Intelligent Transportation Infrastructure,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Dec. 23, 2024
Abstract
Understanding
the
association
between
metro
ridership
and
built
environment
is
crucial
for
promoting
integrated
transportation
land
use
planning.
However,
prior
research
has
rarely
examined
temporally
varying
and/or
non-linear
associations
environment.
To
address
this
gap,
study
collects
data
in
Chengdu,
China,
January
of
each
year
2019
2022
uses
light
gradient-boosting
machine
(LightGBM)
SHapley
Additive
exPlanations
(SHAP)
models
to
examine
complex,
over
four
years.
Our
findings
highlight
nature
environment’s
influence.
The
key
predictors
remained
relatively
stable
throughout
years,
including
number
entrances
(the
top
predictor
across
all
years),
employment
density,
floor
area
ratio.
influence
factors,
such
as
land-use
mix,
residential
micro-district
distance
city
center,
shows
great
temporal
variations,
underscoring
importance
incorporating
dynamics
into
analyses
interactions
This
offers
a
valuable
reference
urban
planners
crafting
tailored
policies
station-area
transit-oriented
development
(TOD).
Land,
Journal Year:
2025,
Volume and Issue:
14(4), P. 665 - 665
Published: March 21, 2025
Understanding
how
land
use
affects
urban
rail
transit
(URT)
ridership
is
essential
for
facilitating
URT
usage.
While
previous
studies
have
explored
the
way
that
impacts
ridership,
few
figured
out
this
impact
evolves
over
time.
Utilizing
turnstile
and
data
in
Beijing,
we
employed
panel
analysis
methods
to
verify
existence
of
temporal
heterogeneity
capture
heterogeneity.
The
results
identified
time-varying
on
boarding
alighting
trips
weekdays
non-weekdays
also
demonstrated
rationality
mixed
effects
coefficient
(TVC-P)
model
capturing
accurately.
TVC-P
revealed
density
appealed
commuting
during
weekday
morning
peak
times,
it
triggered
generation
commutes
evening
rush
hours.
diversity
promoted
an
extended
period
non-weekdays.
Additionally,
study
specific
ridership.
These
insights
provide
both
theoretical
empirical
support
developing
policies
actions
improve
efficiency
transportation
systems
foster
alignment
between
transport.
ISPRS International Journal of Geo-Information,
Journal Year:
2024,
Volume and Issue:
13(8), P. 266 - 266
Published: July 26, 2024
The
growing
relevance
of
promoting
a
transition
urban
mobility
toward
more
sustainable
modes
transport
is
leading
to
efforts
understand
the
effects
built
environment
on
use
railway
systems.
In
this
direction,
there
are
challenges
regarding
creation
coherence
between
locations
metro
stations
and
their
surroundings,
which
has
been
explored
extensively
in
academic
community.
This
process
called
Transit-Oriented
Development
(TOD).
Within
context
Latin
America,
study
seeks
assess
influence
ridership
metropolitan
area
Valparaíso,
Chile,
testing
two
approaches
definition,
one
fixed
distance
from
stations,
other
based
origin
destination
survey
area.
analysis
Ordinary
Least
Squares
regression
(OLS)
identify
factors
environment,
affects
metro’s
ridership.
Results
show
that
models
defined
through
explain
better.
Moreover,
reveals
system
Greater
Valparaíso
was
not
planned
harmony
with
development.
demonstrate
an
inverse
effect
ridership,
contrasting
expected
outcomes
station
designed
following
approach.
International Journal of Geographical Information Science,
Journal Year:
2024,
Volume and Issue:
unknown, P. 1 - 20
Published: Oct. 7, 2024
Scale
in
multiscale
geographically
weighted
regression
(MGWR)
directly
impacts
the
accuracy
of
coefficient
estimates
and
shapes
comprehensive
evaluation
intensity
spatially
non-stationary
relationships.
Presently,
MGWR
primarily
utilizes
back-fitting
for
sequentially
optimizing
multiple
scales
(MGWR-BF).
However,
set
individual
optima
obtained
through
sequential
optimization
may
not
necessarily
represent
global
optimum.
To
address
this
issue,
paper
proposes
a
multi-scale
cooperative
within
(MGWR-GA)
model.
Specifically,
MGWR-GA
employs
genetic
algorithm
to
simultaneously
input
potential
scale
combinations,
each
comprising
P
scales.
Subsequently,
it
introduces
dedicated
overall
estimation
designed
these
scales,
ultimately
determining
optimal
combinations
based
on
AICc.
Simulation
experiments
have
shown
that,
at
least
stationarity,
by
approximate
true
values
across
twelve
different
test
environments.
Additionally,
bias
is
lower
than
that
MGWR-BF,
especially
low
signal-to-noise
ratio
settings.
Empirical
further
confirm
effectiveness
identifying
both
globally
stationary
locally
Furthermore,
outperforms
MGWR-BF
terms
goodness-of-fit,
adjusted
AICc
spatial
autocorrelation
residuals.
These
findings
indicate
can
serve
as
valuable
tool
modeling
