Quantitative analysis of the protective performance of bicycle helmet with multi-direction impact protection system in oblique impact tests

Abstract

Purpose

The current study aimed to assess the protective performance of helmets equipped with multi-directional impact protection system (MIPS) under various oblique impact loads.

Methods

Initially, a finite element model of a bicycle helmet with MIPS was developed based on the scanned geometric parameters of an actual bicycle helmet. Subsequently, the validity of model was confirmed using the KASK WG11 oblique impact test method. Three different impact angles (30°, 45°, and 60°) and 2 varying impact speeds (5 m/s and 8 m/s) were employed in oblique tests to evaluate protective performance of MIPS in helmets, focusing on injury assessment parameters such as peak linear acceleration (PLA) and peak angular acceleration (PAA) of the head.

Results

The results demonstrated that in all impact simulations, both assessment parameters were lower during impact for helmets equipped with MIPS compared to those without. The PAA was consistently lower in the MIPS helmet group, whereas the difference in PLA was not significant in the no-MIPS helmet group. For instance, at an impact velocity of 8 m/s and a 30° inclined anvil, the MIPS helmet group exhibited a PAA of 3225 rad/s² and a PLA of 281 g. In contrast, the no-MIPS helmet group displayed a PAA of 8243 rad/s² and a PLA of 292 g. Generally, both PAA and PLA parameters decreased with the increase of anvil angles. At a 60° anvil angles, PAA and PLA values were 664 rad/s² and 20.7 g, respectively, reaching their minimum.

Conclusion

The findings indicated that helmets incorporating MIPS offer enhanced protection against various oblique impact loads. When assessing helmets for oblique impacts, the utilization of larger angle anvils and rear impacts might not adequately evaluate protective performance during an impact event. These findings will guide advancements in helmet design and the refinement of oblique impact test protocols.

1. Introduction

As environmental consciousness burgeons, the concept of green and low-carbon transportation gains momentum, leading more individuals to opt for bicycles in short-distance commuting. However, this surge in cycling has coincided with an increase in bicycle accidents. The 2018 Global Status of Road Safety reported 1.35 million fatalities worldwide due to road traffic accidents, among which 40,500 were cyclists.¹ In China alone, there were 53,200 bicycle accidents in 2020, resulting in 2774 fatalities and 8174 injuries.²

Previous studies indicate that head injuries are the primary cause of fatalities among bicyclists.^3,4 Helmets have been identified as effectively in reducing the risk of head injuries for cyclists, particularly in instances involving impacts with the ground.^3,4 Typically, the helmet liner material comprises expanded polystyrene (EPS), known for its ability to absorb energy during vehicle or ground impacts, thereby lessening the linear impact loads on the cyclist's head.5, 6, 7 However, during a vehicle collision, the cyclist's head is exposed to both linear and rotational forces. Conventional EPS-only liner helmets face challenges in effectively reducing rotational loads. Recent studies have revealed that incorporating multi-directional impact protection system (MIPS), a thin layer of low-friction plastic, between the EPS liner of the helmet and the wearer's head, is proved effectivly in diminishing rotational head motion.^8,9 This reduction in rotational motion is crucial as it generates angular acceleration, which often leads to shear deformation of brain tissue. Such deformation is recognized as the primary mechanism underlying brain tissue injuries7, 8, 9, 10, including diffuse axonal injury and concussion^6,8,9,11.

Bicycle helmets undergo safety testing according to different national standards, such as the 16CFR1203-2018 of the United States, the EN 1078-2012 of the European Union, the GB 24429-2009 of China, and the AS/NZS 2063-2008 of Australia and New Zealand. However, these standards do not encompass oblique impact testing. Studies have indicated that oblique impacts often produce higher levels of angular acceleration, potentially leading to concussions, intracranial hemorrhage, and neural injuries.12, 13, 14

Bonin et al.¹⁵ conducted helmet tests using inclined anvils set at angles of 30°, 45°, and 60° to investigate the impact of angles on head rotation injuries. Additionally, Baker et al.¹⁶ observed that during bicycle crashes, the typical impact speed ranges between 5 m/s and 8 m/s, with helmet impacts commonly occurring at the side, front, and rear positions. To assess the effectiveness of helmets protection in oblique collisions, the anvil angles and impact speeds serve as the basis for establishing an oblique impact simulation.

2. Methods

2.1. Finite element modeling of MIPS bicycle helmet

The study utilized one of the commonly available bicycle helmets featuring MIPS, as depicted in Fig. 1. This helmet is compliant with the GB 24429-2009 standard and comprises components, such as an outer shell with metallic properties, a detachable inner liner, MIPS pads, a foam liner, and straps.

Fig. 1.

Finite element model and physical object of MIPS helmet.

MIPS: multi-directional impact protection system.

The precise geometric model of a helmet was acquired using a HandySCAN 3D scanner. Subsequently, the helmet geometry underwent optimization for reverse reconstruction employing Geomagic Warp software. The resulting finalized helmet geometry model was imported into Hyperworks software, where a refined finite element helmet model was developed. The entire modeling process is illustrated in Fig. 2.

Fig. 2.

Finite element modeling process for bicycle helmets with MIPS.

MIPS: multi-directional impact protection system, STL: stereolithography, IGES: initial graphics exchange specification, FE: finite element.

In this paper, the material properties for the shell, foam liner, MIPS gel pad, and strapping material of the helmet model are derived from existing literature. It is noted that varying foam densities possess different energy absorption capacities, consequently influencing the protective capability of a helmet.¹⁷ In this paper, the MATA_63 material was used to represent the foam material, which has a foam density of 80 kg/m³. Table 1 presents the material parameters relevant to the model.

Table 1.

Helmet parts materials and other finite element model material parameters.

Finite element models	Authentic materials	Principal structure models	Densities ρ/g·cm−3	Elastic modulus E/MPa	Poisson's ratios	Reference (year)
Shell	Acrylonitrile butadiene styrene	MAT 24	1.055	3000	0.30	Teng et al.18 (2013)
Foam-lined	Expanded polystyrene	MAT 63	0.0125	5	0.01	Ouellet et al.19 (2006)
Lace-up	Polyethylene terephthalate	MAT 24	1.4	1000	0.44	Teng et al.18 (2013)
MIPS gel pads	polycarbonate	MAT 1	1.2	500	0.39	Mu et al.20 (2022) Ji et al.21 (2017)

MIPS: multi-directional impact protection system.

2.2. Verification of model construction

The model construction of the MIPS helmet coupled to the headform is shown in Fig. 3A. The MIPS helmet is covered by a notably smooth surface. Based on the research of Bonin et al.¹⁵, we model the interaction between the MIPS gel pad and the bicycle helmet liner as “face-to-face” contact (contact_automatic_surface_to_surface). To simulate the sliding effect during contact between MIPS and the helmet liner, we set the coefficient of friction between the MIPS and helmet foam at 0.17. The coupling between the helmet strap and the head model is done by defining the surface-to-surface contact¹⁷, which is shown in Fig. 3B. The correlation between different helmet components and their corresponding friction coefficients was established in a previous study conducted by Han et al.¹⁷ Considering the presence of an anti-slip pad on the inside of the MIPS and building upon the findings of Han et al.'s¹⁷ study, we set the friction coefficients between the headform and the MIPS, as well as between the helmet and the inclined anvil, at 0.5 and 0.55, respectively. For modeling purposes, the material of the international standard organization (ISO) headform was represented using MAT 20, with Young's modulus and Poisson's coefficient set at 200 GPa and 0.2, respectively.²² Similarly, the material for the inclined anvil was modeled using MAT 24, with Young's modulus and Poisson's coefficient set to 200 GPa and 0.3, respectively.²²

Fig. 3.

Verification model for the MIPS helmet-headform. (A) The model construction of the MIPS helmet coupled to the headform. (B) The surface-to-surface contact coupling between the helmet strap and the head model.

ISO: international standard organization; MIPS: multidirectional impact protection system.

The Italian helmet manufacturer KASK released its industry-standard rotational impact testing protocol for bicycle helmets on October 27, 2022, known as WG11 rotational impact test.²³ This test, derived from ECE 22.05, is specifically designed to evaluate helmet performance concerning rotational impacts. Fig. 4 illustrates the 4 designated impact positions aligned with ECE 22.05 standards: impact_0°, impact_135°, impact_180°, and impact_270°.

Fig. 4.

Bicycle helmet crash test area schematic calibration.

By the WG11 rotational impact test standard and ECE 22.05 requirements, 4 helmets were utilized for assessment. These helmets underwent drop tests on a 45° inclined anvil at a minimum speed of 6 m/s across 4 specified impact positions: impact_0°, impact_135°, impact_180°, and impact_270°. In the experiment, we measured angular acceleration by integrating an accelerometer into the ISO headform. The accelerometer was securely attached to the headform using a thread. Fig. 5 shows the simulated conditions, encompassing fall and boundary parameters used in impact simulations, aligning with the experimental drop tests conducted on the helmets.

Fig. 5.

Drop experiments and simulation conditions and boundary conditions of MIPS helmet.

MIPS: multi-directional impact protection system.

To assess the protective effectiveness of the bicycle helmet equipped with MIPS, the enhancement ratio of peak kinematic injury parameters between helmets with and without MIPS was determined. This ratio, denoted as ΔMIPS and represented by equation (1), quantifies the improvement in injury parameters attributed to MIPS. ΔMIPS_ peak linear acceleration (PLA) refers to the difference in PLA between helmets with and without MIPS, while ΔMIPS_ peak angular acceleration (PAA) signifies the difference in PAA in the presence and absence of the MIPS helmet.

$$\Delta MIPS=\{\begin{array}{c}{\Delta MIPS}_{PLA}=\frac{{PLA}_{noMIPS}-{PLA}_{MIPS}}{{PLA}_{noMIPS}}\\ {\Delta MIPS}_{PAA}=\frac{{PAA}_{noMIPS}-{PAA}_{MIPS}}{{PAA}_{noMIPS}}\end{array}$$ (1)

3. Results

3.1. Validation results of helmets with MIPS

Fig. 6 shows the linear and angular acceleration curves of the head's center of mass obtained from simulations and experiments conducted during helmet drop tests at the 4 helmet positions on the inclined anvil. It is noted that the kinematic parameters in the simulation were filtered at a frequency of 1000 Hz. A comprehensive and comparative analysis between the simulations and test results indicates significant agreement in the linear acceleration curves, angular acceleration curves, as well as in the peak values of both linear and angular acceleration at the 4 test positions (Table 2).

Fig. 6.

The results against validation of MIPS bicycle helmet of FE model. (A) Impact_0°, (B) Impact_135°, (C) Impact_180°, and (D) Impact_270°.

MIPS: multi-directional impact protection system; FE: finite element.

Table 2.

Helmets rotation experiments compared with simulation.

Test points	Parametric	Experiment values	Simulation values	Error ratio (%)
Impact_0°	PLA (g)	97	103	6.19
	PAA (rad/s2)	1810	1890	4.42
Impact_135°	PLA (g)	80	82	2.50
	PAA (rad/s2)	791	840	6.19
Impact_180°	PLA (g)	60	57	5.00
	PAA (rad/s2)	1054	1051	0.28
Impact_270°	PLA (g)	142	139	2.11
	PAA (rad/s2)	1192	1236	3.69

PLA: peak linear acceleration; PAA: peak angular acceleration.

3.2. Kinematics of helmets with and without MIPS in different oblique impact tests

Fig. 7 presents the kinematic behavior of headforms wearing helmets with and without MIPS, at impact velocities of 5 m/s and 6 m/s, in both frontal and oblique positions. The initial (0 ms) and final (25 ms) motion states are displayed in the kinematic representation of the oblique impact simulation. While the initial kinematics of the helmet-headform with and without MIPS exhibited similarity during the initial impact, noticeable differences emerged during the final phase of the impact.

Fig. 7.

Kinematics of the MIPS and no-MIPS helmet at front impact position with different impact velocities and different inclined anvils.

MIPS: multi-directional impact protection system.

Box plots of uniform size (180 mm × 160 mm) were utilized to visualize the motion patterns of the helmet-headform at specific time points (0 ms and 25 ms). This standardized representation enables a comprehensive assessment of kinematic variances occurring between these time intervals, offering a clear view of any differences in kinematics.

Among both the MIPS and no-MIPS helmet groups, increased impact velocity correlates directly with the amplified rotational motion of the headform. In particular, in the anvil 30° test, there was a significant difference in helmet-head morphology kinematics between impact velocities of 5 m/s and 8 m/s, resulting in greater rotational movement of the head morphology at higher speed. Substantial kinematic differences were observed between the MIPS and no-MIPS helmet groups at similar velocities. Headforms in no-MIPS helmets demonstrate pronounced rotational movement, while those using MIPS helmets effectively mitigate this motion. Specifically, at an impact velocity of 5 m/s on the 30° inclined anvil, the initial kinematics of the helmet-headform between MIPS and no-MIPS helmet were similar at 0 ms, but substantial differences at 25 ms.

In further exploration of variations at different impact positions in the oblique tests, Fig. 8 illustrates the kinematics of MIPS and no-MIPS helmet groups in the simulated oblique test at a 45° anvil angle and an impact velocity of 5 m/s. Differences in helmet-headform kinematic results in the no-MIPS helmet group were significant, particularly in the side and rear impact positions, where rotational motion was evident in the side position, resulting in motion completely beyond the visible region. These comparative outcomes indicated that more moderate helmet-headform kinematics and less severe rotational motion are displayed in the MIPS helmet group compared to the no-MIPS helmet groups.

Fig. 8.

Kinematics of the MIPS and no-MIPS helmet at 5 m/s impact velocity, anvil 45° of oblique impact test.

MIPS: multi-directional impact protection system.

3.3. Peak kinematic parameters of helmets with and without MIPS in different oblique impact tests

Based on the kinematic results, substantial differences are observed between the MIPS helmet group and the no-MIPS helmet group. To further investigate factors influencing peak kinematic parameters of the helmet groups with and without MIPS across various impact test conditions, the peak PLA and PAA values from 36 helmet tests in the oblique impact simulations were divided into 4 groups: MIPS and no-MIPS of impact velocities of either 5 m/s or 8 m/s. The statistical outcomes are shown in Fig. 9. The graph displays 4 categorized conditions for the MIPS and no-MIPS helmet groups. The light blue solid box represents the condition with an impact velocity of 5 m/s for the MIPS helmet group, while the dark blue dashed box represents the condition with an impact velocity of 8 m/s for the same group. The light green solid box represents the condition with an impact velocity of 5 m/s for the no-MIPS helmet group, while the dark green dashed box represents the condition with an impact velocity of 8 m/s for the same group.

Fig. 9.

PLA and PAA with and without MIPS helmet group under different oblique impact tests. (A) PAA and PLA on a 30° anvil, (B) PAA and PLA on a 45° anvil, and (C) PAA and PLA on a 60° anvil.

MIPS: multi-directional impact protection system; PAA: peak angular acceleration; PLA: peak linear acceleration.

For PAA, it is evident that higher impact velocities resulted in greater angular acceleration. Among the 3 impact positions (side, front, and rear), the lowest PAA occurs at the front impact position, which was 664 rad/s² in the 60° anvil impact. In terms of PLA, higher impact velocities also led to higher linear velocities. The lowest PLA was observed at the rear impact location, which was at 20.7 g in the 60° anvil impact.

It becomes apparent that for the 30° inclined anvil, both PLA and PAA are the highest, with PAA approximately below 10,000 rad/s² and PLA below 300 g. For the 45° inclined anvil, both PLA and PAA are the next highest, with below 8000 rad/s² and 250 g, respectively. For the 60° inclined anvil, PLA and PAA are the lowest, with PAA below 5000 rad/s² and PLA below 100 g.

In this current paper, to quantitatively evaluate the effectiveness of MIPS within the same type of helmet, we calculated ΔMIPS for both PLA and PAA parameters across all working conditions, following the ΔMIPS calculation method defined in the Method section. The results were then categorized into 3 groups based on different anvil types: anvil 30°, anvil 45°, and anvil 60°.

The quantitative results reveal that in all simulation outcomes, there is an average reduction of approximately 54.8% in average PAA, while average PLA shows a reduction of about 15.6% (Fig. 10). This highlights the efficacy of MIPS technology in decreasing rotational kinematic parameter, demonstrating an average reduction of 54.8%. Conversely, the effectiveness of linear kinematic parameter is comparatively lower, resulting in an average reduction of PLA approximately 15.6%. Therefore, MIPS technology exhibits greater efficiency in mitigating rotational impacts to the head, particularly in reducing angular acceleration, while showing slightly less effectiveness in mitigating linear acceleration.

Fig. 10.

ΔMIPS on the 30°, 45°, and 60°inclined anvil.

MIPS: multi-directional impact protection system; PRA: peak rotational acceleration; PLA: peak linear acceleration.

4. Discussion

The kinematics of helmet-headform interactions are significantly influenced by the conditions of oblique impact tests. Bliven et al.¹³ have delved into helmet protection under varying inclined angles of the anvil. Our research aligns with these considerations by incorporating impact angles into our oblique impact tests.

In Fig. 11, we present a line graph illustrating PAA and PLA for categorized groups, where each data point encapsulates the outcome of a specific simulation condition. The graphs reveal a consistent trend: both PLA and PAA values exhibit a diminishing pattern as the anvil angle increases. Additionally, it is evident that the PLA and PAA values for helmets equipped MIPS are consistently lower compared to no-MIPS equipped.

Fig. 11.

PAA and PLA on the 30°, 45°, and 60° inclined anvil.

MIPS: multi-directional impact protection system; PAA: peak angular acceleration; PLA: peak linear acceleration.

This observation can be attributed to 2 primary factors. Firstly, the alteration in the anvil angle changes load distribution, subsequently influencing subsequent helmet-headform kinematics. Secondly, the incorporation of MIPS technology dissipates energy from short-duration, high-linear impacts through rotational effects. This cushioning effect enhances the involvement of the helmet foam in the energy absorption process, optimizing its capacity to absorb energy and resulting in a reduction of the high peak values without MIPS.

Baker et al.¹⁶ have reported that head impact velocities typically range between 5 m/s and 16 m/s, with accidents concentrating between 5 m/s and 8 m/s. The angle of impact is usually between 10° and 80°, with a concentration of accidents between 30° and 50°. Their findings also highlight the front and side areas as more frequently impacted. Fig. 9, Fig. 10 integrate results based on the angle of inclination of the anvil and impact velocities and show that both PLA and PAA values exhibit systematic changes across different impact positions of the helmet (side, front, and rear) in the absence of MIPS-equipped helmets. Specifically, PLA values decrease as the impact position changes, reaching their lowest values at the rear position. This trend is consistent across inclined anvils at 30°, 45°, and 60° and becomes more pronounced, especially at higher impact velocities. Significantly, substantial PLA changes are evident at impact positions on a 30° inclined anvil and at an impact velocity of 8 m/s. This observation suggests that testing helmets with a steeper angle of inclined anvil in helmet impact standards might lead to an overestimation of helmet protective efficacy, potentially affecting the accurate assessment of helmet effectiveness.

The study conducted by Bonin et al.¹⁵ highlights the significant influence of MIPS technology on helmet kinematics during oblique impact scenarios, leading to variations in injury parameters. Our study quantitively demonstrates differences in PAA parameters between helmets equipped with MIPS and those without MIPS. Fig. 10 illustrates the quantitative differences in kinematic parameters, specifically PAA and PLA, between the MIPS and no-MIPS helmet groups. This analysis demonstrates the effectiveness of MIPS technology on helmet-headform kinematics. Interestingly, the ΔMIPS_PAA for the head exceeds the ΔMIPS_PLA, indicating that MIPS helmets primarily focus on reducing rotational head movement, leading to a subsequent decrease in PAA.

In this study, a single headform was exclusively employed, aligning with the general requirements of helmet testing protocols. However, it is noteworthy that disparities exist between results obtained from a single headform and those derived from a complete human body model, which merits further exploration in subsequent investigations. It is well known that varying injury criteria and headform type could yield distinct outcomes in assessments. We concentrated on linear and rotational motion considerations, specifically focusing on 2 kinematic parameters: PLA and PAA. To enhance the evaluation of helmet protection performance in inclined impact testing, future studies could contemplate the incorporation of comprehensive evaluation parameters.

Additionally, this study substantiates the efficacy of the finite element helmet model through empirical tests and provides valuable insights into the helmet protective efficacy in simulated inclined impact tests. However, it is crucial to acknowledge the limitation that this verification solely pertains to simulated tests. For a comprehensive understanding of the protective efficacy of MIPS helmets, future research endeavors could study real accident cases.

In this study, the inclined anvil parameters, impact velocities, and impact positions were adjusted quantitatively to evaluate the protective performance of helmets with and without MIPS. The conclusions derived from this analysis are: (1) Based on the simulation results of the oblique impact test, the peak head kinematic parameters of the helmet impacting at the 60° inclined anvil are low. This suggests that angled anvils with larger angles may overestimate the protective performance of the helmet in the experiment due to the lower peak kinematic values. (2) The MIPS significantly influences helmet kinematics during oblique impact conditions, particularly reducing rotational head movements of the headform. (3) Quantitative analysis reveals that MIPS reduces angular acceleration by an average of 54.8% while providing a more modest reduction of linear acceleration, which averages 15.6%. These conclusions contribute to our understanding of the effectiveness of MIPS-equipped helmets in mitigating the impact of oblique forces on the head across various simulated conditions.

Funding

This work was supported by the Natural Science Foundation Project of Xiamen City, China (3502Z20227223), and Fujian Provincial Technological Innovation Key Research and Industry Development Project (2022G43) and (2023G048).

Ethical statement

There were no human subjects in this article, so informed consent was not applicable.

Declaration of competing interest

All authors declare no relevant relationships and competing interests.

Author contributions

Conception and design: Yong Han.

Experimental design and conduct: Yong Han, Hao Yang, He Wu.

Data collection: Hao Yang, Di Pan.

Analysis and interpretation of results: Hao Yang, He Wu, Bing-Yu Wang.

Draft manuscript preparation: Yong Han, Hao Yang.

All authors reviewed the results and approved the final version of the manuscript.