4 ways generative AI addresses manufacturing challenges

The
manufacturing
industry
is
in
an
unenviable
position.
Facing
a
constant
onslaught
of
cost
pressures,
supply
chain
volatility
and
disruptive
technologies
like
3D
printing
and
IoT.
The
industry
must
continually
optimize
process,
improve
efficiency,
and
improve
overall
equipment
effectiveness.

At
the
same
time,
there
is
this
huge
sustainability
and
energy
transition
wave.
Manufacturers
are
being
called
to
reduce
their
carbon
footprint,
adopt
circular
economy
practices
and
become
more
eco-friendly
in
general.

And
manufacturers
face
pressure
to
constantly
innovate
while
ensuring
stability
and
safety.
An
inaccurate
AI
prediction
in
a
marketing
campaign
is
a
minor
nuisance,
but
an
inaccurate
AI
prediction
on
a
manufacturing
shopfloor
can
be
fatal.

Technology
and
disruption
are
not
new
to
manufacturers,
but
the
primary
problem
is
that
what
works
well
in
theory
often
fails
in
practice.
For
example,
as
manufacturers,
we
create
a
knowledge
base,
but
no
one
can
find
anything
without
spending
hours
searching
and
browsing
through
the
contents.
Or
we
create
a
data
lake,
which
quickly
degenerates
to
a
data
swamp.
Or
we
keep
adding
applications,
so
our
technical
debt
continues
to
increase.
But
we
are
unable
to
modernize
our
applications,
as
logic
that
is
developed
over
the
years
is
hidden
there.

The
solution
lies
in
generative
AI
 

Let’s
explore
some
of
the
capabilities
or
use
cases
where
we
see
the
most
traction:

1.
Summarization

Summarization
remains
the
top
use
case
for
generative
AI
(gen
AI)
technology.
Coupled
with
search
and
multi-modal
interaction,
gen
AI
makes
a
great
assistant.
 Manufacturers
use
summarization
in
different
ways.

They
may
use
it
to
design
a
better
way
for
operators
to
retrieve
the
correct
information
quickly
and
effectively
from
the
vast
repository
of
operating
manuals,
SOPs,
logbooks,
past
incidents
and
more.
This
allows
employees
to
focus
more
on
their
tasks
and
make
progress
without
unnecessary
delays.

IBM®
has
gen
AI
accelerators
focused
on
manufacturing
to
do
this.
Additionally,
these
accelerators
are
pre-integrated
with
various
cloud
AI
services
and
recommend
the
best
LLM
(large
language
model)
for
their
domain.

Summarization
also
helps
in
n
harsh
operating
environments.
If
the
machine
or
equipment
fails,
the
maintenance
engineers
can
use
gen
AI
to
quickly
diagnose
problems
based
on
the
maintenance
manual
and
an
analysis
of
the
process
parameters.

2.
Contextual
data
understanding

Data
systems
often
cause
major
problems
in
manufacturing
firms.
They
are
often
disparate,
siloed,
and
multi-modal.
Various
initiatives
to
create
a
knowledge
graph
of
these
systems
have
been
only
partially
successful
due
to
the
depth
of
legacy
knowledge,
incomplete
documentation
and
technical
debt
incurred
over
decades.

IBM
developed
an
AI-powered

Knowledge
Discovery
system
that
use
generative
AI
to
unlock
new
insights
and
accelerate
data-driven
decisions
with
contextualized
industrial
data.
IBM
also
developed
an
accelerator
for
context-aware
feature
engineering
in
the
industrial
domain.
This
enables
real-time
visibility
into
process
states
(normal/abnormal),
alleviates
frequent
process
obstructions,
and
detects
and
predicts
golden
batch.

IBM
built
a
workforce
advisor
that
uses
summarization
and
contextual
data
understanding
with
intent
detection
and
multi-modal
interaction.
Operators
and
plant
engineers
can
use
this
to
quickly
zero
in
on
a
problem
area.
Users
can
ask
questions
by
speech,
text,
and
pointing,
and
the
gen
AI
advisor
will
process
it
and
provide
a
response,
while
having
awareness
of
the
context.
This
reduces
the
cognitive
burden
on
the
users
by
helping
them
do
a
root
cause
analysis
faster,
thus reducing
their
time
and
effort.

3.
Coding
Assistance

Gen
AI
also
helps
with
coding,
including
code
documentation,
code
modernization,
and
code
development.
As
an
example
of
how
gen
AI
helps
with
IT
modernization,
consider
the
Water
Corporation
use
case.

Water
Corporation
adopted
Watson
Code
Assistant,
which
is
powered
by
IBM’s
gen
AI
capabilities,
to
help
their
transition
into
a
cloud-based
SAP
infrastructure.

This
tool
accelerated
code
development
by
using
AI-generated
recommendations
based
on
natural
language
inputs,
significantly
reducing
deployment
times
and
manual
labor.
With
Watson
Code
Assistant,
Water
Corporation
achieved
a
30%
reduction
in
development
efforts
and
associated
costs
while
maintaining
code
quality
and
transparency.

4.
Asset
Management

Gen
AI
has
the
power
to
transform
asset
management.

Generative
AI
can
create
foundation
models
for
assets.
When
we
must
predict
multiple
KPIs
on
the
same
process
or
there
is
a
fleet
of
similar
assets.
It
is
better
to
develop
one
foundation
model
of
the
asset
and
fine-tune
it
multiple
times.

Gen
AI
can
also
train
for
predictive
maintenance.
Foundation
models
are
very
handy
if
failure
data
is
scarce.
Traditional
AI
models
need
lots
of
labels
to
provide
reasonable
accuracy.
However,
in
foundation
models,
we
can
pretrain
models
without
any
labels
and
fine-tune
with
the
limited
labels.

Also,
generative
AI
can
provide
technician
support
and
training.
Manufacturers
can
use
gen
AI
technologies
to
create
a
training
simulator
for
the
operators
and
the
technicians.
Further,
during
the
repair
process,
gen
AI
technologies
can
provide
guidance
and
generate
the
best
repair
procedure.

Build
new
digital
capabilities
with
generative
AI

IBM
believes
that
the
agility,
flexibility,
and
scalability
that
is
afforded
by
generative
AI
technologies
will
significantly
accelerate
digitalization
initiatives
in
the
manufacturing
industry.

Generative
AI
empowers
enterprises
at
the
strategic
core
of
their
business.

Within
two
years,
foundation
models
will
power
about
a
third
of
AI
within
enterprise
environments.

In
IBM’s
early
work
applying
foundation
models,
time
to
value
is up
to
70%
faster
than
a
traditional
AI
approach.
Generative
AI
makes
other
AI
and
analytics
technologies
more
consumable,
which
helps
manufacturing
enterprises
realize
the
value
of
their
investments.

Build
new
digital
capabilities
with
generative
AI