbreviated Injury Scale) was introduced (NSW, 2019) and is a system that classifies the relative risk of 'threat to life' in a person who was injured in an accident. AIS divides these risks into severity levels [11][12][13][14][15][16][17][18][19][20]: The interval that the collision last (t1 and t2) is from 15ms to 36ms which produced the maximal HIC and a(t) is the resultant linear acceleration of time history (in g) of the centre of gravity of the head.
The assessment of the HIC is important to the highway and airline safety community as both industry standards will need HIC evaluation for certification (SAE, 1966 andNHTSA, 1971). According to the Federal Motor Vehicle Safety Standard (FMVSS) 208 and FMVSS 213 for occupant crash protection and child restraint systems respectively, specify that the HIC value must be less than 1000 which there is an 18% chance of a severe injury, 55% of a serious injury and a 90% chance of a moderate injury. Furthermore, HIC is related to the AIS injury criteria which assess the probability of skull fracture using the calculated HIC value. Equation (2) is the formula calculating the risk of AIS 4+ injury (Mertz, 1996) and Equation (3) is the risk of AIS 3+ injury (NHTSA, 1995). The femur load criterion is a measure of injury to the femur.
It is the compression force acting axially on each femur and the threshold limit according to FMVSSS 208 and NCAP (New Car Assessment Programme) is 10kN for the 50th percentile male which corresponds to a 35% probability of AIS 2+ injury occurring.
The AIS formula involving the femur are shown below: The probability of an AIS 2+ injury is calculated by:

Methodology
The primary requirement for this research was to conduct a pedestrian crash analysis using software simulations to identify the injuries made on the pedestrian during a collision. This section of the thesis will explain the methodology used to conduct this research which consists of the simulation set-up, process of gathering injury data, and analysis and evaluation of the results.

Vehicle Model
The initial phase of this work was to generate a complete finite element model (FEM) of a small utility car which was used alongside the dummy to conduct a car to pedestrian crash analysis. For this research a Toyota Yaris (2010) was used. The vehicle model was initially meshed at approximately 5mm using Altair Hyperworks and since only a frontal collision test will be conducted, the back end of the vehicle was removed leaving the front section of the car as shown in Figure 3 below. This was done in order to reduce the duration of the simulation. To be able to get a more 'realistic' representation of a collision between a pedestrian and a car, the underlying components at the front end of the car was included such as the bumper beam, engine, radiator, and support frames. It was identified that as the pedestrian collides with the vehicle, the outer parts such as the hood and the bumper also have contact with the inner components (Ashton, n.d). The dummy used in this project is a Hybrid III 50 th percentile dummy. The hybrid III dummy is the most widely used dummy for both frontal crash and automotive safety testing. It is used by engineers in order to observe the human kinematics during a collision and to gather injury data which will allow an accurate prediction of the severity of potential injuries. For this project, a total of 3 dummies at heights of 170cm, 185cm, and 200cm with respective weights of 75kg, 82kg, and 90kg were used to evaluate the effects of the pedestrian's height and weight to the severity of the injuries (Figures 4-9). The 3 dummies are:

Design of Experiment
LS Dyna, a finite element software was used to carry out the simulations as it incorporates the used of either explicit, or implicit analysis to simulate complex real-life problems. For this project, the explicit analysis was used because an explicit computation has the capability to compute rigid bodies and large displacements which are likely to be present within a crash simulation, while implicit analysis is limited to handle these different variables. The simulation process was carried out to give results on 3 different dummies in a stationary and walking stance impacting the front end of the vehicle at 25mph in 3 different positions. Figure 6 shows the height of the dummies used for this study relative to the vehicle's bumper height which is around 200mm from the ground and was not changed all throughout the investigation.
In real-life situations, traffic accidents involving pedestrians usually occurs when the vehicle is making a turn. In this study, the dummy was positioned accordingly into 3 different places along the front of the car model. Figure  will be further discussed in section 3.0.

Results
The results presented in this study have been referred in this section. The process of acquiring and evaluating the injury data are as follows

Head Injury Criteria
The value of the HIC had been acquired by generating the head acceleration graph which was obtained through an accelerometer inside the dummy's head. The accelerometer is located at the head's centre of gravity and is labelled as node 1. The output (binout) file from the simulation was then utilised in Hypergraph 2D to generate an acceleration vs time graph, and then a data filtering method CFC180 SAE was applied in order to reduce any unwanted 'noise' in the data. Using the equation explained in section 5.2.1 and the injury tool from Hypergraph 2D, the HIC value was obtained and evaluated using the criteria shown in Table 2. The assessment process consists of colour-coding each HIC value in terms of its injury level, for example, if the HIC value was more than 1000 but less than 1700, the value in the results table will be highlighted in yellow indicating that the pedestrian will suffer severe head injuries [21][22][23][24][25][26][27][28][29][30][31][32][33][34].   (Tables 3 & 4) in which each value is highlighted into different colours indicating its severity level. For instance, when the femur load acting on the pedestrian was found to be more than 25kN, the value will be highlighted in red which indicates that fatal injuries will occur. Additionally, if the tibia load was found to be less than 5 kN, the value will be highlighted in green which signifies that the pedestrian will only suffer moderate injuries.   force which is shown in the data acquired in Table 5 above.         By using the assessment method mentioned earlier in this section, the HIC values shown in Table 6   Lower Extremity Injury-170cm, 75kg From the data acquired (Figures 14 & 15), it can be observed that when the dummy is standing still (stationary), the right femur  in higher forces acting on the lower right side of the dummy which was proven in the data acquired. Figure 17 shows the simulation of the dummy walking at position P2. It can be clearly seen that there's two points of contact occurring in the lower leg during the collision with the vehicle in which the right leg was initially hit at t=0.04s followed by an impact to the left leg at t=0.05s. ( Table   6) is the assessment of the probability of injury severity for the femur for both stationary and walking scenarios. The values are colour-coded in relation to its severity level as explained earlier in this section. When the dummy is stationary, fatal (red) lower leg injuries will occur as the left and right femur loads were more than 25kN. Moreover, when the dummy is walking, at P2 it experiences serious injuries (yellow) in both right and left leg, at P1 it will suffer both moderate (green) and serious injuries in the left and right leg respectively, and finally at P3 similar to P1 the dummy will experience both moderate and serious injuries. (Table 7) below shows the probability of injury severity for the tibia for the stationary and walking scenarios. When the dummy was in position P3, only moderate injuries will occur as highlighted in green. At position P2 serious injuries (yellow) occur, but at P1 fatal (red) injuries occur in the right leg as the tibia load was more than 15kN.  The graph below Figure 18 shows the Head Injury Criteria value on the head during the accident for the second dummy at 185cm  Tables 8 & 9 were evaluated and colour-coded in terms of its severity level. At P2 for both stationary and walking scenarios, the dummy suffers moderate head injury, P1 and P3 for stationary the dummy experiences severe head injury, and a fatal injury occurs to the dummy when walking at P1. (Figures 21 & 22) shows the head acceleration graph of the dummy during the accident for both stationary and walking simulations which was used to calculate the HIC values.

Lower Extremity Injuries-185cm, 82kg
The results when the dummy is stationary were similar to the first dummy (170cm) where the right femur and tibia loads are  Figure   25 shows the initial contact between the vehicle and the stationary dummy when the collision occurs at the centre of the vehicle front.
Just like the short dummy (170cm), the right leg was hit first (at t=0.03s) which immediately makes contact with the left leg. This results in higher forces acting on the lower right side of the dummy which proven the plausibility of the data acquired. Figure 26 shows the simulation of the dummy walking at position P2. It can be clearly seen that there's two points of contact occurring in the lower leg during the collision with the vehicle in which the right leg was initially hit at t=0.04s followed by an impact to the left leg at t=0.05s. A probability table to investigate the severity of injuries was created as shown in Table 9 for the femur. When the dummy is stationary, it will suffer fatal injuries (red) as the left and right femur loads were more than 25kN. However, when the dummy is walking, there's chance of moderate and serious injuries in the left and right leg at all positions. Table 10 below shows the probability of injury severity for the tibia for the stationary and walking scenarios. When the dummy was in position P3, there will be moderate injuries when the dummy is stationary and a chance of serious injuries when the dummy is walking, at position P2. At P1 fatal injuries will occur in the right leg as the tibia load was more than 10kN.    Table 11 shows that at all the scenarios that the dummy is        Figure 34 shows the initial contact between the vehicle and the stationary dummy when the collision occurs at the centre of the vehicle front. An initial contact was made to the right leg, at t=0.03s, which immediately makes contact with the left leg resulting in higher load acting on the lower right side of the dummy which is shown in the data acquired. Figure 34 shows the simulation of the dummy walking at position P2. There's two points of contact occurring in the lower leg during the collision with the vehicle in which the right leg was initially hit at t=0.03s followed by an impact to the left leg at t=0.04s. (Table 12) depicts the injury severity of the femur for both stationary and walking simulations. When the dummy is stationary, it will suffer fatal injuries when placed in positions P1 and P2 as the left and right femur loads were more than 25kN but severe injuries will occur when the dummy is in P3. However, in all the collision positions when the dummy is in a walking stance only moderate injuries will occur as the loads are below 10Kn (Table 13)

Discussion
The main purpose of this research is to conduct a series of    Figures a-e).

Conclusion
To conclude, a total of 18 simulations were completed for this study which involves simulating car to pedestrian collisions at 3 different pedestrian heights and 3 different positions. The methodology for this research was conveyed which consists of the simulation set-up, process of gathering injury data, and analysis and the evaluation of the results using the injury severity probability assessment process shown in the results section of this thesis.
From the data acquired it has been proven that the pedestrian's, height, weight and stance by played a big role to the outcome of an accident. It can be clearly seen that the pedestrian suffers minor injuries while walking in comparison to when the pedestrian is at a stationary position. Problems that were raised during the generation of the results were the fact that the data on the neck injuries were not generated due to insufficient information to calculate the neck injuries. Additionally, running the simulations using local server was found not ideal for this study due to the long duration in order to run the simulation. Instead a High-Performance Computer (HPC) was used to run all the simulations as provides the processing power to finish the simulations as quick as possible.

Future Scope
After the conclusion of this report, more research could be undertaken to extend the scope of the project: a) Performing simulations at different vehicle speeds i.e. 30mph and 40mph.
b) Velocity can be applied to the dummy to simulate walking and running scenarios and evaluate its effect on injury severity.

c)
Examination of other damage criterion in car to pedestrian collisions that has not been referred in this dissertation such as the neck and thoracic injury criterion.
d) Utilising THUMS model to conduct the accident simulations as it can simulate many aspects of the human form, from skin and bone to muscle tissue and internal organs, which will allow a better understanding of the real injuries people might suffer in a traffic accident.
Utilising THUMS model to conduct the accident simulations as it can simulate many aspects of the human form, from skin and bone to muscle tissue and internal organs, which will allow a better understanding of the real injuries people might suffer in a traffic accident.