This
study
investigates
the
important
management
factors
contributing
to
carbon
emissions
of
infrastructure
construction
project
in
Thailand.
Specifically,
research
seeks
identify
which
stages
process
most
significantly
cause
emissions.
At
beginning,
were
reviewed
and
selected.
Then,
interviews
with
managers
related
persons
projects
conducted
investigate
real
context.
Next,
discussion
five
experts
who
have
relevant
experience
emission
for
then
arose
validate
possible
key
factors.
Based
on
expert
comments,
questionnaire
was
developed
its
validity
clarity.
After
that,
distributed
829
certified
from
Thai
Green
Building
Institute
(GBI),
employs
exploratory
factor
analysis
(EFA)
adequacy
data,
achieving
a
significant
Kaiser-Meyer-Olkin
(KMO)
measure
0.805.
The
identifies
three
primary
components
responsible
emissions:
(1)
manufacturing
materials
process,
an
eigenvalue
4.396,
accounting
39.964%
variance,
highlights
impact
raw
material
production;
(2)
transportation
1.599,
explains
14.534%
underscoring
environmental
implications
transporting
waste;
(3)
1.279,
contributes
11.624%
focusing
directly
linked
on-site
activities.
results
demonstrate
strong
statistical
correlations
among
measured
latent
variables,
indicating
robust
model
fit.
Key
highest
loadings
include
concrete
production,
steel
manufacturing,
vehicle
distance
used
waste
debris
disposal,
others.
not
only
sources
sector
but
also
provides
insights
into
potential
areas
effective
reduction.
Buildings,
Год журнала:
2024,
Номер
14(6), С. 1845 - 1845
Опубликована: Июнь 18, 2024
(1)
Background:
Artificial
intelligence
(AI)
and
machine
learning
(ML)
techniques
are
being
more
widely
employed
in
the
field
of
wind
engineering.
Nevertheless,
there
is
a
scarcity
research
on
comfort
pedestrians
terms
conditions
with
respect
to
building
design,
particularly
historic
sites.
(2)
Objectives:
This
aims
evaluate
ML-
computational
fluid
dynamics
(CFD)-based
pedestrian
(PWC)
analysis
outputs
using
novel
method
that
relies
sophisticated
handling
image
data.
The
goal
propose
assessment
enhance
efficiency
AI
models
over
different
urban
scenarios.
(3)
Methodology:
stages
include
climate
data,
CFD
OpenFOAM,
ML
Autodesk
Forma,
comparisons
results
similarity
based
SSIM,
MSE,
PSNR
metrics.
(4)
Conclusions:
study
effectively
demonstrates
considerable
potential
utilizing
as
supplementary
tool
for
evaluating
PWC.
It
maintains
high
degree
accuracy
precision,
allowing
rapid
effective
assessments.
methodology
precise
comparison
two
visual
absence
numerical
data
allows
objective
pertinent
comparisons,
it
eliminates
any
distortions.
(5)
Recommendations:
Additional
can
explore
integration
case
studies,
thus
expanding
scope
studies.
Buildings,
Год журнала:
2025,
Номер
15(4), С. 592 - 592
Опубликована: Фев. 14, 2025
The
severe
global
warming
driven
by
the
large-scale
emission
of
greenhouse
gases
has
made
reduction
carbon
emissions
a
critical
priority
for
economic
and
social
development.
Among
various
sectors,
construction
industry
stands
out
due
to
its
significant
consumption
natural
resources
throughout
building
process,
resulting
in
considerable
environmental
burden.
In
China,
from
account
approximately
40%
total
emissions.
Therefore,
mitigating
this
sector
is
utmost
importance.
This
study
develops
an
evaluation
model
low-carbon
production
management
enterprises,
utilizing
Analytic
Hierarchy
Process
(AHP).
Through
case
study,
research
identifies
practical
challenges
implementing
offers
actionable
recommendations.
Theoretically,
provides
valuable
reference
future
on
energy
conservation
industry.
practice,
it
guidance
enterprises
achieving
transition.
Mathematics,
Год журнала:
2025,
Номер
13(9), С. 1481 - 1481
Опубликована: Апрель 30, 2025
Global
carbon
dioxide
(CO2)
emissions
are
increasing
and
present
substantial
environmental
sustainability
challenges,
requiring
the
development
of
accurate
predictive
models.
Due
to
non-linear
temporal
nature
data,
traditional
machine
learning
methods—which
work
well
when
data
structured—struggle
provide
effective
predictions.
In
this
paper,
we
propose
a
general
framework
that
combines
advanced
deep
models
(such
as
GRU,
Bidirectional
GRU
(BIGRU),
Stacked
Attention-based
BIGRU)
with
novel
hybridized
optimization
algorithm,
GGBERO,
which
is
combination
Greylag
Goose
Optimization
(GGO)
Al-Biruni
Earth
Radius
(BER).
First,
experiments
showed
ensemble
such
CatBoost
Gradient
Boosting
addressed
static
features
effectively,
while
time-dependent
patterns
proved
more
challenging
predict.
Transitioning
recurrent
neural
network
architectures,
mainly
BIGRU,
enabled
modeling
sequential
dependence
on
data.
The
empirical
results
show
GGBERO-optimized
BIGRU
model
produced
Mean
Squared
Error
(MSE)
1.0
×
10−5,
best
tested
approach.
Statistical
methods
like
Wilcoxon
Signed
Rank
Test
ANOVA
were
employed
validate
framework’s
effectiveness
in
improving
evaluation,
confirming
significance
robustness
improvements
due
framework.
addition
accuracy
CO2
forecasting,
integrated
approach
delivers
interpretable
explanations
significant
factors
emissions,
aiding
policymakers
researchers
focused
climate
change
mitigation
data-driven
decision-making.
Buildings,
Год журнала:
2025,
Номер
15(9), С. 1541 - 1541
Опубликована: Май 2, 2025
For
more
than
two
decades,
computational
analysis
has
been
pivotal
in
expanding
architectural
capabilities,
enabling
sustainable
design
through
detailed
environmental
analysis.
Central
to
creating
environments
is
the
profound
understanding
of
wind
dynamics,
which
significantly
influence
comfort
levels
around
buildings.
Traditionally,
tunnel
experiments,
situ
measurements,
and
fluid
dynamics
(CFD)
simulations
have
employed
assess
speeds
urban
settings.
However,
advent
machine
learning
(ML)
introduced
innovative
methodologies
that
extend
beyond
these
conventional
approaches,
offering
new
insights
applications
design.
This
study
focuses
on
evaluating
pedestrian-level
using
ML
techniques,
with
a
comparative
against
traditional
measurements
CFD
simulations.
Our
findings
reveal
can
predict
sufficient
accuracy
for
preliminary
phases.
One
primary
challenges
addressed
integration
visual
outputs
from
models
quantitative
data,
necessary
step
enhance
model
reliability
applicability.
By
developing
novel
techniques
this
integration,
our
research
marks
significant
contribution
field,
benchmarking
effectiveness
established
methods.
The
results
validate
model’s
capability
accurately
estimate
speeds,
thereby
supporting
comfortable
environments.