IISE Transactions,
Год журнала:
2024,
Номер
unknown, С. 1 - 12
Опубликована: Окт. 7, 2024
To
address
concerns
regarding
economic
risks
and
reliability
issues
in
existing
maintenance
practices,
this
study
introduces
a
novel
condition-based
model
that
considers
failure
terms
of
both
cost
availability.
Utilizing
Markov
decision
process,
determines
inspection
intervals
policies
aimed
at
minimizing
the
nested
conditional
Value-at-Risk
cumulative
costs
while
satisfying
operational
availability
constraints.
By
applying
to
plasma
etching
we
demonstrated
its
effectiveness
compared
models.
Additionally,
found
higher
risk
levels
do
not
necessarily
lead
stricter
policies,
whereas
achieving
better
incurs
additional
costs.
These
findings
highlight
importance
balancing
when
determining
an
optimal
policy.
Journal of Manufacturing Systems,
Год журнала:
2024,
Номер
74, С. 397 - 410
Опубликована: Апрель 11, 2024
This
paper
develops
a
data-driven
approach
to
dynamically
integrate
tactical
production
and
predictive
maintenance
planning
for
multi-state
system
composed
of
several
series-parallel
machines.
The
objective
is
determine
an
integrated
lot-sizing
preventive
strategy
that
will
minimize
the
sum
costs,
while
satisfying
demand
all
products
over
entire
horizon.
A
rolling
horizon
adopted
continuously
update
plans
based
on
new
data
obtained
through
sensors.
Unlike
existing
models,
we
develop
hybrid
deep
learning
(DL)
coordinate
decisions
multiple
To
accurately
predict
health
condition
each
machine,
developed
DL
method
combines
powers
convolutional
neural
network
(CNN),
long-short-term
memory
(LSTM),
attention
technique.
We
use
reliability
theory
estimate
capacity.
Furthermore,
genetic
algorithm
solve
large-scale
problems.
Benchmarking
are
used
compare
results
our
with
model-based
approach,
pure
LSTM,
CNN-LSTM
approach.
comparison
prediction
accuracy,
solution
quality,
computational
time.
show
superiority
suggested
CNN-LSTM-attention
framework
integrating
production.
Reliability Engineering & System Safety,
Год журнала:
2024,
Номер
249, С. 110199 - 110199
Опубликована: Май 18, 2024
This
paper
presents
a
Deep
Reinforcement
Learning
(DRL)-based
optimization
approach
for
determining
the
optimal
inspection
and
maintenance
planning
of
scrap-based
steel
production
line.
The
DRL-based
recommends
adequate
time
inspections
activities
based
on
monitoring
conditions
line,
such
as
machine
productivity,
buffer
level,
demand.
Some
practical
aspects
system,
uncertainty
duration
variable
rate
machines,
were
considered.
A
line
was
modeled
multi-component
system
considering
components
dependencies.
simulation
model
developed
to
simulate
dynamics
assist
with
development
DRL
approach.
proposed
is
compared
traditional
policies,
reactive
maintenance,
time-based
condition-based
maintenance.
In
addition,
different
algorithms
PPO
(Proximal
Policy
Optimization),
TRPO
(Trust
Region
DQN
(Deep
Q-Network)
are
investigated
in
case-based
scenario.
findings
indicated
potential
significant
financial
savings.
Therefore,
demonstrates
adaptability
has
be
powerful
tool
industrial
competitiveness.
