Frontiers in Marine Science,
Journal Year:
2024,
Volume and Issue:
11
Published: Nov. 12, 2024
To
address
human
errors
in
collision
avoidance
tasks
of
remotely
controlled
ships,
this
study
aims
to
develop
a
comprehensive
framework
for
error
analysis
within
the
context
autonomous
ships.
Firstly,
Hierarchical
Task
Analysis
method
is
utilized
identify
crew
associated
with
traditional
ship,
and
these
are
then
dissected
into
different
operational
stages
using
Information
Decision
Action
Crew
cognitive
model.
Secondly,
combination
fault
hypothesis
expert
opinions
used
potential
that
may
occur
during
operations
Thirdly,
an
integrated
approach
proposed
build
quantitative
risk
assessment
model,
which
combines
Failure
Mode
Effects
Analysis,
Evidential
Reasoning,
Belief
rules-based
Bayesian
Network.
Then,
axiomatic
verify
robustness
applicability
Finally,
based
on
results
assessment,
specific
measures
enhancing
safety
process
The
findings
show
uncoordinated
interactions
human-computer
systems
decision-making
stage
pivotal
factor
process.
Therefore,
future
design
efforts
remote-control
centre
should
prioritize
improving
clarity
interaction
interfaces.
Ocean Engineering,
Journal Year:
2024,
Volume and Issue:
307, P. 118174 - 118174
Published: May 23, 2024
Despite
the
progress
in
autonomous
ship
technology,
unknown
risks
persist
design,
operation,
and
regulation
of
maritime
surface
ships.
A
comprehensive
literature
review
for
hazard
identification
risk
analysis
method
ships
is
currently
lacking.
Based
on
a
database
62
relevant
literatures,
this
study
presents
distribution
literatures
by
journal,
year
publication,
country
or
region
authorship,
institution.
To
gain
further
insights
into
research
hotpots
frequently
neglected
influential
factors,
are
classified
four
groups
based
categories
list
factors
compiled.
this,
content
analysed
with
respect
to
human
ship-related
environmental
technology
factors.
Furthermore,
statistical
conducted
23
related
systematic
terms
data
sources
methods,
noting
that
researchers
commonly
utilize
datasets
combination
methods.
This
not
only
contributes
understanding
current
status
challenges
but
also
provides
potential
future
directions.
Journal of Marine Engineering & Technology,
Journal Year:
2024,
Volume and Issue:
unknown, P. 1 - 20
Published: Oct. 15, 2024
Maritime
Autonomous
Surface
Ships
(MASS)
have
gained
much
attention
as
a
safer
and
more
efficient
mode
of
transportation
potential
solution
to
reduce
the
workload
seafarers.
Despite
highly
sophisticated
autonomous
systems
that
enable
MASS
make
independent
decisions,
presence
humans
on
board
or
in
loop
safety
management
highlights
need
for
effective
human-machine
interaction.
However,
potentially
systematic
review
critical
aspects
human-MASS
interaction
has
not
yet
been
conducted.
In
this
paper,
we
aim
fill
gap
by
reviewing
literature
related
from
four
crucial
perspectives:
state
art
interaction,
situational
awareness
MASS,
collision
avoidance
methods
within
mixed
waterborne
transport
system
(MWTS),
human
trust
MASS.
Our
reveals
efficiency
mainly
focuses
key
aspects:
(i)
factors,
(ii)
available
technologies
supporting
autonomy
(iii)
analysis
design
(iv)
requirements
regarding
regulations.
Moreover,
provide
detailed
discussion
three
fundamental
factors
influence
including
awareness,
decision-making
system,
Finally,
based
our
analysis,
propose
an
integrated
framework
which
these
are
taken
into
account.
We
anticipate
their
will
receive
improve
Applied Sciences,
Journal Year:
2025,
Volume and Issue:
15(6), P. 3114 - 3114
Published: March 13, 2025
Analyzing
risk
coupling
effects
in
highway
accidents
provides
guidance
for
preventive
decoupling
measures.
Existing
studies
rarely
explore
the
differences
between
primary
(PA)
and
secondary
(SA)
from
a
quantitative
perspective.
This
study
proposes
method
to
measure
of
PA
SA
on
highways
examine
their
differences.
A
domain-pretrained
named
entity
recognition
(NER)
model,
TRBERT-BiLSTM-CRF,
is
proposed
identify
factors
types
based
431
accident
investigation
reports
published
by
emergency
management
departments
China.
The
N-K
model
was
applied
calculate
values
different
scenarios
SA,
Wilcoxon
signed-rank
test
performed
them.
Finally,
were
compared,
targeted
prevention
recommendations
are
provided.
results
showed
that
our
NER
achieved
best
macro-F1
score
traffic
recognition.
Most
increased
with
number
types,
but
value
five
lower
than
four
factors,
indicating
do
not
always
superimpose
each
other
complex
scenarios.
Moreover,
there
significant
mechanisms
SA.
suggest
likelihood
occurrences
should
be
reduced
through
standardized
vehicle
inspections
flexible
control
measures,
respectively,
thereby
enhancing
safety.
