Head injury of Vietnamese pedestrian in crash accident with SUV using numerical simulation

Use your smartphone to scan this QR code and download this article ABSTRACT Crash test simulation using finite-element method is more andmore popular in the automobile industry because of its feasibility and cost saving. Themajority of finite element dummymodels used in crash simulation are built based on anthropometrical and biomechanical data of the USA and European bodies. Thus, it is necessary to develop a scaling algorithm to scale a reference dummy size into a desired one without rebuilding the entire model. In this paper, a brief review of scaling method to archive V-dummy is presented. V-dummy is scaled from the Hybrid III dummy model provided by LS-DYNA to suit Vietnamese anthropometry. In particular, head injury risk of a pedestrian involved in a frontal impact with a SUV is studied. Total of fifteen simulations were conducted to investigate the head injury by using HIC (Head Injury Criterion). When striking by a car, the head will hit on car bonnet first and fly in the air later before impacting to ground. The position of head impact on car strongly depends on impact angle, but landing posture on ground of pedestrian is not correlated with impact speed or angle. In the phase of car-impact, pedestrians suffer from a low level of head injury, whereas in the ground-impact, HIC exceeds the threshold of human tolerance except speed of 20 km/h. When pedestrian is hit by SUV at angle 90o , the shoulder strike on car bonnet before head, it causes HIC in impact angle 90o is lower than 0o and 45o . It cannot be recognized the correlation between HIC and impact speed in the phase of ground-impact.


INTRODUCTION
Traffic accidents still occur every day, every hour, taking away the health and lives of many people. Traffic accidents not only cause consequences for victims but also lead to many other corollaries, many families have to face difficulties, debts and deadlock. The rate of road traffic fatalities per 100,000 populations in South-East Asia still higher compared to the average rate of the world in 2013 and 2016 (Figure 1) 1 . Traffic accidents in Vietnam in 2018 decreased slightly compared to 2017, nationwide there were 18,232 traffic accidents causing 8,200 deaths as shown in Figure 2; 5,124 wounded people and 9,070 minor injuries as shown in Figure 3 1 -General Statistics Office of Vietnam. Compared to the previous year, traffic accidents decreased in all three criteria in 2018, the number of accidents decreased by 9.2%, the number of deaths decreased by 1.9%; the number of injured people decreased by 8.3% and the number of minor injuries decreased by 20.8%. Of the total of 9,446 traffic accidents from less serious or more, 9,202 cases (accounting for 97.4%) occurred on the road, causing 7,958 deaths and 5,060 injuries. In 2018, there were 50 traffic accidents per day throughout the country (5 cases down compared to 2017), including 26 traffic accidents from less serious and more and 24 traffic collisions, making 22 people died, 14 were injured and 25 were slightly injured. With the attempt to develop the government's transport infrastructure, the rate of deaths and injuries due to traffic accidents decreased each year.

RESEARCH METHODOLOGY
For the study of car-pedestrian crashes, it is two common methods that can be employed: conducting crash tests with mechanical dummies and simulating car crashes on computer. The former is a traditional way and gives good results compared with real life car impact; however, its disadvantage is very expensive test equipment and generally more time-consuming than the latter because after every crash test, experimental vehicles as well as dummies need repairing to be ready for the next experiments. On the other hand, with the advent of digital computers, car crash simulation using finite element method (FEM) becomes more common in car industries in these days because it opens a new modern way for engineers to run crash tests inside computers rather than on roads. In addition, not only does it save time and be less expensive   than real crash tests, it also gives designers and engineers many chances to modify and customize designs for each parts of the car without making changes to the whole models. Consequently, more virtual crash tests run, more insights engineers can gain to fully understand their design to give the best product. As a result, car crash simulation using FEM is a promising and suitable method for the research. For this study, LS-DYNA is chosen because of its capability to simulate highly nonlinear problems including car crashes. Hybrid III 50th dummy is popular use for human modal in kinematics analysis. Because this dummy is built based on fiftieth percentile male (equal in height and weight of the average North American), V-dummy was firstly developed in Ly et al. (2019) 2 by scaling algorithm presented in Hyncik (2007) 3 , Untaroiu (2007) 4 . Hybrid III 50 th dummy is transformed to a dummy model representing Vietnamese. Dummy and Vietnamese anthropometry are shown in Table 1. The scaling algorithm comprises three steps -Scaling of dummy geometry -Scaling of inertial properties -Scaling of joint properties

HEAD INJURY
Pedestrian head injuries are the main causes of pedestrian fatalities and disabilities in pedestrian to mo-tor vehicle collision 7 . The mechanisms and behaviors of pedestrian head in collision are unpredictable in real cases. In spite of the development of automotive safety industry, the only injury criteria in wide use is the Head Injury Criterion (HIC), which was developed in the 90s.

Head injury criterion
HIC was first idealized in 1961 by Gadd in his research. He also developed his criterion -Gadd severity index (GSI). After that, it was truly finalized by Versace (1971), which known as a function of average linear acceleration correlated to the Wayne State University tolerance curve. But it was first only published widely by the US National Highway Traffic Safety Administration (NHTSA) and is expressed as: where t 2 and t 1 : two arbitrary times during acceleration pulse. Linear acceleration a is a function of time (seconds), which measured in multiples of gravity acceleration (g's). The average linear acceleration _ a of a(t) between two phases t 2 and t 1 can be expressed as: And the head injury criterion (HIC) can be calculated as:

Science & Technology Development Journal -Engineering and Technology, 3(SI2):SI129-SI144
where HIC is the maximum value between the impact time t 2 and t 1 of the brackets {}, while the index 2.5 based on the real case accidents. Since the HIC index had an important part in automotive safety industry, there are still some limitations on HIC as an injury severity criterion such as: -Rotating acceleration of head is skipped.
-Only hard contacts are taken into account.

Scaling of Head Injury Criterion
Moreover, a simple method developed by Mertz et al. 8 to scale HIC for various dummy sizes. For the sake of convenience, three dimensionless fundamental ratios are Length scale ratio: Mass density ratio: λ m = ρ 1 ρ 2 Modulus of elasticity ratio: where L is where L is the characteristic length, ρ is density, and E is modulus of elasticity. Assume that the density and modulus of elasticity remain constant for two subjects, hence λ m = 1 and λ E = 1. Therefore, a scaling factor for HIC can be written as: This method has the pros and cons. The advantage is it is simple and straightforward to use, but because its derivation is based on the assumption that considering head impact to be represented as a singledegree-of-freedom spring-mass system, so it does not account for the complexity of a real-life impact. However, the main purpose for introducing this method here is to realize that the scale factor of HIC is inversely proportional to the length scale factor, which means the smaller in size, the greater in HIC. This fact will be used to verify results of simulations associated with the scaled dummy.

A proposed method for verification of the scaled dummy
In the scope of this study, it is difficult to verify the scaled dummy because there are no available experimental data or theoretical results associated with Vietnamese pedestrian-car crashes. Therefore, the verification of the scaled dummy is limited to evaluate qualitatively by using Eq. (4) that indicates the inverse proportion between HIC and dummy sizes. Therefore, the values of HIC of the scaled dummy are expected to be larger than those of the Hybrid III 50 th dummy.

Abbreviated injury scale
The Abbreviated Injury Scale (AIS) points out risk of fatality for a given injury level correlated to the head injury criterion (HIC). There are six injury levels from 1 (minor injuries) to 6 (fatal, non-survivable). The correlation between HIC and AIS is presented in Table 2.

RESULT VERIFICATION The Hybrid III 50 th dummy verification
In this section, due to the lack of available experimental data for SUV-pedestrian crashes, a simulation between the Hybrid III 50 th dummy and a sedan is done to compare experimental data 9 and simulation results. Besides, the computer simulation conducted in Elmasoudi (2015) 9 , namely paper simulation which performed pedestrian-sedan crashes, are also added for comparison. The acceleration results from simulation are filtered following SAE CFC1000 specification. Full model is shown in Figure 4. Termination time is set to 0.25 second, and 0.6 is chosen to compute stable time step between iterations. Two main types of contact used are automatic single surface for handling self-contact, and automatic surfaceto-surface for general contact. In terms of boundary conditions, automatic single surface is used for the whole car model, and automatic surface-to-surface is applied for contacts between the dummy and the car, four car tires and the ground, and the dummy and the ground. In this verification of the Hybrid III 50 th dummy, the car's velocity is 40 km/h, and both of the dummy and the car are subjected to gravity.  As can be seen Figure 5, the head accelerations in the x-direction and z-direction in the simulation have the same waveform as those in the experiment. The contact between the dummy's head and the car hood occurs roughly from 13 ms to 14 ms. In the simulation, while a peak acceleration in z-direction is seen at nearly 14ms, the head encounters a negative xcomponent acceleration at around 14 ms, and then it experiences a positive x acceleration. The accelerations in the physical test see the same pattern as those in the simulation but with higher peak intensity. This difference likely results from the discrepancy in the material properties of the car hood, wind shield, or even dummy head between the real model and the FE model. Another reason might be due to contact stiffness chosen during the simulation. The paper's simulation in Elmasoudi (2015) 9 seem quite fluctuating, and have the peak accelerations larger than the experiment ones. It is always good to verify the energy balance. In LS-DYNA, the energy balance can be checked using the energy ratio, which is the ratio between total energy and the sum of initial energy and external work. The total energy includes kinetic energy, internal energy, frictional energy, rigid wall energy, damping energy, and hourglass energy, whereas initial energy is composed of initial kinetic energy and initial internal energy. The energy balance should hold at all times during the analysis, or alternatively the energy ratio should remain as close as possible to one. As can be seen from Figure 6, the energy ratio in the simulation satisfies energy balance as it is nearly close to one with a small variation.

The V-dummy verification
A comparison of HIC between the Hybrid III 50 th dummy and the V-dummy for various cases will be made. Car model used for these simulations is C5000 Pickup. The magnitudes of velocity of the car are 40, 50, and 60 km/h, and three angles of impact is 0, 45, and 90, which are shown in Figure 7. Boundary conditions is the same as those in the verification of the SI133 Figure 5: Head acceleration in g's.
Hybrid III 50 th . Note that HIC in this verification is caused by the first impact between the dummy and the car. As illustrated in Table 3, HIC for the Hybrid III 50 th dummy in most cases is smaller than that for the Vdummy with the exception of the case of 40 km/hangle 0 where the trend is reversed. In general, the simulation results show the same pattern as Eq. (4) does; that is HIC is inversely proportional to dummy size. Here, a typical example is chosen to show the energy ratio and mass scaling during the simulation are described in Figure 8 and Figure 8 in the case of 50 km/h -angle 0 o respectively. It is proposed that the percentage of mass increase should be less than 1% of physical mass, and energy ratio close to 1. As seen from the figures, these criteria are well fulfilled. Note that the mass increase stems from mass scaling that is a technique in which nonphysical mass is added to a structure to obtain a larger time step size. Status of impact of the dummy with the Pickup is shown in Figure 10. It indicates that when colliding with the Pickup, the dummy's head comes into contact with the hood.

Vehicle modal is chosen in this research is SUV-2002
Ford Explorer 10 . The impact position and landing posture of pedestrian V-dummy when directly hit by SUV are illustrated in Table. 4. When striking by a car, the head of human body is hit on car bonnet which is name car-impact at first and finally impact SI134 Figure 6: Energy ratio during the simulation. to ground which is name ground-impact. These two phases of impact are demonstrated from Figure 11 to Figure 16. SUV, which also known as high ground clearance vehicles, at the case of side-on impact 90 o , the HIC at 40 km/s is much lower than any other cases. This can be reasoned that in the simulation, according to the key frames from Figure 12 to Figure 16, it shows that dummy's head did not hit perpendicularly to car bonnet in the car-impact phase but make a small sliding on it, which leads to a significant drop in con-  In the first phase of contact, car-impact, the trend of HIC is increased linearly for both factors -impact angle and initial velocity as described in Table 4 and drawn graphically in Figure 17. During the carimpact, the peak value of HIC is around 1200 (being directly hit -as impact angle equals 0 o , by high ground clearance vehicle), which may cause the skull fracture, but without severe neurological damages; no damages of spiral cord -according to the AIS table -Table. 2. Moreover, as demonstrated in Figure 17, there is an increase in HIC when the speed rises from 20 km/h to 60km/h. At the impact direction at 45 0 and 90 0 , the HIC is below 1000, which corresponds to AIS below 3. Of three angles, 0 o degree is the most dangerous, and it is followed by 45 o and 90 o degrees respectively.

Science & Technology Development Journal -Engineering and Technology, 3(SI2):SI129-SI144
In conclusion, if car velocity is below 40 km/h, victims just suffer from less-than-15-minute unconsciousness without involving any skull fracture, whereas if velocity is larger than 40 km/h, pedestrians will encounter more-than-15-minute unconsciousness.
For ground impact, pedestrian-to-ground collision, all of the head injury criterion exceeds the threshold of human tolerance except speed of 20 km/h. HIC is described in Table 5 and drawn graphically in Figure 18. It should be pointed out that since the ground is considered as an absolutely rigid body, contact forces it applies on the dummy is far larger than in the first impact. As a result, HIC in ground impact is very large, and it suggests that ground impact is the key cause of severe head injuries or even death.

CONCLUSION
In this study, a series of simulation is conducted to verify the Hybrid III 50 th with a comparison with experimental data, and qualitatively assess how well the results of scaled dummy simulations match with the anticipated trend. A set of fifteen crash simulations between SUV and V-dummy is performed, and the findings show that in the first impact, pedestrians are likely to suffer spinal injuries, whereas in the ground impact, head injuries are the principal cause of death. When pedestrian is hit by SUV at angle 90 o , shoulder strike on car bonnet before head, it causes HIC in impact angle 90 o is lower than the two others. HIC in case of pedestrian-to-ground collision exceeds the threshold of human tolerance except speed of 20 km/h. It cannot be recognized the correlation between HIC and impact speed in the phase of groundimpact.