Commercial air travel remains the safest mode of transport
available in OECD countries. Commercial airlines in Australia do
not require infants under the age of 24 months to occupy their own
seats during flight. However, the children carried in the arms of
adult passengers must be restrained during taxi, take-off, landing
and turbulence.

The aims of this project were to review the developments in safe
transport of children in aircraft and to conduct a test program
based on current Australian child restraint systems (CRS). This
initial program was later extended to include the assessment of
infant carrier systems (commonly referred to as baby slings) for
use as infant restraints in aircraft.

A US Civil Aerospace Medical Institute (CAMI) study found that
lap-held restraint systems allowed excessive forward body excursion
of the test dummies, resulting in severe head impact with the seat
back directly in front. The tests showed how a lap-held infant
could be crushed between the forward seat back and the accompanying
adult during impact (Gowdy & De Weese 1994). Following the CAMI
study, the US Federal Aviation Administration (FAA) banned the use
of booster seats and all lap-held restraint devices in aircraft
during take-off, landing and taxi. This has resulted in lap-held
children travelling wholly unrestrained in aircraft.

The travelling public is likely to expect that the level of
safety offered to child passengers in commercial aircraft is
equivalent to that of adult passengers restrained by lap belts. The
use of an appropriate child restraint system can offer the highest
level of safety for young children travelling in aircraft, both in
turbulence and in crash situations. However, the compatibility of
current Australian automotive CRS with aircraft seating has not
been investigated and their performance in aircraft emergency
situations is unknown.

There are very few preventable child deaths in aircraft crashes.
Newman, Johnston and Grossman (2003) found that the use of CRS
would prevent 0.4 child air-crash deaths per year. They concluded
that making infant air-seats compulsory would raise air travel
costs which could result in a net increase in deaths and injuries
as families opt for automobile travel - a higher-risk mode of
transport per kilometre of travel.

Child restraint testing

A selection of automotive CRS available in Australia was chosen
for testing in this study to cover the range of common child
restraint types. The DME Corporation PlaneSeat, certified for use
in both motor vehicles and aircraft in the United States, was also
examined. The testing was completed in three stages:

  1. Fit test

    The CRS were fitted to an economy class aircraft seat row to check
    for compatibility. Twenty Australian standard CRS were fitted to
    the aircraft seat according to the manufacturer's instructions.
    Fourteen of the CRS models had problems in this test. Either they
    did not fit within the 31-inch seat pitch or they were difficult to
    fit due to interference with the latching mechanism of the aircraft
    seat lap belt. One restraint was designed for use only with a top
    tether strap requiring an anchorage system not available in
    commercial aircraft.

  2. Turbulence (inversion) test

    The CRS were subjected to the FAA seat inversion test for
    turbulence. This test caused no difficulty for the Australian CRS,
    which have 6-point harnesses for the child. Booster seats were not
    tested in this series.

  3. Dynamic sled test

    The Australian automotive CRS were subjected to the requirements
    of the dynamic FAA aircraft seat test, without the top tether
    normally required in motor vehicle installation. The CRS were
    installed on a single aircraft seat row by the lap belt and
    subjected to a 16G longitudinal test with a velocity change of more
    than 45 km/h. Forty-two sled tests were conducted involving 11
    models of Australian CRS together with tests where dummies were
    restrained only by the aircraft seat lap belt. The average sled
    deceleration for the tests was 18.9G and the mean entry velocity
    was 47.6 km/h.

    The dummies were retained in the CRS in all sled tests. However,
    all the CRS exhibited significant forward motion, rotation, and
    rebound motion. This less controlled movement, in comparison with
    typical automotive testing of CRS, was due to the following:

    • the upper tether could not be installed;
    • the more vertical geometry of the aircraft seat lap belt;
    • the poor compatibility of the aircraft seat lap belt design and
      the CRS belt paths;
    • the poor interaction of the CRS with the aircraft seat base
      cushion and frame;
    • a rebound phase that was poorly controlled due to the more
      extensive forward motion of the CRS.

In tests where the child dummies were restrained only by the
aircraft seat lap belt, excessive forward motion of the dummy head
and torso occurred due to the lack of upper body restraint and the
folding over of the aircraft seat back. This motion is likely to
result in impact with the forward seat back.

Infant carrier testing

Four commercially available infant carriers were chosen as
representative and were tested to evaluate their performance with
respect to retention of the child, forward excursion, and crushing
by the adult. Two samples of the standard 'supplementary loop
belts' (or belly belts) were included for comparative testing. The
testing was conducted in two stages:

  1. Turbulence (inversion) test

    The infant carriers were subjected to an inversion test to
    simulate turbulent conditions. An infant dummy was placed in the
    carrier and fitted to an adult dummy restrained by a lap belt in an
    aircraft seat. The tests demonstrated that infants could be
    adequately restrained when exposed to 1G of vertical acceleration
    provided the carrier was securely fastened.

  2. Sled test

    A lap-belt restrained adult dummy in an aircraft seat, with an
    infant dummy in a carrier, was subjected to a 9G dynamic sled test.
    The severity of the pulse was based on the results of a static load
    test. The commercially available infant carriers tested were not
    able to restrain infants under crash situations.

The infant carriers could be redesigned to ensure that the
infant was restrained in dynamic loads equivalent to the test
pulse. If this was done, then an infant carrier would form an
alternative to the supplementary loop belt.

Suggested actions

The following suggestions are made based on the findings of this
study and the principle that infants and young children are
entitled to the same level of protection, both in flight and during
emergency landing situations, that is afforded to adults.

  1. The use of CRS by infants and young children on flights in
    Australia is to be encouraged. The CRS used could be either
    designed specifically for use in aircraft, or, Australian
    automotive CRS approved for use in aircraft as per suggestion
    number 3.

  2. Testing should be conducted of the system of an upper tether
    strap for Australian automotive CRS with a non-breakover aircraft
    seat back, as currently used by Qantas.

  3. An approval system should be established to ensure that any
    Australian automotive CRS to be used in aircraft fits in the
    aircraft seat and is compatible with the aircraft lap belt. The
    approval could be in the form of an extra test added to the
    existing motor vehicle requirements similar to the FAA approval
    system.

  4. Improvements in the crash performance of Australian automotive
    CRS in aircraft could be achieved by making changes to the seating
    systems in the aircraft to minimise forward excursion of the CRS in
    the seat. In order of priority, these suggested improvements
    are:

    1. Supply a properly mounted upper tether, either as used by Qantas
      should testing show that this is effective or, by supplying
      attachment points in the aircraft for CRS use. This could be
      achieved by restricting CRS use to the seats forward of a bulkhead
      and by requiring a modified bulkhead design with appropriate
      attachment points built in for the tether.

    2. Change lap belt geometry (angled at 45 to 60 degrees instead of
      vertical) for use with a CRS to reduce the initial forward
      excursion of the base. However, such seat belt geometry may not be
      appropriate for other users of the belt.

    3. Make changes to the seat base cushion to ensure its retention
      under CRS dynamic loads.

  5. Improvements in the crash protection offered in aircraft to an
    infant seated on the lap of an adult could be achieved if some
    seats were fitted with lap sash or harness type seat belts for use
    by parents holding infants. These seats, possibly adjacent to a
    bulkhead could be forward- or rearward-facing. Controlling the
    upper torso motion of the adult has the potential to reduce crash
    loading to an infant seated on the lap of an adult.
  6. If suggestion 5 was implemented, then an approval system for
    infant carriers (slings) for use in aircraft should be put in
    place. A sling system could be designed and developed as a
    replacement for the belly belt. This type of infant carrier could
    offer improved retention and comfort in turbulent conditions; in
    conjunction with appropriate seating fitted with a lap/sash or
    harness for the parent, it could offer improved safety for the
    infant in a crash.

  7. The changes resulting from the incorporation of ISO rigid
    anchorage systems (ISO-fix or latch systems), which are becoming
    mandatory worldwide, need to be studied and accommodated for use in
    aircraft.

Publication Mode
Publication date
Publication type
Review Date
Authors
Gibson, T, Thai, K and Lumley, M
Subject Matter