Subido por stuvka

The Use of Intelligent Systems to Support the Decision–Making Process in Lean Maintenance Management

Anuncio
13th IFAC Workshop on Intelligent Manufacturing Systems
13th
IFAC
Workshop
on
Manufacturing
Systems
Available online
at www.sciencedirect.com
13th
IFAC
Workshop
on Intelligent
Intelligent
Manufacturing
Systems
August
12-14,
2019. Oshawa,
Canada
13th
IFAC
Workshop
on Intelligent
Manufacturing Systems
August
12-14,
2019. Oshawa,
Oshawa,
Canada
August
12-14,
2019.
Canada
13th
IFAC
Workshop
on
Intelligent
Manufacturing
Systems
August 12-14, 2019. Oshawa, Canada
August 12-14, 2019. Oshawa, Canada
ScienceDirect
IFAC PapersOnLine 52-10 (2019) 148–153
The
Use
of
Intelligent
Systems
to
Support the
Decision–Making
Process
The
Use
of
Intelligent
Systems
to
the
Decision–Making
Process
The
Use of
of Intelligent
Intelligent Systems
Systems to
to Support
Support
the
Decision–Making
Process
The
Use
Support
the
Decision–Making
Process
in
Lean
Maintenance
Management
The Use of Intelligent
Systems
to
Support
the
Decision–Making
Process
in
Lean
Maintenance
Management
in
Lean
Maintenance
Management
in
Lean Maintenance
Maintenance Management
Management
in Lean
Katarzyna
Antosz*, Lukasz Pasko**,
Arkadiusz Gola***
Katarzyna
Lukasz Pasko**,
Pasko**, Arkadiusz Gola***
Gola***
Katarzyna Antosz*,
Antosz*, Lukasz
 Pasko**, Arkadiusz
Katarzyna Antosz*, Lukasz
Arkadiusz Gola***

Katarzyna
Antosz*, Lukasz
Arkadiusz Gola***
*
Rzeszow
Mechanical
 Pasko**,
*
Rzeszow University
University of
of Technology,
Technology, Faculty
Faculty of
of
Mechanical Engineering
Engineering and
and Aeronautics,
Aeronautics, Poland
Poland
* Rzeszow
University
of
Technology,
Faculty
of
 Mechanical Engineering and Aeronautics, Poland
(e-mail:
[email protected]).
* Rzeszow University of Technology,
Faculty
of
Mechanical
Engineering
and
Aeronautics,
Poland
(e-mail:
[email protected]).
(e-mail: Faculty
[email protected]).
* Rzeszow University
of Technology,
of Mechanical
Engineering and
Aeronautics, Poland
**
Faculty
(e-mail: [email protected]).
** Rzeszow
Rzeszow University
University of
of Technology,
Technology,
Faculty of
of Mechanical
Mechanical Engineering
Engineering and
and Aeronautics,
Aeronautics, Poland
Poland
**
Rzeszow
University
of
Technology,
Faculty
of
Mechanical
Engineering
and
Aeronautics,
Poland
(e-mail:
[email protected]).
(e-mail:
[email protected]).
** Rzeszow University of Technology,
Faculty
of
Mechanical
Engineering
and
Aeronautics,
Poland
(e-mail:
[email protected]).
(e-mail:
[email protected]).
*****
Rzeszow
Faculty
ofofMechanical
Engineering
andInformation,
Aeronautics, Poland
LublinUniversity
Universityof
ofTechnology,
Technology,
Institute
Technological
Systems
(e-mail:
[email protected]).
*** Lublin
Institute
of
Systems of
of Information,
Information, Poland
Poland
***
Lublin University
University of
of Technology,
Technology,
Institute
of Technological
Technological Systems
of
Poland
(e-mail:
[email protected]).
(e-mail:
[email protected])}
*** Lublin University of Technology,
Institute
of Technological Systems of Information, Poland
(e-mail:
[email protected])}
(e-mail:
[email protected])}
*** Lublin University of Technology, Institute of Technological Systems of Information, Poland
(e-mail: [email protected])}
Abstract: Manufacturing
Manufacturing companies
companies(e-mail:
continually
aim at
at increasing
increasing the
the performance
performance and
and effectiveness
effectiveness of
of
[email protected])}
Abstract:
continually
aim
Abstract:
Manufacturing
companies
continually
aim
at increasing
the performance
and
effectiveness
of
maintenance
processes.
The
emphasis
is
put
on
the
elimination
of
unexpected
failures
which
generate
unAbstract:
Manufacturing
companies
continually
aim
at increasing
the performance
and
effectiveness
of
maintenance
processes.
The
emphasis
is
put
on
the
elimination
of
unexpected
failures
which
generate
unmaintenance
processes.
The
emphasis
is
put
on
the
elimination
of
unexpected
failures
which
generate
unAbstract:
Manufacturing
companies
continually
aim
at
increasing
the
performance
and
effectiveness
of
necessary
costs
and
production
losses.
The
element
that has
has an
an impact
impact
on the
the efficiency
efficiency
of maintenance
maintenance
is
maintenance
processes.
The emphasis
is
putelement
on the elimination
of unexpected
failures which
generate unnecessary
costs
and
production
losses.
The
that
on
of
is
necessary
costs
and
production
losses.
The
element
that
has
an
impact
on
the
efficiency
of
maintenance
is
maintenance
processes.
The
emphasis
is
put
on
the
elimination
of
unexpected
failures
which
generate
unnot
only
the
selection
of
an
appropriate
conservation
strategy
and
the
use
of
appropriate
methods
and
necessary
costs
and production
losses. Theconservation
element thatstrategy
has an impact
onuse
the of
efficiency
of maintenance
is
not
only
the
selection
of
an
appropriate
and
the
appropriate
methods
and
not
only
the
selection
of
an
appropriate
conservation
strategy
and
the
use
of
appropriate
methods
and
necessary
costs
and
production
losses.
The
element
that
has
an
impact
on
the
efficiency
of
maintenance
is
tools
to
support
the
decision-making
process
in this
this area.
area.strategy
The aim
aimand
of this
this
workofis
isappropriate
to present
present the
the
possibility
not
only
the selection
of an appropriate
conservation
the use
methods
and
tools
to
support
the
decision-making
process
in
The
of
work
to
possibility
tools
to
support
the
decision-making
process
in
this
area.
The
aim
of
this
work
is
to
present
the
possibility
not
only
the
selection
of
an
appropriate
conservation
strategy
and
the
use
of
appropriate
methods
and
of
using
intelligent
systems
to
support
decision-making
processes
in the
the
implementation
ofpossibility
the Lean
Lean
tools
to support
the decision-making
process
in this area. The
aim of this
work
is to present theof
of
using
intelligent
systems
to
decision-making
processes
in
implementation
the
of
using
intelligent
systems
to support
support
decision-making
processes
in the
implementation
ofpossibility
the Lean
tools
to support
the decision-making
process
in this
area.
The
aim
of this
work
is to
present the
Maintenance
concept,
which
allows
to
increase
the
operational
efficiency
of
the
company's
technical
inof
using
intelligent
systems
to
support
decision-making
processes
in
the
implementation
of
the
Lean
Maintenance
concept,
which
to
the
efficiency
the
inMaintenance
concept,systems
which allows
allows
to increase
increase
the operational
operational
efficiency
ofimplementation
the company's
company's technical
technical
inof
using intelligent
to support
decision-making
processes
in theof
of the Lean
frastructure.
Maintenance
concept,
which
allows
to
increase
the
operational
efficiency
of
the
company's
technical
infrastructure.
frastructure.
Maintenance concept, which allows to increase the operational efficiency of the company's technical infrastructure.
Keywords:
decision
–– making
process,
maintenance
management,
Maintenance,
logic,
decifrastructure.
© 2019, IFAC
(International
Federation
Automatic Control)
HostingLean
by Elsevier
Ltd. All fuzzy
rights reserved.
Keywords:
decision
process,
management,
Lean
Maintenance,
fuzzy
logic,
Keywords:
decision
– making
making
process,ofmaintenance
maintenance
management,
Lean
Maintenance,
fuzzy
logic, decidecision
tree,
the
theory
of
rough
sets
Keywords:
decision
– making
process, maintenance management, Lean Maintenance, fuzzy logic, decision
tree,
theory
rough
sion
tree, the
thedecision
theory of
of
rough sets
sets
Keywords:
– making
process, maintenance management, Lean Maintenance, fuzzy logic, decision tree, the theory of rough sets
sion tree, the theory of rough sets
These activities
activities are
are related
related to
to the elimination
elimination of
of losses, e.g.
e.g.
These
1.
These
activities
are in
related
to the
the elimination
of losses,
losses,
e.g.
1. INTRODUCTION
INTRODUCTION
failures
occurring
maintenance
(Clarke
et
al.,
2010).
1.
INTRODUCTION
These
activities
are in
related
to the elimination
of losses,
e.g.
failures
occurring
maintenance
(Clarke
et
al.,
failures
occurring
in
maintenance
(Clarke
et
al., 2010).
2010).
1. INTRODUCTION
These
activities
related
to
the elimination
of appropriate
losses,
e.g.
Moreover,
theseare
activities
require
the
use of
of
For
the
last
few
decades,
maintenance
was
thought
as
unnecfailures
occurring
in maintenance
(Clarke
et appropriate
al., 2010).
1.
INTRODUCTION
Moreover,
these
activities
require
the
use
For
the
last
few
decades,
maintenance
was
thought
as
unnecMoreover,
these
activities
require
the
use
of
appropriate
For
the
last
few
decades,
maintenance
was
thought
as
unnecfailures
occurring
in
maintenance
(Clarke
et
al.,
2010).
methods and
andthese
toolsactivities
to support
the decision
making
process
essary
an
enterprise
management.
That
because
issue
Moreover,
require
the use making
of appropriate
For
thein
last
few
decades,
maintenance
wasis
thought
asthe
unnecmethods
tools
to
the
process
essary
in
an
enterprise
management.
That
is
because
issue
methods
andthese
toolsactivities
to support
support
the decision
decision
process
essary
inlast
an few
enterprise
management.
That
isthought
because
the
issue
Moreover,
require
the2018,
use making
of appropriate
For the
decades,
maintenance
was
asthe
unnec(Jasiulewicz-Kaczmarek
and
Żywica,
Valis
and
Mawas
limited
to
the
specific
functions
that
are
normally
used
in
methods
and
tools
to
support
the
decision
making
process
essary
in
an
enterprise
management.
That
is
because
the
issue
(Jasiulewicz-Kaczmarek
and Żywica,
Żywica,
2018, making
Valis and
and
Mawas
limited
to
the
functions
that
are
normally
used
in
and
2018,
Valis
Mawas
limited
to
the specific
specific
functions
that
are
normally
used
in (Jasiulewicz-Kaczmarek
methods
and2018).
tools to support
the decision
process
essary
in ansituations
enterprise
management.
That
is because
the
issue
zurkiewicz,
emergency
such
as
machine
failures
(Gola,
2019).
(Jasiulewicz-Kaczmarek
and
Żywica,
2018,
Valis
and
Mawas
limited
to
the
specific
functions
that
are
normally
used
in
zurkiewicz,
2018).
emergency
situations
such functions
as machine
machine
failures
(Gola, used
2019).
2018).
emergency
situations
such
as
failures
(Gola,
2019).
(Jasiulewicz-Kaczmarek
Żywica,
2018, Valis
and Mawas limitedthis
to the
specific
that
are normally
in zurkiewicz,
In the
the literature,
literature,
there are
areand
different
solutions
concerning
the
However,
practice
is
no
longer
acceptable
because
the
zurkiewicz,
2018).there
emergency
situations
such
as machine
failures (Gola,
2019).
In
different
solutions
concerning
the
However,
this
practice
is
no
longer
acceptable
because
the
In
the
literature,
are different
solutions
concerning
the
However,
this
practice
is
no
longer
acceptable
because
the
2018).there
emergency situations
such as
machine
failures (Gola,
2019). zurkiewicz,
supporting
decision
processes
in
maintenance
management.
importance
of
maintenance
has
been
recognised
as
a
strategic
the literature,
there
are different
solutions concerning
the
However,
this
practice is no
acceptable asbecause
the In
decision
processes
in
management.
importance
of maintenance
maintenance
haslonger
been recognised
recognised
strategic
supporting
decision
processes
in maintenance
maintenance
management.
importance
of
has
been
aa strategic
In
the work
literature,
there
are
different
solutions
concerning
the
However,ofthis
practice is
no
longer
acceptable asbecause
the supporting
In the
the
(Taghipour
et al.,
al.,
2011),
the authors
authors
present how
how
element
income
for
enterprises.
decision
processes
in maintenance
management.
importance
ofgenerating
maintenance
has been
recognised
asNowadays,
a strategic supporting
In
work
(Taghipour
et
2011),
the
present
element
of
generating
income
for
enterprises.
Nowadays,
In
the
work
(Taghipour
et
al.,
2011),
the
authors
present
how
element
of
generating
income
for
enterprises.
Nowadays,
supporting
decision
processes
in
maintenance
management.
importance
of
maintenance
has
been
recognised
as
a
strategic
to identify
identify
and
prioritizeetcritical
critical
devices
to
mitigate
functionmaintenance
is
considered
aa key
that
influences
In
the workand
(Taghipour
al., 2011),
the to
authors
present
how
element
of generating
income
for element
enterprises.
Nowadays,
to
prioritize
devices
mitigate
functionmaintenance
is
element
that
influences
to
identify
prioritize
devices
mitigate
functionmaintenance
is considered
considered
a key
key
element
that Nowadays,
influences
In
the workand
(Taghipour
etcritical
al.,a 2011),
the to
authors
present
how
element of competitiveness
generating
income
forcost
enterprises.
al
failures.
The
authors
used
multi-criteria
decision-making
company’s
as
its
constitutes
the
main
to
identify
and
prioritize
critical
devices
to
mitigate
functionmaintenance
is considered as
a key
element
that influences
al
failures.
The
authors
used
a
multi-criteria
decision-making
company’s
competitiveness
its
cost
constitutes
the
main
al
failures.
The
authors
used
a
multi-criteria
decision-making
company’s
competitiveness
as
its
cost
constitutes
the
main
to
identify
and
prioritize
critical
devices
to
mitigate
functionmaintenance
is
considered
a
key
element
that
influences
to
prioritize
and
to
propose
appropriate
part
of
operational
costs
of
Thus,
an
al
failures.
The
authorsdevices
used a multi-criteria
decision-making
company’s
competitiveness
asanitsenterprise.
cost constitutes
the unexmain model
model
to
prioritize
and
to
appropriate
part
of
operational
costs
of
Thus,
an
unexmodel
to The
prioritize
devices
and ‘preventive
to propose
propose
appropriate
part
ofsystem
operational
costs
ofasan
anitsenterprise.
enterprise.
Thus,
an
unexal failures.
authorsdevices
used
a multi-criteria
decision-making
company’s
competitiveness
cost
constitutes
theoffered
main maintenance
strategies
including
maintenance’
pected
failure
may
influence
the
quality
of
the
model
to
prioritize
devices
and
to
propose
appropriate
part
of
operational
costs
of
an
enterprise.
Thus,
an
unexmaintenance
strategies
including
‘preventive
maintenance’
pected
system
failure
may
influence
the
quality
of
the
offered
maintenance
strategies
including
‘preventive
maintenance’
pected
system
failure
may
influence
the
quality
of
the
offered
model
to
prioritize
devices
and
to
propose
appropriate
part
of
operational
costs
of
an
enterprise.
Thus,
an
unexand
‘user
training’.
Moreover,
in
the
works
(Bashiri
et
product,
the
availability
of
machines
or
tools,
the
environmaintenance
strategies
including
‘preventive
maintenance’
pected
system
failure
may
influence
the
quality
of
the
offered
and
‘user training’.
training’.
Moreover,
in the
the
works (Bashiri
(Bashiri
et al.,
al.,
product,
the availability
availability
ofinfluence
machines
orquality
tools,ofthe
the
environand
‘user
Moreover,
in
works
et
al.,
product,
the
of
machines
or
tools,
maintenance
strategies
including
‘preventive
maintenance’
pectedand
system
failure may
the
theenvironoffered
2011;
Zhaoyang
et
al.,
2011,
Galar
et
al.,
2012),
the
ment
an
operator
(Sobaszek
et
al.,
2017;
Burduk
and
and
‘user
training’.
Moreover,
in the et
works
(Bashiritheet imal.,
product,
the
availability
of machines
or tools,
the
environ2011;
Zhaoyang
et
al.,
2011,
Galar
al.,
2012),
imment
and
an
operator
(Sobaszek
et
al.,
2017;
Burduk
and
2011;
Zhaoyang
et
al.,
2011,
Galar
et
al.,
2012),
the
imment
and
an
operator
(Sobaszek
et
al.,
2017;
Burduk
and
and ‘useroftraining’.
Moreover, in the
works
(Bashiri et al.,
product, the 2015,
availability
ofetmachines
or tools,
the environ- portance
preventive
the
maintenance
Jagodziński,
Szwarc
al.,
2019).
implementation
Zhaoyang
et al.,maintenance
2011, Galarin
et
al.,
2012), themanimment
and an2015,
operator
(Sobaszek
et al.,The
2017;
Burduk and 2011;
portance
of
preventive
maintenance
in
the
maintenance
manJagodziński,
Szwarc
et
al.,
2019).
The
implementation
portance
of
preventive
maintenance
inthe
theal.,
maintenance
manJagodziński,
Szwarc
eta al.,
2019).
implementation
2011; Zhaoyang
et
al.,
2011, Galar
et
2012),
thebased
imment
andManufacturing
an2015,
operator
(Sobaszek
etimpact
al.,The
2017;
Burduk
and
agement
process
of
equipment
and
of
risk
of
Lean
had
great
the
change
of
portance
of
preventive
maintenance
inthe
the role
maintenance
manJagodziński,
2015, Szwarc
eta al.,
2019).
Theon
implementation
agement
process
of
equipment
and
role
of
risk
based
of
Lean
Manufacturing
had
great
impact
on
the
change
of
agement
process
of
equipment
and
the
role
of
risk
based
of
Lean
Manufacturing
had
a
great
impact
on
the
change
of
portance
of
preventive
maintenance
in
the
maintenance
manJagodziński,
2015,
Szwarc
et
al.,
2019).
The
implementation
was
shown.
perceiving
the
role
of had
maintenance
in
an on
enterprise.
This
agement
process
of equipment and the role of risk based
of
Lean Manufacturing
a great impact
the change
of maintenance
maintenance
was
perceiving
the
role
maintenance
in
enterprise.
This
maintenance
was shown.
shown.
perceiving
the
role of
of had
maintenance
in an
anmanagement
enterprise.
This
agement
process
of et
equipment
and
the
role propose
of risk to
based
of
Lean Manufacturing
a great impact
on
the change
of In
the
work
(Zeineb
al.,
2017),
the
authors
use
philosophy
is
currently
a
preferably
used
conmaintenance
was
shown.
perceiving
the
role of amaintenance
in anmanagement
enterprise. This
In
the
work
(Zeineb
et
al.,
2017),
the
authors
propose
to
use
philosophy
is
currently
preferably
used
conIn
the
work
(Zeineb
et
al.,
2017),
the
authors
propose
to
use
philosophy
is
currently
a
preferably
used
management
conmaintenance
was
shown.
perceiving
the
role
of
maintenance
in
an
enterprise.
This
the
Analytical
Hierarchy
Process,
the
technique
for
order
cept
in
enterprises.
A
number
of
organisations
have
started
In
the
work
(Zeineb
et
al.,
2017),
the
authors
propose
to
use
philosophy
is
currently
a
preferably
used
management
conthe
Analytical
Hierarchy
Process,
the
technique
for
order
cept
in
enterprises.
A
number
of
organisations
have
started
the
Analytical
Hierarchy
Process,
the
technique
for
order
cept
in
enterprises.
A
number
of
organisations
have
started
In
the
work
(Zeineb
et
al.,
2017),
the
authors
propose
to
use
philosophy
is
currently
a
preferably
used
management
conpreference
by
similarity
to
ideal
solution,
and
the
mathematiusing
“lean
tools”
mainly
in
order
to
reduce
losses
in
producthe
Analytical
Hierarchy
Process,
the
technique
for
order
cept
in
enterprises.
A
number
of
organisations
have
started
preference
by similarity
similarity
to Process,
ideal solution,
solution,
and the
the mathematimathematiusing
“lean
tools”
mainly
in
order
to
reduce
losses
in
producpreference
by
to
ideal
and
using
“lean
tools”
mainly
in
order
to
reduce
losses
in
producthe
Analytical
Hierarchy
the
technique
for
order
cept
in
enterprises.
A
number
of
organisations
have
started
cal
optimization
to
determine
the
criticality
of
equipment,
to
tion
(Gornicka
and
Burduk,
2018).
However,
it
beby similarity
to ideal
andofthe
mathematiusing
“lean tools”
in order
reduce losses
in produccal
optimization
to determine
determine
thesolution,
criticality
equipment,
to
tion
(Gornicka
andmainly
Burduk,
2018).to
However,
it quickly
quickly
be- preference
cal
optimization
to
the
criticality
ofthe
equipment,
to
tion
and
Burduk,
2018).
it
quickly
bepreference
by similarity
to ideal
solution,
and
mathematiusing(Gornicka
“lean
tools”
mainly
in order
to However,
reduce in
losses
in producrank
the
maintenance
strategies
and
to
select
the
optimal
came
apparent
that
the
presence
of
losses
manufacturing
optimization
to determine
the criticality
of equipment,
to
tion
(Gornicka
and Burduk,
2018).
However,
it quickly be- cal
rank
the
maintenance
strategies
and
to
select
the
optimal
came
apparent
that
the
presence
of
losses
in
manufacturing
rank
the
maintenance
strategies
and
to
select
the
optimal
came
apparent
that
the
presence
of
losses
in
manufacturing
cal optimization
to determine
the
criticality
of equipment,
to
tion (Gornicka
and Burduk,
2018). processes
However,such
it quickly
be- maintenance
strategy
taking
into
consideration
the
total
processes
is
influenced
by
auxiliary
as,
among
rank
the
maintenance
strategies
and
to
select
the
optimal
came
apparent
that
the
presence
of
losses
in
manufacturing
maintenance
strategy
taking
into
consideration
the
total
processes
is influenced
influenced
by
auxiliary
processes
such
as, among
among maintenance
strategy
taking
into
total
processes
is
by
auxiliary
as,
rank the maintenance
strategies
andconsideration
to
select
thethe
optimal
came apparent
that the
presence
ofprocesses
lossesorganisations
in such
manufacturing
maintenance
cost.
In
the
work
(Wang
et
al.,
2006),
the
auothers,
maintenance.
That
is
why,
some
have
maintenance cost.
strategy
taking
into
consideration
thethetotal
processes
is influencedThat
by auxiliary
processes
such as, among
In
work
(Wang
et
al.,
2006),
auothers,
is
why,
some
organisations
have
maintenance
In the
the
work
(Wang
ettool
al., to
2006),
auothers, maintenance.
maintenance.
ismethods
why, processes
some
organisations
have maintenance
strategy
taking
into
consideration
thethetotal
processes
is influencedThat
by auxiliary
suchof
as,mainteamong
thors used
used the
thecost.
Monte
Carlo
simulation
forecast
the
started
implementing
lean
in
the
area
cost.
In the
work
(Wang ettool
al., to
2006),
the auothers,
maintenance.
That
ismethods
why, some
organisations
have maintenance
thors
Monte
Carlo
simulation
forecast
the
started
implementing
lean
in
the
area
of
maintethors
used
the
Monte
Carlo
simulation
tool
to
forecast
the
started
implementing
lean
methods
in
the
area
of
maintemaintenance
cost.
In
the
work
(Wang
et
al.,
2006),
the
auothers,
maintenance.
That
is
why,
some
organisations
have
risk
of
equipment
failures.
According
to
(Cruz
et
al.,
2012),
nance
defined
as
Lean
Maintenance.
thors
used
the
Monte
Carlo
simulation
tool
to
forecast
the
started
implementing
lean methods in the area of mainte- risk
of
equipment
failures.
to tool
(Cruz
al.,
2012),
nance
as
ofused
equipment
failures.
According
(Cruz
etforecast
al.,
2012),
nance
defined
as Lean
Lean Maintenance.
Maintenance.
thors
the Monte
CarloAccording
simulation
to et
the
starteddefined
implementing
lean methods in the area of mainte- risk
in the
the
equipment
maintenance
processto
the
focus
should
be
risk
of equipment
equipment maintenance
failures. According
tothe
(Cruz
et al.,
2012),
nance defined as 2.
Lean
Maintenance.REVIEW
in
process
focus
should
be
LITERATURE
in
the
equipment
maintenance
process
the
focus
should
be
risk
of
equipment
failures.
According
to
(Cruz
et
al.,
2012),
nance defined as 2.
Lean
Maintenance.
2. LITERATURE
LITERATURE REVIEW
REVIEW
paid
on
the
risk
caused
by
equipment
failure.
In
the
work
in
the
equipment
maintenance
process
the
focus
should
be
paid
onequipment
the risk
risk caused
caused
by equipment
equipment
failure.
In should
the work
work
the
failure.
the
in
theon
maintenance
process
the
focusIn
be
2. is
LITERATURE
REVIEW activities aimed paid
(Jamshidi
et al.,
al.,
2015)
theby
authors
applied,
among
others,
the
Lean Maintenance
Maintenance
a concept
concept implementing
implementing
paid
on the
risk2015)
caused
by
equipment
failure.
Inothers,
the work
2.
LITERATURE
REVIEW
(Jamshidi
et
the
authors
applied,
among
the
Lean
is
a
activities
aimed
(Jamshidi
et
al.,
2015)
the
authors
applied,
among
others,
the
Lean
Maintenance
is
a
concept
implementing
activities
aimed
paid
on
the
risk
caused
by
equipment
failure.
In
the
work
fuzzy
failure
modes
and
effects
analysis
method
to
prioritize
at increasing
increasing
the efficiency
efficiency
and
effectiveness
of technical
technical
et al.,
2015)
authors
applied,
amongtoothers,
the
Lean
Maintenance
is a conceptand
implementing
activities
aimed (Jamshidi
fuzzy
failure
modes
andthe
effects
analysis
method
prioritize
at
the
effectiveness
of
modes
and
analysis
method
prioritize
at
increasing
the
efficiency
and
effectiveness
of
technical
(Jamshidi
et al.,
theeffects
authors
applied,
among
others,
the
Lean
Maintenance
is a concept
implementing
activities
aimed fuzzy
devicesfailure
and
to 2015)
facilitate
the classification
classification
of to
devices
for
infrastructure
(Jasiulewicz
and
Saniuk,
2018).
fuzzy
failure
modes
and
effects
analysis
method
to
prioritize
at
increasing
the
efficiency
and
effectiveness
of
technical
devices
and
to
facilitate
the
of
devices
for
infrastructure
(Jasiulewicz
andand
Saniuk,
2018). of technical devices
and modes
to facilitate
the classification
of to
devices
for
infrastructure
and
Saniuk,
2018).
fuzzy failure
and effects
analysis method
prioritize
at increasing (Jasiulewicz
the efficiency
effectiveness
devicesbyand
to facilitate
the reserved.
classification of devices for
infrastructure
(Jasiulewicz
and Saniuk,
2018).of Automatic Control) Hosting
2405-8963
© 2019,
IFAC (International
Federation
Elsevier
Ltd. All rights
devices and
to facilitate
the classification of devices for
infrastructure
(Jasiulewicz
and Saniuk,
2018).
Peer review under
responsibility of International Federation of Automatic
[email protected]
2019 IFAC
148Control.
[email protected]
2019 IFAC
IFAC
148
10.1016/j.ifacol.2019.10.037
[email protected]
2019
148
[email protected] 2019 IFAC
148
[email protected] 2019 IFAC
148
2019 IFAC IMS
August 12-14, 2019. Oshawa, Canada
Katarzyna Antosz et al. / IFAC PapersOnLine 52-10 (2019) 148–153
maintenance activities based on their criticality. In the work
(Bayar et al., 2016) the authors propose to use ontologies and
multi-agent systems as an approach for disruptions and risks
monitoring. The system was designed to support a decision
making process. In the work (Carnero, 2016) the author proposes to use a multicriteria decision making approach to improve maintenance policies in a healthcare organization
where the condition of healthcare equipment may influence
human life. In turn, in the work (Li et al., 2014) the authors
incorporated machine learning techniques together with the
distributed learning and hierarchical analytical approaches to
explore historical and real-time data concerning railway conditions and to predict failures. Additionally in the work (Loska, 2013) the methodology of operational modelling of the
decision-making process with the use of scenario methods
was proposed. Although there are many methods and tools
supporting the decision making process in maintenance management, but in the current research, the use of intelligent
systems which are dedicated to support the implementation of
the Lean Maintenance concept is often not taken into consideration.
The main goal of this work was to present the possibility of
using intelligent systems to support decision-making processes in the Lean Maintenance concept, which, allows to increase the operational efficiency of the company's technical
infrastructure. The third section of the article presents the
results of research on the use and effectiveness of Lean
Maintenance in production enterprises. The aim of the research was to gather information on technical infrastructure
management systems in enterprises, with particular emphasis
on Lean Maintenance methods and tools as well as the ability
to identify factors affecting the efficiency of their application.
The fourth section presents the use of artificial intelligence
methods (decision trees and the theory of rough sets) to
search for dependencies between specific activities implemented as part of the Lean Maintenance implementation and
the obtained effects. The last section of this paper concludes
the work and presents its limitations and as well as the future
research.
149
results of the studies were shown in the work (Antosz and
Stadnicka, 2014).
Based on the studies conducted in the enterprises of
podkarpackie voivodship and their results considering the
identification of the ways of technical infrastructure
management, on the activities undertaken as part of the Lean
Maintenance implementation in enterprises as well as the
assessment of the effectiveness of the technical infrastructure
management ways of enterprises, it was able to draw the
following conclusions:
- The analysed enterprises identify different kinds of wastes
in the technical infrastructure management process. In
order to eliminate wastes more and more enterprises
implement Lean Maintenance philosophy.
- The companies that have identified the wastes such as
machine failures or long changeover times plan to
implement or have implemented different Lean
Maintenance methods and tools related to the machine
use and the improvement of the organization of work
stands.
- The number of enterprises that implement TPM is
becoming smaller. Mostly, medium enterprises resign
from the implementation of the TPM method.
- The research proved that most companies decided to
implement TPM because of high machine failure rate, and
the TPM system is most often implemented on all
machines if the mean time to repair ranges between 1 to 8
hours.
- The enterprises realize different activities in terms of the
TPM method implementation. The presented results show
that in large enterprises the kind of activities within this
method implementation has practically not changed.
- The most common problems that appear during the TPM
implementation include the lack of skilled workers to
eliminate irregularities on machines and the lack of
engagement of the top management into the activities
related to TPM.
- Enterprises implement 5S method increasingly. However,
less and less companies declare to have implemented 5S
in the whole enterprise. They notice benefits from the
implementation of 5S method, and its effectiveness is
mainly assessed with 5S audits.
- The results show that the knowledge and use of the
SMED (Single Minute Exchange of Die) method have not
significantly changed for the last few years. SMED is
most often implemented in small and micro enterprises.
However, only among large companies there were such
companies that analysed all changeovers. Enterprises
most often assess the effectiveness of the SMED method
by the changeover time reduction.
- Most of the analysed enterprises still don’t use the OEE
(Overall Equipment Effectiveness) indicator. This
indicator is used in medium enterprises. It is most
frequently used in the enterprises where a single
production is used and in the enterprises from aviation,
metalworking and automotive industries. The lowest
values of the OEE indicator, below 30%, are obtained in
small enterprises, in the metalworking industry and in the
enterprises with Polish capital.
3. THE ASSESSMENT OF LEAN MAINTENANCE
EFFECTIVENESS CONCEPT IN ENTERPRISES –
STUDY RESULTS
The first stage of the study was to collect information
considering the systems of technical infrastructure
management, in particular, the methods and tools of Lean
Maintenance such: TPM (Total Productive Maintenence),
SMED (Single Minute Exchange of Die), 5S and OEE
(Overall Equipment Effectiveness) indicator, and the
possibility to identify the factors that influence the
effectiveness of their use. The studies were carried out in 150
manufacturing enterprises in podkarpackie voivodship in two
stages: stage I – years 2010 to 2014 and stage II in years 2014
to 2017. The surveyed enterprises were classified according
to the following criteria: an organisation size, a production
type, an industry, an ownership type, capital, a company’s
condition and the type of the owned machines. Among the
analysed enterprises, there were ones that realized a few
production types or operated in a few industries. The detailed
149
2019 IFAC IMS
150
August 12-14, 2019. Oshawa, Canada
Katarzyna Antosz et al. / IFAC PapersOnLine 52-10 (2019) 148–153
-
Large enterprises achieved the greatest benefits from
implementing Lean Maintenance tools. In most of the
analysed areas they achieved higher benefits than
expected.
Additionally, the statistical analysis with using chi square test
of the obtained results, allowed to identify the factors that
influence
the
Lean
Maintenance
implementation
effectiveness. On the basis of the studies it was drawn the
following conclusions: the factors that influence the use of
Lean Maintenance methods and tools are: the enterprise size,
industry and the capital owned, the factors that influence the
SMED method implementation are: the company’s condition,
the kind and means of supervising, and the type of machines
owned, the factors that influence the 5S method
implementation are: the company’s condition and kind of
supervision, the main factor influencing the TPM method
implementation is the type of machines owned, the activities
realized within the TPM method implementation depend
mainly on the type of ownership and the company’s
condition, the classification of machines and spare parts
realized within TPM depends on the company’s size, its
capital, type of the machines owned and kind of supervision.,
the reduction of changeover times depends on an enterprise
type, the OEE value depends on the ownership type,
enterprise industry and mean time to repair, the reduction of
unexpected downtimes depends on the implementation of the
SMED method and the type of machines owned.
The studies allowed to identify the factors that influence the
achieved results after the Lean Maintenance implementation
in enterprises. It should be noted that the studies often
showed that single factors don’t have a significant impact on
the studied areas, although their interaction with other factors
may have a substantial impact on the analysed area.
However, the problem is that analysing a process with so
many variables is very difficult. Therefore, it was proposed
the concept of using the method of building and analysing
decision trees, and the theory of rough sets for the
quantitative interpretation of the degree Lean Maintenance is
used.
turing enterprises. The 13 decision rules were generated for
predicting the value of the variable “OEE value”. In order to
assess the quality of the generated decision rules, the rules
were validated. The aim of the validation was to prove in a
documented manner and in accordance with the assumptions
that the generated decision rules actually lead to the planned
results.
Fig. 1. Calculation of the OEE indicator in enterprises.
The assessment of the developed decision rules was
conducted in the following stages: re-conducting the surveys
in 20 randomly chosen enterprises, developing an expert
system based on the generated decision rules (within the
framework programme PC-Shell of Aitech Sphinx suite),
using the survey results to study general classification ability
of the generated decision rules with the developed expert
system, and assessing qualitatively the obtained results with
quality classification measures.
The structure of the created knowledge base in the expert
system consists of two blocks: facets and rules. The facet
block is used to declare the decision attributes and their
values. Decision attributes are explanatory variables placed in
the nodes of the decision tree. In addition, for each decision
tree, the so-called the result attribute (target attribute) that
represents the result of the system inference. After
completing the inference, the application presents the
determined value of the resulting attribute "OEE value"
placed in a separate window. On the figure 2 the result of the
expert system consultation window is presented. On the
figure 3 the window of justifying the result of applying the
expert system is presented. The quality analysis consisted of
developing binary matrices of errors for the classifier for the
classes that most commonly appear in the conducted studies.
In the developed binary matrices, the class analysed at a
particular moment was assumed positive (TP – true positive)
while the remaining classes were negative (TN – true
negative).
4. DECISION TREE AND THE THEORY OF ROUGH
SETS TO SUPPORT THE DECISION – MAKING
PROCESS IN LEAN MAINTENANCE MANAGEMENT
Decision tree (classification) were developed for decision
variable: the average value of OEE after the Lean Maintenance implementation. The indicator is one of the indicators
recommended in the literature for the assessment of the efficiency of the machines owned and the TPM implementation.
It was designed a classification tree for enterprises which
were analysed a given indicator and implemented the TPM
method. It was developed a classification tree for a dependent
variable – an average value of OEE – 24 out of the studied
group of enterprises. The study results shows, that most of
the analysed enterprises still don’t use the OEE indicator
(stage I: 60,38%, stage II: 73,96%). This indicator is used for
all machines merely by a few percent of the enterprises (stage
I: 7,55%, stage II: 5,21%) (Fig. 1).
The developed classifier allowed to generate a set of decision
rules that may be the basis for determining the directions and
effects of the Lean Maintenance implementation in manufac150
2019 IFAC IMS
August 12-14, 2019. Oshawa, Canada
Katarzyna Antosz et al. / IFAC PapersOnLine 52-10 (2019) 148–153
Fig. 2. The result of the expert system consultation window.
of the lack of an answer in the survey. The rough set theory
uses decision tables that include a higher number of items.
Therefore, it considers a greater number of data while
generating the rules. This fact allowed to discover new
relations between the explaining and explained variables that
had not been discovered with the decision trees.
The decision rules were generated based on the proposed
algorithm LEM2 (Learning from Examples Module, version
2), together with RSES software. Then, the assessment of
their correctness was done. It was conducted the assessment
of the developed decision rules in the following stages:
generating a decision table and error matrix, developing an
expert system based on the generated decision rules,
application of the obtained results of the survey for
examining a general classification ability, applying the
obtained survey results for examining a general classification
ability of the generated decision rules with the developed
expert system, and assessing the quality of the obtained
results with quality classification measures.
Fig. 3. The window of justifying the result of applying the
expert system.
Table 1 and 2 present error matrices for the classifier - the
value of OEE for the two most-emerging classes: 30-50% and
70-85%. Based on a binary matrix of errors, we can designate
numerical indicators (Table 3). In detail, these indicators
have been presented and discussed, among others in the
works (Costa et al., 2007, Fawcelt, 2006; Sokolova and
Lapalme, 2009).
Table 3. Indicators used to test the quality of classifier (Pasko
and Setlak, 2016).
Ind.
Designation
Formula
TP  TN
Acc 
Acc
accuracy,
TP  TN  FP  FN
overall
error
FP  FN
Err 
Err
rate,
TP  TN  FP  FN
true positives
TP
TPR 
TPR rate,
TP  FN
Table 1. Matrix of errors for the classifier value of OEE
indicator - 30 - 50% class.
Real classes
TP
TN
TP
7
0
Predicted classes
TN
0
13
Table 2. Matrix of errors for the classifier value of OEE
indicator - 70-85% class.
Real classes
TP
TN
151
TNR
Predicted classes
TP
TN
5
1
2
12
PPV
The obtained indicator values for the assessment of
classification measure, e.g. of an error (Err) at the level of
0,00 to 0,15 proved high usefulness of the developed
classifiers, and thereby, their possibility to apply them by
manufacturing enterprises for the effectiveness assessment of
the Lean Maintenance methods and tools implementation.
The rough set theory was used to assess the degree of the
Lean Maintenance application. The conducted studies
described are based on the same input data sets as were used
for decision tree. However, it was suggested a means of
improving the accuracy of the obtained assessment through
incorporating the incomplete data.
Data considered
organisational and technical area of activity of the most
enterprises. Applying the rough set theory allowed to extend
the group of the considered enterprises, what increased the
scope of the data used for calculating and for the inference
process. For instance, while generating the decision rules for
the described variable “an average OEE value” with a
decision tree, 24 enterprises were taken into consideration.
Applying the rough set theory allowed to increase the group
of the considered enterprises to 34. The additional 10
enterprises were described with a variable set, of which at
least one variable doesn’t have a designated value as a result
NPV
FPR
FDR
FNR
true negatives
rate,
positive
predictive
value,
negative
predictive
value
false positive
rate
false discovery
rate,
false negatives
rate
TNR 
TN
TN  FP
PPV 
TP
TP  FP
NPV 
TN
TN  FN
FPR 
FP
 1  TNR
FP  TN
FDR 
FP
FP  TP
FNR 
FN
 1  TPR
TP  FN
MCC
Matthew’s
correlation
coefficient
MCC 
F1
F1-score
F1 
J
Youden’s
statistic
J
TP  TN  FP  FN
(TP  FN )(TP  FP)( FN  TN )( FP  TN )
2  PPV  TPR
PPV  TPR
J  TPR  TNR  1
The figure 4 presents the RSES program window with the
error matrix for the classification of learning objects based on
the decision rules generated by the LEM2 algorithm for the
variable "Average value of OEE".
151
2019 IFAC IMS
152
August 12-14, 2019. Oshawa, Canada
Katarzyna Antosz et al. / IFAC PapersOnLine 52-10 (2019) 148–153
gence methods (decision tree and the theory of rough sets) to
apply them in manufacturing enterprises for the effectiveness
assessment of the Lean Maintenance methods and tools implementation is presented. The positive results obtained during the conduct of the described research lead to the conclusion that the activities in these areas should be continued. In
particular, they ought to be: studies considering the assessment of the effectiveness of using other methods and tools
recommended in the literature within the Lean Maintenance
implementation, the possibility of extending functionality
designed in a computer application, the use of other methods
of data exploration for generating decision rules and comparing their classification quality.
Fig. 4. The RSES program window with the error matrix for
the variable "Average value of OEE".
The indicator values obtained for examining the classifier
quality, e.g. the error (Err) at the level of 0,10 and Acc at the
level of 0,90 to 1,00, proved high usefulness of the decision
rules generated with the rough sets, and, thereby, their
applicability by manufacturing enterprises for the assessment
of the Lean Maintenance effectiveness. Table 4 shows the
comparison of the achieved values of the indicators used for
the assessment of the classifier quality for the models
obtained with decision tree (DT) and rough set theory (RST).
Table 4. Comparison of the achieved values for the models
obtained with the decision trees (DT) and the rough set
theory (RST).
Indicators
Acc
TPR
TNR
PPV
NPV
MCC
F1
J
Err
FPR
FDR
FNR
REFERENCES
Antosz, K. (2018). Maintenance – identification and analysis
of the competency gap. Eksploatacja i Niezawodnosc –
Maintenance and Reliability, vol. 20 (3), pp. 484–
494.
Antosz, K., Stadnicka, D. (2014). The results of the study
concerning the identification of the activities realized in
the management of the technical infrastructure in large
enterprises. Eksploatacja i Niezawodnosc-Maintenance
and Reliability. vol. 16 (1), pp.112-119.
Bashiri, M., Badri, H., Hejazi, T.H. (2011). Selecting optimum maintenance strategy by fuzzy interactive linear assignment method. App. Math. Mod. vol. 35 (1), pp. 152164.
Bayar, N., Darmoul, S., Hajri-Gabouj, S., Pierreval, H.
(2016). Using immune designed ontologies to monitor
disruptions in manufacturing systems. Com in Ind, vol.
81, pp. 67-81.
Burduk, A., Jagodziński, M. (2015). Assessment of Production System Stability with the Use of the FMEA Analysis and Simulation Models. In: Jackowski K., Burduk R.,
Walkowiak K., Wozniak M., Yin H. (eds) Intelligent Data Engineering and Automated Learning – IDEAL 2015.
Lecture Notes in Computer Science, vol. 9375. Springer,
Cham, pp. 216-223.
Carnero, M.C., Gomez, A. (2016) A multicriteria decision
making approach applied to improving maintenance policies in healthcare organizations. BMC Med Inform Decis
Mak, vol. 16, Springer, London.
Clarke, G., Mulryan, G, Liggan P. (2010), Lean Maintenance
– A Risk-Based Approach, Pharmaceutical Engineering,
The Official Magazine of ISPE, 30(5), pp. 1-6.
Costa E.P., Lorena A.C., Carvalho A.C.P.L.F., Freitas A.A.
(2007). A review of performance evaluation measures
Quantitative techniques for medical equipment maintenance management for hierarchical classifiers. Evaluation Methods for Machine Learning II: papers from the
AAAI-2007 Workshop, AAAI Press, pp.182÷196.
Cruz, A.M., Rincon A. M.R. (2012). Medical device maintenance outsourcing: Have operation management research
and management theories forgotten the medical engineering community? A mapping review. Eur. J. Op.
Res. vol. 221(1), pp. 186-197.
Fawcelt T. (2006). An introduction to ROC analysis. Pattern
Recognition Letters, Vol. 27(8) Elsevier, New York, pp.
861÷874.
Classifier: an average OEE value
Marked class
30-50%
70-85%
DT
RST
DT
RST
1,00
0,90
0,85
0,90
1,00
1,00
0,83
1,00
1,00
0,88
0,86
0,86
1,00
0,60
0,71
0,75
1,00
1,00
0,92
1,00
1,00
0,73
0,66
0,80
1,00
0,75
0,77
0,86
1,00
0,88
0,69
0,86
0,00
0,10
0,15
0,10
0,00
1,00
0,67
1,00
0,00
0,40
0,29
0,25
0,00
0,00
0,17
0,00
The indicators that achieved more favourable values were
marked green. When the values achieved for particular
indicators were identical, they were marked yellow. While
analysing the presented results, it should be noticed that the
models generated with the rough set theory achieved much
better results than in decision trees. The decision rules
indicated only better values for all indicators in OEE
classifier for the 30-50% class. The obtained results prove the
possibility to apply decision trees and the rough set theory for
supporting the decision making process in Lean Maintenance
management.
6. CONCLUSIONS
The work presents the possibility of using intelligent systems
to support decision-making processes in the Lean Maintenance management. In the work the use of artificial intelli152
2019 IFAC IMS
August 12-14, 2019. Oshawa, Canada
Katarzyna Antosz et al. / IFAC PapersOnLine 52-10 (2019) 148–153
Galar, D., Gustafson, A., Tormos, B., Berges, L. (2012).
Maintenance decision making based on different types of
data fusion, Eksploatacja i Niezawodnosc-Maintenance
and Reliability, vol. 2, pp. 135-144.
Gola, A. (2019). Reliability analysis of reconfigurable manufacturing system structures using computer simulation
methods. Eksploatacja i Niezawodnosc – Maintenance
and Reliability, 21 (1): 90–102.
Gornicka, D., Burduk, A. (2018). Improvement of Production
Processes with the Use of Simulation Models. in Intelligent Systems and Computing vol.657, pp. 265-274.
Jamshidi, A., Rahimi, S.A., Aitkadi, D., Ruiz, A. (2015). A
comprehensive fuzzy risk-based maintenance framework
for prioritization of medical devices. App. Soft Comp.
vol. 32, pp. 322-334.
Jasiulewicz-Kaczmarek M., Saniuk, A. (2018), How to make
maintenance processes more efficient using lean tools?
Book Series: Advances in Intelligent Systems and Computing, , Springer, Cham, vol. 605, pp. 9 – 20.
Jasiulewicz-Kaczmarek, M., Żywica P. (2018). The concept
of maintenance sustainability performance assessment by
integrating balanced scorecard with non-additive fuzzy
integral. Eksploatacja i Niezawodnosc – Maintenance
and Reliability, 20 (4): 650–661.
Li, H.F., Parikh, D., He, Q., Qian, B.Y., Li, Z.G., Fang, D.P.,
Hampapur, A. (2014). Improving rail network velocity:
A machine learning approach to predictive maintenance.
Transportation Research Part C-Emerging Technologies, vol. 45, pp. 17-26.
Loska A. (2013). Exploitation assessment of selected technical objects using taxonomic methods. Eksploatacja i
Niezawodnosc – Maintenance and Reliability, 15(1), pp.
1-8.
Pasko Ł., Setlak G. (2016). Badanie jakości predykcyjnej
segmentacji rynku. Zeszyty Naukowe Politechniki
Śląskiej, Seria Informatyka, 37, 1 (123): 83-97 (in
polish).
Sobaszek Ł., Gola A., Kozłowski E. (2017). Application of
survival function in robust scheduling of production jobs,
[in:] Ganzha M., Maciaszek M., Paprzycki M., Proceedings of the 2017 Federated Conference on Computer
Science and Information Systems (FEDCSIS), IEEE,
New York, pp. 575-578.
Sokolova M., Lapalme G. (2009). A systematic analysis of
performance measures for classification tasks. Information Processing and Management, Vol. 45 (4), pp.
427÷437.
Szwarc E., Bocewicz G., Banaszak Z., Wikarek J. (2019)
Competence allocation planning robust to unexpected
staff absenteeism, Eksploatacja i Niezawodnosc - Maintenance and Reliability vol. 21(3), pp. 440-450. DOI:
10.17531/ein.2019.3.10
Taghipour, S., Banjevic, D., Jardine, A. (2011). Prioritization
of medical equipment for maintenance decisions. J. Oper
Res Soc, vol. 62 (9), pp. 1666–1687.
Valis D., Mazurkiewicz D. (2018) Application of selected
Levy processes for degradation modelling of long range
mine belt using real-time data, Archives of Civil And Mechanical Engineering, vol. 18(4), pp. 1430-1440.
153
Wang, B., Furst, E., Cohen, T., Keil, O.R., Ridgway, M.,
Stiefel, R. (2006). Medical Equipment Management
Strategies. Biom. Instr.Tech., vol. 40(3), pp.233-237.
Zeineb, B. H., Malek, M., Ahmad, Al H., Ikram, K., Faouzi,
M. (2017). Quantitative techniques for medical equipment maintenance management. Eur. J. of Ind. Eng., vol.
10(6), pp. 703-723.
Zhaoyang, T., Jianfeng L., Zongzhi W., Jianhu Z. Weifeng H.
(2011). An evaluation of maintenance strategy using risk
based inspection. Saf. Sci. vol. 49 (6), pp. 852-860.
153
Descargar