Revista de Gestão Social e Ambiental,
Год журнала:
2023,
Номер
17(4), С. e03561 - e03561
Опубликована: Июнь 27, 2023
Objectives:
The
Internet
of
Things
(IoT)
framework
is
crucial
for
improving
monitoring
applications
smart
cities
and
controlling
municipal
operations
in
real
time.
most
significant
issue
with
to
has
been
the
handling
solid
waste,
which
may
have
negative
consequences
on
health
well-being
people.
Waste
management
become
a
problem
that
developing
developed
nations
must
face.
waste
exciting
affects
habitats
all
around
world.
Thus,
it
necessary
create
an
efficient
method
eliminate
these
issues
or,
at
very
least,
reduce
them
manageable
level.
Theoretical
framework:
This
work
proposed
Improved
Particle
Swarm
Optimization
Deep
Learning-based
Municipal
Solid
Management
(IPSODL-MSWM)
cities.
Methods:
IPSODL-MSWM
approach
aims
identify
various
types
materials
enable
sustainable
management.
A
Single
Shot
Detection
(SSD)
model
enables
object
detection
paradigm.
Then,
feature
vectors
were
generated
using
MobileNetV2
based
deep
Convolutional
Neural
Network
(CNN).
IPSO
obtained
by
hybrid
Genetic
Algorithm
(GA)
PSO
algorithm.
Results
Conclusion:
IPSODL
employed
automatic
hyperparameter
tuning
since
manual
trial-and-error
time-consuming.
Implications
research:
uses
Support
Vector
Machine
(SVM)
accurate
excess
categorization
this
work.
implies
better
city
development.
Originality/value:
With
optimal
accuracy
99.45%,
many
simulations
show
model's
enhanced
capability
classification.
Scientific Reports,
Год журнала:
2025,
Номер
15(1)
Опубликована: Янв. 23, 2025
Effective
construction
waste
(CW)
management,
mainly
concrete,
brick,
and
steel,
is
a
critical
challenge
due
to
its
significant
environmental
economic
impacts.
This
study
addresses
this
by
proposing
multiple
linear
regression
models
predict
generation
in
residential
buildings
within
the
Egyptian
industry,
considering
influence
of
factors
such
as
building
design
site
management
features.
Using
data
from
25
case
studies,
demonstrated
high
predictive
accuracy,
with
adjusted
R²
values
0.877,
0.893,
0.889
for
bricks,
steel
waste,
respectively.
These
R2
indicate
that
explain
approximately
88-89%
variance
buildings,
highlighting
their
effectiveness
enhancing
resource
planning
strategies.
The
findings
suggest
incorporating
variables
total
area,
consistency,
organization
significantly
improves
accuracy
predictions.
Although
show
acceptable
performance,
future
research
should
aim
expand
dataset,
incorporate
additional
variables,
test
across
different
types
projects
validate
further
refine
these
tools.
offer
valuable
insights
practices,
minimizing
supporting
sustainable
development
Egypt's
industry.
With
accurate
forecasts
generation,
help
project
managers
stakeholders
plan
CW
more
effectively,
mitigating
unnecessary
material
consumption
reducing
adopt
improved
recycling
processes
decreased
dependence
on
landfills,
support
Vision
2030.
Automation in Construction,
Год журнала:
2023,
Номер
152, С. 104898 - 104898
Опубликована: Май 10, 2023
Accurate
quantification
and
detailed
classification
of
construction
waste
are
paramount
to
improving
their
management.
Over
the
last
decades,
various
models
have
been
developed
measure,
manage,
report
generation.
A
understanding
those
is
essential
explore
applications
across
life-cycle
stages
a
built
asset.
Existing
reviews
primarily
focused
on
analysing
functions
methodologies,
but
digital
information
standards
automate
process
under-explored
in
existing
literature.
review
adopted
analyse
papers
published
from
2012
2022.
Out
279
articles
retrieved,
71
meeting
eligibility
criteria
were
included.
critical
analysis
indicates
that
unified
data
structure,
standard
information,
approach
interoperability
between
BIM
knowledge
bases
vital
reinforce
efficiency.
Based
findings,
conceptual
framework
demonstrate
workflow
for
building
projects.
The
outcomes
will
facilitate
researchers
identify
prevailing
gaps
enhance
system
meet
demands.
