All Episodes

June 18, 2024 28 mins

The manufacturing process is a carefully orchestrated system where each step is as important as the next. But oftentimes there is limited real-time inspection of parts, and defects are detected too late or missed. Enter Eigen Innovations, the Intel-supported AI system that allows workers to be more efficient and helps manufacturers avoid losing money on returns and recalls of defective products. In this episode, Eigen executive Jon Weiss discusses what’s next at the intersection of manufacturing and technology, including the crucial role Intel will play in an essential industry that drives the global economy.

Learn more about how Intel is leading the charge in the AI Revolution at

See for privacy information.

Mark as Played

Episode Transcript

Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:04):
Take a second to think about every single item in
your home. Your television, your refrigerator, your desk, lamp, your laptop,
even the smartphone you might be using to hear my
voice right now. All of these things, and so many
more items in our lives, began in a factory. There
are more than six hundred and twenty thousand manufacturing businesses

in the United States right now, responsible for nearly twelve
percent of the total US economic output. The numbers are
even more staggering in China, which makes up nearly twenty
nine percent of the total global output. For manufacturing. Factories
have been around since the late eighteenth century, and today
they're used everywhere from South Korea to southern California to

make cars, airplanes, textiles, and even space vehicles, and each
one depends on a carefully choreographed system of steps, each
one as essential as the next before the final product
rolls off the production line. Mistakes, however, are also an
unavoidable part of this process. Manufacturers simply can't check every

piece of every product, and it's nearly impossible to achieve
perfection when some manufacturing plants produce thousands of items a day.
So how can technology help an industry so crucial to
our daily lives, how can factories use AI to reduce
and even prevent defective products? Welcome to Technically Speaking, an

Intel podcast produced by iHeartMedia's Ruby Studio in partnership with Intel.
In every episode, we explore how AI innovations are changing
the world and revolutionizing the way we live. Hey there,
I'm gram class, and today we're headed into the world
of manufacturing, an expansive and essential industry that drives the

global economy and both the history dating back nearly two
hundred and fifty years, we've seen manufacturing create a revolution,
resurrect nation's economies, connect people around the globe, and even
send mankind into space. But what's next at the intersection
of manufacturing and technology. In this episode, we'll be focusing

on how AI technology can help optimize manufacturing and improve
quality thanks to no small part to the minds at
Intel and at Eigen Innovations, a company committed to helping
organizations unlock the power of machine vision to automate quality inspections.
Before we go any further, let's welcome our guest joining

us today is John Weiss, the chief revenue officer at
Eigen innovations. John oversees all revenue generation activities at Eigen,
including driving sales in Eigen's machine vision software and engineering services.

Speaker 2 (02:53):
Welcome to the show, John, thanks for having me. Graham's
great to be here.

Speaker 1 (03:00):
Let's start with a bit of background on manufacturing and
the role it plays in our society. I mean it's
fair to say that I phone, our car, laptop, even
the food we eat involves some sort of manufacturing process.
I'd like to get your thoughts on just the importance
and scale of manufacturing plants around the world.

Speaker 2 (03:17):
Yeah, sure, Well, like you said, just about everything in
our daily lives comes from factories or plants. But sure,
depending on if you commute on a train or in
a car, lots of those components are coming from factories.
Very little these days are really kind of hand crafted
and handmade and smile batch, especially large scale consumer items.

And there's many different types of processes and many different
types of ways things are made.

Speaker 1 (03:46):
And look, I know there's a multitude of ways and
types of manufacturing processes. Like a Volkswagen built in Germany
is going to be very different from an iPhone built
in China. Do you find common things, reads or similarities
across manufacturing industries.

Speaker 2 (04:05):
Yeah, definitely, so, I guess maybe as a kind of
a starting point, it's important to understand there are two
types of manufacturing processes or approaches. One is called process manufacturing.
This is things like chemicals, plastics, things that can't really
be broken down or deconstructed easily. Or you have discrete manufacturing,
which is much more of the process of putting stuff together.

Think about a watch or a car. Now, both of
those processes, discrete and process manufacturing, they're quite different, but
there are certainly similarities. And between the two methods, you
basically have all of the things that we use every day,
right then oftentimes actually they kind of bleed into one another.
Most things have a little bit of process manufacturing involved
and then a little bit of discrete manufacturing as well. However,

I would say the commonalities across both are really heavily
reliant on technology. We see a very large push for
data driven decision making. We see large patterns or trends
in both realms of manufacturing around empowering the workforce, trying
to opt skill workers via technology to get them to

be focused on more mission critical tasks or higher value
activities while letting some of the technology do more of
the mundane tasks.

Speaker 1 (05:21):
Yeah. In a previous job that I had, we were
doing consumer electronics, and we struggled quite a bit with
the quality side of things and being able to ensure
a good product can you manufactured in our China plant.
And one thing that struck me was that there was
a very manual process in terms of the quality inspection,

and it will take samples in one out of every
ten and the test that and then if that worked,
then okay, and then we assume the rest kind of work.
I'm wondering if you could share any stories or examples
of I guess problems with quality or defective products that
stick in your mind.

Speaker 2 (05:59):
Yeahsolutely. I mean that's all we do at Eigen, right.
All we do is industrial machine vision for inline quality inspection.
So a couple that stick into my mind. Actually very
relevant to what you said, sample testing. Lots of manufacturers
do this, right if they're a high volume shop or
a high volume process. For example, we have some customers
that use our technology to inspect upwards of forty thousand

units a week per facility. The challenge is if you
do find a problem, now you're kind of scratching your
head wondering how many in between the last one hundred
or last fifty also had a problem right, And unfortunately
you tend to find out the hard way when you
get returns or warranty claims that maybe something wasn't right
in that process. And so technology is a great way,

especially the arena that we operate in computer vision, we're
helping customers actually get away from that. A great example
is one of our manufacturing customers who makes fuel tanks
for a variety of different vehicles, and they do what's
called destructive testing. They don't just test, they actually break
the fuel tank, that cut it up, and they look

at all of the plastic components inside and they see
was it molded correctly, was it welded correctly? And if
they have a problem, well, now they have to reverse
engineer a whole bunch of stuff and try to figure out,
holy cow, what went wrong right? And how do we
ensure that no bad fuel tank gets on a truck.
They started the journey with us about three years ago,
and fast forward today, we're builds back on every new

machine that gets put into those plants for fuel tank inspection.
So they know unequivocally, every single product that they ship
out the door is of the highest quality standard and
if it's not. If something happens, now they have complete
traceability on everything they've made, so they can figure out
exactly what went wrong in the process.

Speaker 1 (07:42):
What John is talking about here is the output of
the manufacturing process. How can we ensure every fuel tank
that leaves the plant will work as designed? Just as importantly,
we need to consider the quality of the input components.
Everything from the greater steel to the precision of the
fuel gate, These need to be expected to ensure that

these are up to the manufacturer standard. I asked John
for his thoughts about this.

Speaker 2 (08:11):
We don't often look at raw material although it's possible
in some cases, but more often than not our inputs
that we're looking at it's actually process inputs or parameters.
So we're looking at feed rates of raw materials, temperatures
of raw materials, things like this. In the process that
become more of a scientific look of what's happening on

the assembly line and ensuring that everything is inspect We
don't just look at the output of you know, did
you make a good or bad product? But we'll actually
show you all of the process data that went into
making that product. The other side is on the discrete
world where you're actually assembling things. In this instance, what
we do is we'll actually monitor the assembly, so we'll

look at how people are placing door panels into a
doorframe for example, on to motif asset, or look at
tail lamps for lighting purposes right the way that they're
assembled and put together. And what we can do in
real time is tell folks, hey, what you're putting together
is misconfigured, or it's missing components, or it has too

many components. Those are defect types that are pretty common
in the assembly world.

Speaker 1 (09:20):
And what are some of the technology that is used
for that? Is it vision? Is it sensors? Is it combination?

Speaker 2 (09:26):
Everything we do is vision based, so we don't make cameras.
By the way, we are a software provider, we also
act as a system integrator, so a large part of
our business is actually delivering turnkey solutions, not just the software.
But we don't make hardware, which is actually really cool
for us because that means we get to use tons
of different types of options that are available for our

customers and it helps us really find the perfect design
and configuration that is definitely going to solve problems, and
so having the flexibility is really nice, and of course
that's a large and why we partner with Intel. We're
built on the open Veno tech stack, and that means
we can run our software really on any device that

leverages an Intel chip, which gives us tons of options
for deployments. What's really cool about this though, from a
quality perspective, is that it means you now have one
vision system that can integrate with different types of sensors.
So if you want to do, say an optical inspection
for surface defects like scratches and dents, but you also

want to look at perhaps inside that product in a
thermal application, if it's a molded part or something like that,
Well you can look at all those different types of
sensor in one easier to use screen, right, So it
removes the headache of having to have five six different
vision systems to do a variety of inspections.

Speaker 1 (10:47):
And I'm also interested in the deployment of these sorts
of new technologies. I'd like to get your thoughts and
experiences around what's some of the tips and tricks for
people out there trying to deploy not just for manufacturing quality,
but technology and AI in general into a workforce that
maybe is a little bit hesitant.

Speaker 2 (11:07):
Yeah, humans don't like change, that's for sure. I know
I don't. I'm guilty of that. And it's certainly like
that when you go into a factory and you've got
folks that have been on the same line or in
the same steel plant for twenty five thirty years, and
you show up and you've got this bright, new shiny
software and you say, hey, don't worry, data is going

to solve everything. Naturally, people can be quite apprehensive. We
don't often run into technology challenges anymore now, it's really
we run into people challenges and organizational challenges. So first
and foremost, I'll give the advice that I give on
most of the times I'm asked this question. But it's
so true. Is you don't ever start adopting technology just

for the sake of adopting it, just because competitors are
using something, or just because somebody way up the chain says, hey,
we need an AI strategy. Go invest in AI boom,
spend some time and really think about the problems that
you're trying to tackle in my world, in the quality world,
in manufacturing, it's looking at things you can do to
increase yields, increase your throughput, reduce your waste, reduce your rework,

and ultimately lower what's called the cost of quality. Start
with that, find a way that you can or process
that you can optimize by using some of this newer technology,
and then of course do a cost assessment or a
return on your investment analysis, and ensure that the business
justification is there. My experience, that's where a lot of

these projects fall short, and where folks get stuck in
these pilots and pocs is because they get really excited
to try something, but there is no proven business value
or business justification behind it, and naturally then you don't
get the executive sponsorship you need, your budget falls through,
and the project goes nowhere.

Speaker 1 (12:52):
And in your experience, what industries do you find actually
a little bit more advanced in terms of adopting these
new technology both on a technical level but also at
an organizational level that it seems like the teams are
actually involved and successfully deploying these sorts of techniques.

Speaker 2 (13:10):
Yeah, that's a great question. We see pretty advanced deployments
in the automotive world as far as discrete manufacturing goes,
they tend to be far ahead of the curve compared
to say, steel manufacturers or something like that, or concrete manufacturers.
There's a lot of very advanced technology and those automotive
facilities that make sure what you buy is actually perfect. Similarly,

in the process world, pharmaceuticals tends to be on the
continuous process side that tends to be pretty advanced. They
have a lot of vision systems in place looking at
the vaccine vials to ensure the integrity of vile caps
and seals and things like that. Some of the laggers
would be metals, some of the plastics organizations. But there's

also a kind of a bigger dynamic in manufacturing that
I think folks don't really understand that also contributes to
who's advanced and who's which is the sheer size of
these organizations, right, Manufacturers are not all large. Folks tend
to think about John Deere and three M and you know,
the largest players in the world, and the reality is

that makes up such a small fraction of the manufacturing pool,
especially in America, most manufacturing facilities have you know, twenty
people or less. Small to medium manufacturers anywhere from say
like the twenty to two hundred range of employees. That's
who makes up the vast majority of our products. Even
when you buy something really big, you know, whether it's

a whirlpool dishwasher or a hot tub or whatever it
might be, all those little components that make up that
consumer good, Well, it came from probably many different suppliers,
and most of those are small.

Speaker 1 (14:47):
It's nice that you mentioned that because my father has
a small manufacturing facility here. And just to talk a
little bit more of the technology stack that you're using
with open Vino and Intel's edge devices. I'm really interested
to see how some of the smaller guys can actually
use this sort of technology so that it can actually
be more competitive.

Speaker 2 (15:08):
Sure, well, leveraging open Veno helps us have a real
wide range of how on the hardware side, how we
can install our software. What that means for smaller manufacturers
is that we can be quite flexible in the design
of a system and can accommodate just about any budget,
which that alone is pretty significant to understand. I still

think there's a misconception that it's too expensive or too
cumbersome for the little guys, so to speak to really
innovate in their plants, and it's simply not true. You know,
we have customers that make as little as twenty parts
of shift, and even for them, having the flexibility of
how we design and configure these systems, it ensures that
even they can embrace newer technology and provide the highest

amounts of quality to their customers.

Speaker 1 (15:57):
Part of the reason I can can design and configure
of those systems is because the company uses Intel's central
processing units or CPUs, as opposed to GPUs or graphics
processing units. GPUs are specialized processes are originedly designed to
accelerate graphics rendering. The key difference in the manufacturing world

is that CPUs, like the ones Intel provides for Ogen,
are able to perform under harsher or hotter conditions like
the ones you might find in a factory or manufacturing plant. GPUs, meanwhile,
are prone to overhitting without the use of a fan
to cool it down, and most factories won't use fans
so they can avoid spreading dust and debris. There's always

a trade off between designing software optimized for CPUs or
GPUs and a manufacturing plant. I asked John about this,
and I found his answer to be quite illuminating.

Speaker 2 (16:53):
It's always an interesting discussion when people ask, why don't
you just go on GPUs and what's the real difference?
And from a manufacturing perspective, just logically thinking about what
happens in a plant. If you remember, like late nineties,
you remember you had your COMPAC or your Gateway PC,
this big old white box on the floor, and every

so often you'd take the front panel off and it
would just be totally caked in dust. Right, you'd hear
the fans humming, and well, this is what happens to
GPUs and factories. This is why we don't use fans,
because factories are dirty. There's dust everywhere. And what we
found is that when we explored using various types of

mediums to do our processing, what we found is that
fanless intel boxes were not only just as performant and
in some instances probably even more beneficial to use, but
on the maintenance side of it, we didn't have to
worry about dirt and debris, which exists in every single
plant that we deploy these in. We also didn't have

to worry about heat. GPUs generate tons of heat. Had
this discussion with somebody who did deploy GPUs in a
manufacturing environment and they were looking at in tens of
millions of dollars in HVAC improvements just to keep the
factories cool enough to operate effectively. Right. And then the flexibility,
like I mentioned, being able to very easily scale the

hardware for more advanced use cases, if we need two
or three different edge boxes, it's really easy to do,
and also be able to scale down for the smaller
applications where we want to make it a bit more
cost effective for the smaller manufacturers as well.

Speaker 1 (18:33):
Coming up next on Technically Speaking and Intel Podcast.

Speaker 2 (18:37):
Computer vision specifically for quality is becoming more and more common.
I think this will become completely commonplace over the next
twelve years.

Speaker 1 (18:45):
We'll be right back after a brief message from our partner.

Speaker 2 (18:47):
Is that Intel?

Speaker 1 (18:56):
Welcome back to Technically Speaking an Intel podcast. I'm here
now with John Weiss. I'd actually like to get you
to talk a little bit about Eigen Innovations, if you
could tell us a little bit about the company and
its mission.

Speaker 2 (19:13):
Sure so, Iigen Innovations has been around for twelve years.
We started in academia out of the University of New Brunswick.
It was founded by a PhD student and a professor.
We started as a system integrator, so we were going
into factories actually installing vision systems, and over the course
of about a decade, we developed our own software to

make our job as a system integrator easier. And about
I don't know, two and a half years ago or so,
we realized there's actually a ton of value in IP
and the software we created, and so we reinvented the
company and moved away from leading as a system integrator
to actually leading as a software SaaS based company. We
really only do one thing. We do inline quality inspection

and actually, to be more specific, our specialty thermal applications
that leverage AI. So when you think of like injection molding,
blow molding, metal welding, plastic welding, void detection, and plastic goods,
anything that has a heated process that the human eye
can't easily see defects, we do really really well there.

Speaker 1 (20:19):
And we talked a little bit about AI, and I
think we've also talked about the software that utilizes machine vision.
Where do you see AI models and the CPU based
technology being able to compete with machine vision use cases?

Speaker 2 (20:36):
Yeah, it's a good question. Look, I think there are
pros and cons of both approaches. We actually have not
yet come across a project that we had any kind
of processing limitation on being CPU based. We have applications
in production running yet thirty inferences a second across cameras. Right,

that's quite quite fast. There are definitely higher demand applications.
But in our world of process in discrete manufacturing and
the types of projects we typically focus on, speed has
actually not been a problem for us with CPUs, even
at quite aggressive speed. I see the tools getting easier
and easier to use, more and more self service, if

you will. Years ago, we had this phrase of democratizing
data if you remember that, around the days of big data,
kind of empowering everybody to be a data scientist, and
I see the same movement happening in the AI world.
In fact, actually we're a good example of that. You
can use our tool to build deploy train models across
factories and you don't have to touch a line of code.

So I think that's the future. I think the tools
get easier and easier to use, so that my good
friend Jimmy, who's down in Texas at one of our
customer plants, who's been in that same plant for over
thirty years, that he can blow me away with how
he can build a model that does thermal inspection on
metal welding. And years ago, oh, somebody that didn't have

that kind of training from a data science perspective or
a programming perspective, they would never be able to do that.
And today they're building dashboards and building models that are
literally redefining the way these manufacturers operate. It's amazing.

Speaker 1 (22:16):
You heard John say earlier that eigen has been around
for more than a decade and this technology has been
implemented across a variety of manufacturing spaces to thermally inspect
items like metal paper, cardboard, box adhesive, automotive windshields, and
high glass plastics. With such a lengthy track record of achievements,

John spoke about one specific company success story that stuck
out for him.

Speaker 2 (22:44):
A couple that come to mind. I mentioned we inference
about thirty images per second in this one process. This
is a paper process, so it's continuous, very high speed,
and it's for a high glass specialty paper. And what
happens is this high glass coding goes on paper very
rapidly as it's going down the line, and unfortunately there's

a problem where this coding can build up and if
it's not caught in about eight seconds, it will do
roughly one hundred and twenty thousand dollars worth of damage
to the equipment. This can happen multiple times as shift.
This is a very expensive problem if it's not caught.
And so this one's a great example of a thermal application.
It's a heated coating where we look at that we inference,
like I mentioned about thirty images a second, and in

just about one second, we look at all of those images,
we make a determination is there a problem or not,
is it good or is it bad? And we actually
do close loop automation as well. We'll send a signal
back there and trigger a stoppage on the line to
avoid equipment failure. All of that happens in less than
one second. So that's a really good example of speed.
Another good example, I'll give you just one more in

the interest of time, how we can help see things
that folks can't see. Well, I mentioned fuel tanks, and
I mentioned some plastic components and things like that earlier.
Naturally we use thermal vision for that humans can't see.
In thermal patterns of course, so we're able to show
quality engineers inconsistencies in the product that they would never
be able to see with the human eyes. One of

our customers manufacturers the front plates for a dishwasher company,
very large dishwasher manufacturer. And so if you've recently gotten
a new appliance, you probably remember you had to peel
all that film off, right. Well, what you might not
know is that film is on from the raw material
phase and what happens is as it goes down the process,

it gets stamped like a cookie cutter. But that film
is on it the whole time to protect it. So
what's really tough is for the quality engineers to actually
see through the blue film or whatever tint it might be,
to see if there's a scratcher dent. And so this
is one problem we solved for one of our customers
where they were missing the dents, they were missing the
scratches because the humans simply couldn't see through the protective film.

Fast forward to today again, another customer that inspects one
hundred percent of their production on our tooling and gives
them indicators in real time through that blue film if
they have any kind of service defect.

Speaker 1 (25:06):
And you've talked a little bit about the journey twelve
years ago to now. I want to get you to
cast your mind ahead twelve years in the future. Where
do you think Igen will be and in general, where
do you think manufacturing and quality control technology will be
in the next twelve years.

Speaker 2 (25:25):
That's a pretty far horizon. I don't even know if
I could guess the next twelve months, to be honest
with you, just because the industry moves so fast. But
let's say over the course of the next decade, I
would definitely see some of the more innovative technologies becoming mainstream.
So computer vision, there's no doubt about it. Computer vision
specifically for quality is becoming more and more common. I

think this will become completely commonplace over the next twelve years.

Speaker 1 (25:50):
Often ask this of our guess, but if you could
have AI solve one thing in your field that is manufacturing,
what would it be.

Speaker 2 (25:58):
I would like to use AI to clone the entire
Eigen team, because these are some of the most talented
people I've ever worked with, and I just need like
three to four times more of them so I can
go take over the world.

Speaker 1 (26:10):
Yeah. Well, we did have an episode on digital twins
and have a human digital twin, so yeah, you never know.
With that. I'll leave it there. Thank you John for
your time.

Speaker 2 (26:21):
Well, thank you, this was great. Thanks for having me.

Speaker 1 (26:25):
Thank you to John Weiss for his quality insights in
today's episode of Technically Speaking.

Speaker 2 (26:32):
In a world where we.

Speaker 1 (26:33):
Are somewhat preoccupied with virtual and digital goods, I love
hearing stories about the production of real world physical products.
I think we take for granted how much time, effort,
and brain power it takes not only to conceive of
new products, but to design the whole manufacturing process and
get them into the hands of you, the customer. John

highlighted that quality is now non negotiable for consumers and
that manufacturers need to continually reinvent the new technology and
methods to keep producing high quality products as economically as possible.
A common theme in all of our episodes, and one
that I'm always exploring, is whether these new advances in AI,
like the machine and computer vision discussed today, will help

all businesses, regardless of size. So it's pleasing to hear
John say that their technology can help the smaller niche
manufacturers to use the same quality control software and hardware
that the big players have. This is why I'm so
bullish about AI and technology in general, the ability to
lift all people and businesses up, no matter what stage

of life they are in. In our next episode, we
will look at how we can close the AI workforce
gap through education. So join us on July second for
the next edition of Technically Speaking and Intel podcast. Technically
Speaking was produced by Ruby Studio from iHeartRadio in partnership

with Intel, and hosted by me Class. Our executive producer
is Molly Sosher, our EP of Post Production is James Foster,
and our supervising producer is Nika Swinton. This episode was
edited by Sierra Spreen and written by Nick Firshall.
Advertise With Us

Popular Podcasts

Dateline NBC
Let's Be Clear with Shannen Doherty

Let's Be Clear with Shannen Doherty

Let’s Be Clear… a new podcast from Shannen Doherty. The actress will open up like never before in a live memoir. She will cover everything from her TV and film credits, to her Stage IV cancer battle, friendships, divorces and more. She will share her own personal stories, how she manages the lows all while celebrating the highs, and her hopes and dreams for the future. As Shannen says, it doesn’t matter how many times you fall, it’s about how you get back up. So, LET’S BE CLEAR… this is the truth and nothing but. Join Shannen Doherty each week. Let’s Be Clear, an iHeartRadio podcast.

The Dan Bongino Show

The Dan Bongino Show

He’s a former Secret Service Agent, former NYPD officer, and New York Times best-selling author. Join Dan Bongino each weekday as he tackles the hottest political issues, debunking both liberal and Republican establishment rhetoric.

Music, radio and podcasts, all free. Listen online or download the iHeart App.


© 2024 iHeartMedia, Inc.