A Deep Reinforcement Learning-Based Evolutionary Algorithm for Distributed Heterogeneous Green Hybrid Flowshop Scheduling
Processes,
Год журнала:
2025,
Номер
13(3), С. 728 - 728
Опубликована: Март 3, 2025
Due
to
increasing
energy
consumption,
green
scheduling
in
the
manufacturing
industry
has
attracted
great
attention.
In
distributed
involving
heterogeneous
plants,
accounting
for
complex
work
sequences
and
consumption
poses
a
major
challenge.
To
address
hybrid
flowshop
(DHGHFSP)
while
optimising
total
weighted
delay
(TWD)
(TEC),
deep
reinforcement
learning-based
evolutionary
algorithm
(DRLBEA)
is
proposed
this
article.
DRLBEA,
problem-based
heuristic
initialization
with
random-sized
population
designed
generate
desirable
initial
solution.
A
bi-population
global
search
local
used
obtain
elite
archive.
Moreover,
distributional
Deep
Q-Network
(DQN)
trained
select
best
strategy.
Experimental
results
on
20
instances
show
9.8%
increase
HV
mean
value
35.6%
IGD
over
state-of-the-art
method.
The
effectiveness
efficiency
of
DRLBEA
solving
DHGHFSP.
Язык: Английский
A cooperative discrete artificial bee colony algorithm with Q-learning for solving the distributed permutation flowshop group scheduling problem with preventive maintenance
Swarm and Evolutionary Computation,
Год журнала:
2025,
Номер
95, С. 101910 - 101910
Опубликована: Март 19, 2025
Язык: Английский
Lot-Streaming Workshop Scheduling with Operation Flexibility: Review and Extension
Systems,
Год журнала:
2025,
Номер
13(4), С. 271 - 271
Опубликована: Апрель 9, 2025
Lot-streaming
scheduling
methods
with
operation
flexibility
have
been
widely
used
in
aerospace,
semiconductor,
automotive,
pharmaceutical
and
other
manufacturing
enterprises.
Lot-splitting
attracted
much
more
attention
from
academia
industry
due
to
an
urgent
requirement
for
effective
way
improve
the
productivity
of
flexible
workshop
scheduling.
During
past
decade,
many
works
made
on
different
lot-streaming
The
scope
this
review
focuses
journal
publications
collected
Web
Science
database,
among
which
80%
are
high-ranked
journals.
This
paper
aims
provide
a
comprehensive
survey
flexibility.
First,
jobs
discussed
objectives
as
well
constraints
applications
summarized.
Then,
problem
models
their
solution
approaches
reviewed.
Next,
research
trends
applications,
modeling
recalled.
Finally,
potential
future
directions
concluded.
Язык: Английский