Restrain Performance of Child Restraint Systems for 1.5-Year-Old Children on Commercial Airplanes: An Experimental Study

Abstract

This study aims to compare the restraint performance of two child restraint systems (CRSs) used on airplanes—a rear-facing child seat (RFCS) and the child aviation restraint system (CARES)—for 1.5-year-old children, along with their compatibility with different types of aircraft seats. 16 g longitudinal dynamic tests were conducted on two types of aircraft seats with CRSs. Results indicate poor compatibility of CARES with Type A seats, significantly increasing the risk of head, neck, and abdominal injuries, with N_ij exceeding the acceptable limit. In contrast, CARES exhibited good compatibility with Type B seats and effectively protected children. RFCS tests demonstrated effective injury risk reduction on both types of seats. It can be found that the performance of CARES depends on restraint status and seat dimensions; RFCSs provide adequate protection for 1.5-year-olds. Optimal protection can be achieved with smaller restrain angles of CRS and using thinner seat cushions. Compared to CARES, RFCSs better adapt to various aircraft seat structures, offering superior child protection.

1. Introduction

Current airworthiness regulations have established injury criteria for adult occupants, ensuring their protection during flights. However, there are no specific regulations for children, and the injury criteria used are based on adult protection. This leads to children not being adequately protected during flights. A study of accident reports published by the National Transportation Safety Board (NTSB) revealed that between 1980 and 2015, 26 out of 58 accidents involving child injuries resulted in fatal injuries, while 11 resulted in moderate injuries [1]. During takeoff and landing, children are particularly vulnerable due to rapid acceleration and deceleration, which can cause head trauma, abdominal injuries, and spinal damage, especially when appropriate restraint systems are not used [1]. In turbulence, unrestrained children are more likely to suffer lower limb fractures [2,3]. The most severe injuries occur during emergency landings. A survey from the 1980s found that unrestrained children face higher injury risks compared to restrained adults in emergency landings [4]. Therefore, to adequately protect children during flights, using child restraint systems is essential for reducing the risk of injury. According to the regulation FAR/CCAR-121.311 “Seats, Seatbelts, and Shoulder Harnesses”, children under 2 years old can be held by an adult who occupies an approved seat or berth or sits in a child restraint system (CRS) approved by the authority [5,6]. However, the CRS used in aircraft is not as common as that in automobiles, and related research in the last few decades has been limited [7].

Experimental research on aviation CRS was conducted by the Federal Aviation Administration (FAA) in 1990. The results revealed that aircraft seatbelts could cause severe head and abdominal injuries to 2-year-old children; in particular, neither booster seats nor additional loop belts were found to be effective CRSs, as they could cause severe abdominal injury. Rear-facing child seats (RFCSs) have been found to provide better protection for children, but there is still a possibility of hitting the aircraft seatback in rebounding during impact [8,9]. RFCS tests were performed by the Civil Aviation Safety Authority (CASA) of Australia, which also revealed that the head and chest of the dummy might hit the seatback during the rebound process, resulting in increased head and chest acceleration. By using a top tether to assist in restraining the child seat, excessive flipping of the seatback might be caused [10,11].

However, limited by the test conditions in the 1990s, the CRABI and CAMIX child dummies used by the FAA and the P-series dummies used by CASA were relatively behind the current child dummies in terms of bionics; thus, the data obtained in the tests were not comprehensive. The injury assessment was mainly focused on the head and chest, lacking the evaluation of the children’s neck and abdomen [8,10,11].

In addition, the FAA approved a vest-type child restraint device called Child Aviation Restraint System (CARES) based on FAR-21.305 (d) as “an equivalent safety level to TSO-C100c” [12], which could be used in the take-off, landing, and taxiing (TTL) phase. A report by TüV Rheinland Kraftfahrt GmbH Team Aviation suggested that CARES might not be able to be fixed on the seatback normally, and the corresponding aviation seat belt would pose a potential injury risk to the children’s abdomen. However, there is no available published test data regarding the assessment of the CARES system [13].

The objective of the current study was to identify the restraint performance of CARES and RFCS used in current aircraft seats and the potential injury risks of children using these CRSs. An experimental study using Q1.5 ATD was conducted in this study, and CARES performance predicted by TÜV has been verified.

2. Materials and Methods

2.1. Experimental System

Due to the complexity of impact loads in real aircraft accidents and the variation in impact loads across different types of accidents, this study selected the 16 g horizontal impact test specified by the airworthiness regulations CCAR/FAR 25.562 [14,15]. The significance of choosing the 16 g horizontal impact test lies in its representation of the typical impact conditions during emergency landings. It is used to test whether aircraft seats and restraint systems can meet airworthiness requirements under these conditions, thereby ensuring passenger protection performance. Therefore, longitudinal impact tests were conducted using CARES and RFCS.

The experimental system (Figure 1) comprised a sled system, a high-speed camera system, an aircraft seat/restraint system, and a child anthropomorphic test device (ATD).

In each test, a Q1.5 ATD with the sensors listed in

Table 1. Sensors for data collection and filtration.

Type	Measuring Equipment	Measuring Signal	Filter Level
Sled	Acceleration Sensor	Acceleration (g)	CFC 60
Head	Acceleration Sensor	Acceleration (g)	CFC 1000
Neck	Load cell	Payload (N)	CFC 1000
Neck	Load cell	Torque (N·m)	CFC 600
Chest	Acceleration Sensor	Acceleration (g)	CFC 180
Abdominal	Pressure sensor	Pressure (kPa)	CFC 180

was used. The sampling rate of the sensors was 100 kHz (On-board data acquisition systems K3881c, Klstler, Winterthur, Switzerland) and filtered according to SAE J211 (Table 1) [16]. The impact event was recorded by three high-speed cameras (NAC HX-7S) from the two sides and from above at a sampling rate of 1 kHz.

The triangular wave specified in CCAR/FAR 25.562 and SAE AS 5276/1 [17] with a peak pulse magnitude of 16 g and a speed change greater than 13.41 m/s reached within 0.18 s was used. Previous research has found that a 16 g triangular wave load can represent the typical impact conditions during the emergency landing of the aircraft [18]. In Figure 2, five test waveforms are compared with the ideal pulse.

2.2. Aircraft Seat/CRS

Two types of aircraft seats, referred to as Type A and Type B seats, were used in the experiments. The Type A and B seats met the requirements of TSO-C127a [19] and CTSO-C127b [20], respectively. Type A seats are slightly shorter in width, and their cushions are lighter and thinner than those of Type B seats (Figure 3).

Two types of CRSs, i.e., CARES and RFCS, were tested. CARES is a restraint system designed for children aged 1 to 5 years and weighing 10 to 20 kg. The additional belt restrains the system on the aircraft seatback, and the shoulder belts and chest clip restrain the torso of the dummy. The aircraft seatbelt goes through a loop at the bottom of CARES and restrains the pelvis of the dummy. During the installation process, it was found that the aircraft seatbelt could firmly restrain the pelvis of the Q1.5 ATD sitting on the Type B seat. Due to the small seat cushion width of the Type A seat and the small size of the Q1.5 ATD, the seatbelt cannot achieve pre-tensioning when restraining the ATD pelvis. This may lead to the caregiver mistakenly restraining the seatbelt on the child’s abdomen while attempting to tighten the aircraft seatbelt. Therefore, the study adopted two different conditions: one is to restrain the pelvis without pre-tension, and the other is to restrain the abdomen with pre-tension. The ATD posture and restraint conditions are exhibited in Figure 4.

A RFCS with a five-point restrain system was used in the test. The child seat was attached to the aircraft seat, with the seatbelt passing through the belt path opening at the bottom of the child seat. Its installation was realized in accordance with the requirements of SAE AS 5276/1. The pre-tension force of the aircraft seatbelt was 55 N, and the five-point seatbelt integral restraint release force met the CTSO C100c requirement [21]. The ATD posture and restraint conditions for the RFCS are presented in Figure 5. All test conditions are listed in

Table 2. Conditions of the five different sled tests.

Test Number	Seat Type	Restraint System	Test Conditions
A12977	A	CARES	The Q1.5 ATD is restrained using a CARES system equipped with an aircraft seatbelt to restrain the pelvis (without pre-tension).
B15925	A	CARES RFCS	The two Q1.5 ATDs are restrained using CARES and RFCS. When the CARES is used, the corresponding aircraft seatbelt is pre-tensioned, restraining the abdomen of the dummy.
B17091	A	RFCS	The Q1.5 ATD is restrained by the RFCS.
B17115	B	CARES	The Q1.5 ATD is restrained using a CARES system equipped with an aircraft seatbelt to restrain the pelvis (with pre-tension).
B17520	B	RFCS	The Q1.5 ATD is restrained by the RFCS.

.

2.3. Child Injury Criteria

The injury criteria for children are listed in SAE AS 5276/1. However, SAE AS 5276/1 does not contain sufficient criteria for the evaluation of child injuries, and the head injury criterion (HIC) and cumulative 3 ms chest acceleration limits for children are the same as those for adults, which does not allow for an accurate assessment of child injuries. To this end, the ECE R129/r4 and FMVSS 208 standards from the automotive field are referenced here to quantitatively assess ATD injuries [22,23]. Detailed information can be found in

Table 3. Q1.5 ATD injury criteria.

Body Part	Injury Criterion	Injury Limit	Source
Head	HIC15	600	ECE R129/r4
	Cumulative 3 ms acceleration (g)	75
Neck	Tension (N)	780	FMVSS 208
	Compression (N)	960
	Nij	1
Chest	Cumulative 3 ms acceleration (g)	55	ECE R129/r4
Abdomen	Peak abdominal pressure (kPa)	120

.

2.4. Data Analysis

The dynamic response of the restraints and ATDs was analyzed using video data acquired via high-speed cameras (1000 fps) combined with markers that were set up during testing. Data acquisition software (EVAluation PC/NCAP version 2.74, Ingenieurgesellschaft für Automobiltechnik, Inc., Berlin, Germany) was used to collect and filter the ATD sensor data. A custom MATLAB script was used to calculate the Head Injury Criterion (HIC15) and Neck Injury Criterion (Nij) based on the head and neck sensor signals, respectively. The maximum head and chest accelerations with a minimum duration of 3 ms were calculated based on the head and chest accelerometer signals, respectively.

3. Results

3.1. Dynamic Response of ATD and Restraint System

The dynamic responses of the Q1.5 ATD restrained by CARES and RFCS under 16-g longitudinal impact are demonstrated in Figure 6 and Figure 7, respectively.

When CARES was used, the forward motion of the Q1.5 ATD torso and pelvis was restrained by the aircraft seatbelt and the CARES shoulder belt and chest clip. Due to the inertia effect, the head bent forward until the neck flexion limit. Subsequently, under the action of the restraining force and the rebound force due to the seat cushion compression, the dummy moved backward in a rebound motion (Figure 6).

By comparing the situations with and without aircraft seatbelt pre-tension, the maximum forward displacement of the ATD head occurred when the seat belt was not pre-tensioned, reaching up to 240.1 mm. When the aircraft seatbelt was pre-tensioned and constrained at the abdomen, the forward displacement was limited to 190.1 mm (Figure 8).

When the Q1.5 ATD was sitting in the RFCS, its center of gravity was positioned forward. During impact, both the child seat and ATD moved forward, compressing the front portion of the aircraft seat cushion. The ATD was supported by the child seat at the head, back, and buttocks, with the five-point seatbelt restraining its movement. Subsequently, it rebounded backward under the seatbelt restraint (Figure 7).

3.2. Head Injury

In the CARES test, the highest head acceleration of the Q1.5 ATD was on the Type A seat, reaching 41.7 g, while it was significantly lower on the Type B seat (28.2 g), exhibiting a notable difference of 13.5 g. In the RFCS tests, the head acceleration range of the Q1.5 ATD was from 30.2 g to 36.2 g, indicating minimal variation (Figure 9).

Under the CARES, when the aircraft seatbelt was not pre-tensioned, children had the highest risk of head injury of the six tests, with a HIC15 of 131 and a cumulative 3 ms head acceleration of 40.5 g. Conversely, when the seatbelt was pre-tensioned and restrained on the pelvis, the injuries were significantly lower, yielding a HIC15 value of 59 and a head acceleration of 28.0 g, both below the injury limits (Figure 10).

In the RFCS tests, the Q1.5 ATD suffered the greatest head injury when seated in the Type B seat, as manifested by the HIC15 of 107 and the 3 ms acceleration of 35.97 g; conversely, the test conducted in the Type A seat exhibited the least injury, with an HIC15 of 70 and an acceleration of 30.1 g.

3.3. Neck Injury

In the CARES test, the Q1.5 ATD on the Type A seat exceeded the neck tension load limit of 780 N, reaching peak values of 1264 N and 1129 N. In contrast, in the Type B seat test, the load was significantly lower at 586 N. In the RFCS test, the backward movement of the head was limited by the child seat, resulting in a reduced tension load. In the RFCS test, the maximum tension load on the neck was 382 N, which was far less than the injury limit (Figure 11).

The Y-axis neck moment is plotted in Figure 12, where positive values indicate neck flexion. Under the restraints of the CARES and the aircraft seatbelt, the Q1.5 ATD encountered a primary flexion moment due to the forward head movement, with extension moments occurring during rebound movements. The peak neck flexion moments observed across the three CARES tests were relatively consistent, with the maximum value being recorded during the Type A seat test when the abdomen was restrained by the aircraft seatbelt, reaching 16.2 N·m. In the RFCS test, the head of the Q1.5 ATD was positioned in a reclined posture, leading to a predominant extension movement of the neck. The design of the RFCS headrest and backrest mitigated the extension movement, with the maximum extension moment being recorded during the Type A seat test (9.4 N·m).

In the CARES and RFCS tests, the neck of the Q1.5 ATD primarily encountered a tension load with varying bending moments. Proper use of the CARES and the aircraft seatbelt yielded a maximum N_ij value within the injury limit of 1. When the seatbelt was not pre-tensioned or placed incorrectly, the neck tension-flexion (N_tf) peaked at 1.18, surpassing the injury limit. Conversely, when the Q1.5 ATD was placed in the RFCS, the maximum tension-extension (N_te) was calculated to be 0.75, which was within the acceptable limits of injury criteria (Figure 13).

3.4. Chest Injury

In the CARES test, when the aircraft seatbelt was not pre-tensioned, the chest acceleration on the X-axis was noticeably high, reaching 31.68 g (

Table 4. Peak chest acceleration in the X, Y, and Z axes.

Test Number	X-Axis Acceleration (g)	Y-Axis Acceleration (g)	Z-Axis Acceleration (g)
Type A seat-CARES-Non pre-tension	−31.68	5.49	−10.01
	(71.9 ms)	(78.9 ms)	(126.4 ms)
Type A seat-CARES-Abdomen	−23.68	−4.59	4.17
	(104.3 ms)	(87.2 ms)	(96.8 ms)
Type B seat-CARES	−20.6	3.591	−7.39
	(85.1 ms)	(72.5 ms)	(91.4 ms)
Type A seat	27.97	−3.18	−14.63
Child seat-01	(110.8 ms)	(108.8 ms)	(83.5 ms)
Type A seat	35.64	−5.53	−16.32
Child seat-02	(95.1 ms)	(91.8 ms)	(78.7 ms)
Type B seat	26.73	5.29	−18.53
Child seat	(99.9 ms)	(88 ms)	(85 ms)

). Conversely, with a pre-tensioned seatbelt and utilizing the CARES to restrict the movement of the ATD torso, the chest acceleration in the X-axis was significantly decreased to 23.68 g. This adjustment not only lowered the peak chest acceleration substantially but also reduced the cumulative 3 ms chest acceleration from its highest point of 28.97 g to 23.2 g (Figure 14).

In the RFCS test, when the back of the ATD was in contact with the child seat, the torso reclined with the back of the child seat, leading to a significant increase in the Z-axis acceleration of the chest compared to that in the CARES tests. Moreover, the peak resultant acceleration of the chest was the highest in all tests when the child seat was positioned on the right position of the Type A seats, reaching 36.8 g; nonetheless, the peak acceleration lasted less than 5 ms. Concurrently, the cumulative 3 ms chest acceleration also reached its maximum at 29.6 g (Figure 15), which was within the chest injury limit.

3.5. Abdominal Injury

In the CARES test, when the aircraft seatbelt was pre-tensioned and the abdomen restrained, the peak abdominal pressure on the left side reached 80 kPa, which was the highest value observed in the tests. In contrast, when the seatbelt was properly positioned on the pelvis, the abdominal pressure was significantly decreased to its lowest value, with the left abdominal pressure being only 15 kPa. Neither condition exceeded the abdominal injury limit, nor was the abdomen status considered safe (Figure 16).

In the RFCS test, the maximum abdominal pressure on the ATD was 22.7 kPa, which was within 20% of the abdominal injury limit, indicating that the abdomen status was safe. It was also found that the peak abdominal pressure on the left and right sides of the Type B seat was 22.7 kPa and 5.2 kPa, respectively. This significant difference in the pressure was attributed to the left leg of the ATD impacting the aircraft seat, causing a temporary increase in the left abdominal pressure.

3.6. Summary of Protection Performance for Two Types of CRS

The present results indicate that the protective effectiveness of CARES is significantly influenced by the restraint status. More specifically, when the aircraft seatbelt is compatible with CARES, the head, neck, chest, and abdomen of the ATD can be effectively restrained, maintaining the injury index within the acceptable limit. However, when the aircraft seatbelt is not compatible, the injury values of the head and neck of the ATD increase significantly, exceeding the injury limits of the 1-year-old ATD in the FMVSS 208 standard [23].

When the Q1.5 ATD was placed on the RFCS, the five-point seatbelt could effectively restrain the ATD body, preventing it from escaping the protection range of the CRS. The back and headrest of the RFCS supported the ATD head, minimizing the motion amplitude of the neck and reducing the load. The back of the child seat dispersed the impact load by being in full contact with the ATD back and buttocks. The injury values concerning the head, neck, chest, and abdomen of the ATD were all within the injury limits, which was consistent with the conclusions of the TÜV research [13] that recommended the use of RFCS to protect young children.

4. Discussion

This study evaluates the protective performance of various Child Restraint Systems (CRS) and analyses the potential risks of injury to children when using these systems on aircraft. To investigate the precautions when using these CRSs further, the key factors affecting the protective performance of CARES and RFCS are discussed.

4.1. The Influence of Compatibility between CARES and Aircraft Seats on CRS Performance

Incorrect methods of restraint can adversely affect the CRS performance [24]. The incompatibility between CARES and the aircraft seat leads to incorrect restraint of children in two ways. According to the results of this work, when the aircraft seatbelt was restrained on the pelvis without pre-tension, the restraining force generated by the CARES could have caused the seatbelt to slide toward the abdomen of the child. This increased not only the abdominal pressure but also the risk of submarining (Figure 17). This phenomenon is consistent with a previous TÜV study [13], which indicated that, due to their dimensions, current aircraft seatbelts could not properly restrain young children. Furthermore, a non-pre-tensioned seatbelt failed to adequately restrict the forward movement of the buttocks, increasing head rotation. The tension of the neck due to the excessive head swing was far beyond the injury limit during impact, resulting in the most severe injuries. Biomechanical evidence has indicated that the cervical spine of children is notably flexible, making it less prone to fractures [25]. However, during non-contact tensile-flexion movements, the risk of cervical dislocation increased, potentially causing spinal cord injuries. In addition, the high tensile load may cause injuries such as tears in the neck muscles and ligaments.

On the other hand, when the pre-tensioned aircraft seatbelt was erroneously fastened across the abdomen, the impact load was directly transmitted to the abdomen of the dummy through the seatbelt, causing the abdominal pressure to increase significantly, reaching the highest value among the six sled tests. Simultaneously, the incorrect restraint positioning led to an increase in the forward displacement and head rotation amplitude of the ATD, causing the N_tf to exceed the acceptable limit.

This is different from suggestions by TüV [13] that the poor performance by CARES was because it is not able to be fixed on the seatback normally. The reason revealed by this study is the poor compatibility between CARES and aircraft seatbelts. It seems that CARES cannot provide satisfactory protection for a 1.5-year child when using current aircraft seats.

4.2. The Influence of Aircraft Seat Design on the Protective Performance of RFCS

This study examined the protective capabilities of the CRS across different aircraft seat configurations. The ATD injury analysis indicated that RFCSs can better adapt to various aircraft seats, offering superior protection compared to CARES.

Moreover, this study used different aircraft seats to assess the effects of the restraint angle (θ) and the thickness (h) of the aircraft seat cushion on the RFCS motion amplitude (Figure 18). The results revealed that the decrease of θ and h could effectively reduce the forward flipping and rebound motion of the CRS, thereby reducing the collision risk of the CRS and ATD with the aircraft seatback. In this test, the CRS did not exhibit the excessive flipping phenomenon that was reported by Gibson et al. [10] (Figure 19). This may be related to the large θ in the CASA test (about 69°), which was much larger than that of the CRS on the Type A and B seats (48° and 45°, respectively) (Figure 20). Furthermore, the thickness of the airplane seat cushions significantly affects the effectiveness of CRSs. Children seated in RFCSs have a higher and more forward center of gravity. During impact, the child seat compresses the front of the aircraft seat cushion, eventually contacting the seat tubes beneath the airplane seat, which gradually stops its flipping motion. Compared to the 23-mm-thick Type A seat cushion, the 75-mm-thick Type B one increases the gap between the bottom of the child seat and the aircraft seat tube, exacerbating the forward displacement and flipping angle of the child seat. Therefore, seat cushion thickness is one of the reasons for the excessive forward flipping of the child’s seat. Moreover, the child seat is subjected to significant rebound forces due to the cushion deformation, significantly increasing the likelihood of the child seat and ATD impacting the back of the aircraft seat during the rebound phase.

This study, building upon the conclusions drawn by the CASA regarding the significant impact of the restraint angle of aircraft seatbelts on the forward flipping motion of child seats [10], further reveals that a reduction in the restraint angle of the safety belt can significantly enhance the protective performance for children. The research also discovered that the thickness of the cushion is another key factor affecting the protective performance. By reducing the thickness of the cushion, the risk of excessive flipping for children in an accident can be effectively reduced, thereby enhancing the protective performance of the seat.

Child seats used in cars commonly utilize load legs to reduce seat movement and minimize the risk of occupant injury. However, aircraft child seats currently do not incorporate load legs, as the load legs may obstruct passenger evacuation during emergencies. Moreover, the test results indicate that the RFCS will not cause excessive overturning due to limited spacing between aircraft seats, which implies that the installation of support legs is unnecessary.

5. Limitation

The limitation of the present study is that only tests with single-row aircraft seats were included, and the potential injuries to children in the front row caused by rear-row passengers were not taken into consideration during the longitudinal dynamic impact tests.

The rear-facing child seat (RFCS) tests in this study were conducted under ideal conditions, with the aircraft seatbelt tightly securing the child seat through the belt path. However, in cases of incorrect RFCS installation, which can lead to a loose connection between the RFCS and the aircraft seatbelt, the potential injury risk to children and the protective performance of the restraint system requires further investigation.

6. Conclusions

The conclusions show that the protective efficacy of CARES and RFCSs is affected by the aircraft seat. The following conclusions are drawn:

(1)

The restrain performance of CARES depends largely on its compatibility with the aircraft seatbelt. High compatibility between the CARES and the aircraft seat can provide adequate protection for children; however, the lack of belt pre-tension or improper restraint positioning can significantly increase the risk of head, neck, and abdominal injuries.

(2)

The protective capability of RFCSs is associated with the thickness of the seat cushion and the restraint angle of the seatbelt. When the restraint angle and the cushion thickness are decreased, the risk of child injury decreases as well. Upon comparison, RFCSs are adaptable to various aircraft seat configurations and offer better protection for young children compared to the CARES.