March 3, 2026 • 13 mins
We explore how clean, contextualized data turns automation into true operational intelligence and why culture, not hype, defines the ROI of AI. From digital twins to predictive maintenance and OEE as a lever, we show practical steps and a case study that ends guesswork.
• shifting from reactive control to operational intelligence
• data fidelity as the foundation for digital twins
• smart components turning assets into data hubs
• predictive maintenance replacing emergency shutdowns
• OEE moving from lagging metric to predictive lever
• a rail-site case study exposing behavioral root cause
• people elevated by automation and analytics
• cultural discipline, clean architecture, leadership buy-in
Reach out to us at eecoonline.com
Share this podcast out and give us a direct review
Keep Asking Why...
Read our latest article on Industrial Manufacturing here
https://eecoonline.com/inspire/data_manufacturing
Online Account Registration:
Video Explanation of Registering for an Account
Register for an Account
Other Resources to help with your journey:
Installed Asset Analysis Support
System Planning Support
Schedule your Visit to a Lab in North or South Carolina
Schedule your Visit to a Lab in Virginia
Submit your questions and feedback to: podcast@eecoaskwhy.com
Follow EECO on LinkedIn
Host: Chris Grainger
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_00 (00:00):
Welcome to Eco Ask Why, a podcast that dives into
industrial manufacturing topicsand spotlights the heroes that
keep America running.
I'm your host, Chris Granger,and on this podcast, we do not
cover the latest features andbenefits on products that come
to market.
Instead, we focus on advice andinsight from the top minds of
(00:20):
industry because people andideas will be how America
remains number one inmanufacturing in the world.
Welcome to Eco Ask Why.
I'm your host, Chris Granger.
I look forward to spending sometime with you today.
And we're going to be havingsome fun on this episode, diving
into a topic that is it's at thedoorstep of anyone in industrial
(00:45):
manufacturing.
Uh and it's data-drivenmanufacturing.
And what does that look like?
And how can we move beyond AIhype and basic control and
really make sense of this?
And we're super excited for thishere at Electrical Equipment
Company.
Again, we're we're coming intoour 100th year of being in the
business, and it's justsomething we're celebrating big
(01:06):
time.
Looking forward to that bigcelebration.
It's gonna be happening laterthis year, but you'll be hearing
more and more about that,hopefully on EcoSY as we talk
about it.
And uh something that that thatrecently we recognize uh within
industrial transformationconversations have evolved
around two things, okay.
And this is not gonna be earthshattering, but this is
(01:27):
important automation and now AI.
But here's the reality, here'ssome things that we recognize to
be true.
Most plants out there,industrial manufacturers, don't
need more algorithms, right?
What we need is better datadiscipline, okay, because the
true frontier of modernindustrial operations isn't just
(01:47):
smarter machines, it's to shifttowards a fully data-driven
manufacturing ecosystem wherethat operational data becomes a
strategic asset and notsomething like this as an
overthought, okay?
Because AI is powerful.
I mean, we we all can admit thatAI is powerful, but without
clean, contextualized data, it'sjust noise at scale, and that's
(02:12):
not going to serve us at all, nomatter where we're at in
manufacturing.
And the real shift we need tothink about is from reactive
control to operationalintelligence.
Because traditional automationsystems were reactive by design.
Let's just think about that.
Like PLC logic executedpre-programmed rules.
When a signal changes state,this is gonna happen, right?
(02:34):
This alarm's gonna be triggered,uh, and and um alarms will go
off when limits were crossed.
And that system worked verywell, but it really didn't
learn, and that's thedifferentiator because it's not
whether or not your plant'sautomated, most likely it is.
It's whether your plantunderstands itself.
And when manufacturers begincapturing granular operational
(02:58):
data, like machine speeds andtor profiles or any fluctuations
in voltage or or operationaloperator interventions, they
move from reactive to proactive,right?
So we're not just reacting tocontrol, we're proactively being
intelligent, and that's a shiftthat's gonna change everything.
Because the old model, thinkabout that, they really rely
(03:19):
heavily on clipboards, on pens,and just tribal knowledge, all
that stuff, right?
And the new model capturestime-stamped, contextualized
data streams that allow you andyour teams to identify the
bottlenecks invisible to justthe human eye.
Okay, and and you can'toptimize.
Think about this, you reallycan't optimize what you don't
(03:41):
measure, and you definitelycan't simulate it if you can't
capture it.
So you need to be able tosimulate accurate data.
Okay, so data quality is soimportant, it's the foundational
of digital twins and predictivesystems out there.
And there's a there's a really agrowing fascination with this
idea of digital twins.
(04:01):
We spent some time on EcoAs Wildtalking about this because
you're virtually representingphysical systems, and that
allows you to simulateperformance on what that's gonna
look like before you make a realworld change.
But here's what's missed sooften when we think about this
that the digital twin is only asgood as the fidelity of the
(04:22):
input data.
That's it.
So if you don't know exactlylike when a machine is stopped,
or why it's stopped, or underwhich low conditions forced it
to stop, or what upstream ordownstream events influenced it,
then you don't have a digitaltwin, right?
You have a digital guess, andthat's not gonna serve you well.
And AI agents requirestructured, high-resolution,
(04:47):
real-time data to build usablerelationships between these
variables.
And without that foundation, youreally can't optimize, right?
And things are gonna collapseunder ambiguity.
So data-driven manufacturingstarts really at the component
level.
This we're gonna talk about justfor a little bit.
The components, smart componentsare no longer passive devices,
(05:10):
and you have devices out therelike the E300 or C445, these are
smart motor overload relays, andthat's a fundamental shift in
data architecture and how thisworks.
And these are not justproductive devices anymore,
right?
These are data hubs.
I mean, this is really unrealthe type of data you can get.
So instead of just tripping orduring a fall condition, which
they're gonna do, they'recontinuously sampling and
(05:32):
reporting back on voltageprofiles, temperature trends,
vibration, average peak current,current imbalance, you got
torque estimations, you gotenergy consumption.
The list goes on and on.
And that transformation is gonnamake it possible to go from
reactive to predictivemaintenance.
Okay, instead of waiting for acatastrophic failure, you're
you're gonna be able to receivealerts when performance is sort
(05:55):
of starting to drift fromanything that's from the
baseline.
And motors can be servicedduring scheduled downtime.
I have a big lot of a bighistory in being able to do
this.
Uh, and let me tell you what,when you service motors on your
downtime versus an emergencyshutdown, it's so much cheaper,
so much less stress foreveryone.
Damn, that downtime becomesplanned, not chaotic.
(06:19):
And that difference shows upreally big time on your PL and
your balance sheet at the end ofthe day.
So let's talk about OEE for asecond, okay?
Because OEE in the past, it waslike a lagging indicator, right?
But it could be a predictivelever, and and it's usually
thought about overall equipmenteffectiveness is like a
performance metric.
(06:40):
But when you're doing adata-driven ecosystem, this it
changes, it shifts fromhistorical scoreboard to a
predictive lever.
Because when you have hundredsof variables being analyzed and
thousands of variablessimultaneously, like conductor
speed or filler speed or torqueor operator type of
interventions, optimization ofopportunities start to come to
(07:05):
the surface, right?
And this is what so think aboutshifting the the this question.
What was our OEE yesterday?
The question could be whatvariables can we adjust
virtually to increase OEEtomorrow, right?
Without increasing any risk.
And that's the power ofsimulation when you layer it on
clean data, and it's howmanufacturers find those really
(07:28):
sweet spots without breaking anyphysical equipment.
So you really want to move fromguessing to root cause precision
and data that doesn't justoptimize output, it clarifies
accountability.
Uh we we've had instances at anelectrical equipment company
where we saw a railroadmanufacturing facility, they had
(07:49):
repeat motor failures, and thatled to assumptions of electrical
defects, and the costs were thisextravagant costs.
But we were able to put in someof these smart motor relays and
told a different story.
It showed how these littlesnapshots and load history
revealed how the how theequipment was being operated and
how it was being overfed, right?
(08:10):
And that really that uhoverfeeding of the equipment
caused the issues, okay.
The issue wasn't hardware, theissue was our behavior.
But without the data, wewouldn't have known it, right?
We just kept replacingequipment.
But with the data, we couldaddress the root cause, and
that's the difference betweenguessing and knowing.
And the human advantage is is isdata, right?
(08:34):
Uh in the data-driven culture welive in.
And the greatest, like if youstart thinking about a
misconception out there about AIand automation, the big
misconception is that it's goingto replace people.
Like, no, bro, it's not going toreplace people.
They elevate people, right?
Because when you have automatedsystems that are that are
handling repetitive monitoringand data storing, engineers and
(08:55):
technicians are freed up at thatpoint to focus on strategic
optimization or cross-functionalsystem thinking or creative
problem solving for things thatare just popping up, you know,
that that's newly risen to thesurface or continually
improving, right?
That's what it's all about.
Talented people, here it is.
Talented people don't want tofight daily fires, they want to
(09:19):
solve meaningful problems.
And data-driven environmentscreate space to do that, and
they help build, uh, bridgethese generational transitions
because you have youngertechnicians and you have uh
these guys are you know theygrew up on smartphones and
laptops and all that, right?
Uh and dashboards, and nowthey're able to come together
(09:39):
and learn so much about thesystems that they're integrating
with and that they're workingwith.
And this is not about replacingexperience.
We're about we need to amplifyit.
Amplify it.
That's really what it's allabout.
And becoming data-driven is notlike just something you just go
buy some software.
Like it's it starts withculture, it's a cultural
commitment.
(10:00):
It requires discipline, really,it's a discipline and data
collection.
Because if we're if we're gonnaget good out, we've got to put
good in.
But it also means we need tomake investment because you need
to have sensors that can canhandle and provide you the data
that you need to make the rightdecisions.
You need clean architecturebetween components and
analytics, right?
You got to have that cleanarchitecture to be able to
(10:20):
actually uh work with the data.
You gotta have a leadershipcommitment, right?
And you have to be willing toconfront uncomfortable truths
when the metrics show you that,right?
You can't just hide from it andthen run from it.
And the plants that thrive inthe next decade are not the ones
that just gonna have AI, they'llhave data integrity coupled with
(10:40):
it, they'll have systems thatlearn, and they'll have teams
that empower to act on insightinstead of instinct.
Now that instinct is bad, butinstinct, when you couple it
with insight, that's a powerfulcombination.
And the future of manufacturingisn't just about smarter
machines alone, it's about theoverall smarter ecosystem.
(11:03):
When raw operational databecomes strategic intelligence,
right?
That's a big deal because theybecause you'll, as the
manufacturer, become morepredictive, more optimized,
you'll be reducing any type ofwaste, you'll have more engaged
teams, and you'll create, andeverybody's gonna love this, a
substantial competitiveadvantage.
(11:25):
So think about that.
AI, when used correctly as atool, changes everything.
But data, clean, accurate datais the fuel.
Culture's the engine.
All right, so now manufacturerswho line and all three of these
are going to define the next eraof industrial performance.
And we're here to help you,we're here to serve it to serve
(11:47):
you to walk along with you.
If you need help with this, ifyou'd like to talk about it, if
you'd like to get one of ourexperts to come to your
manufacturing facility and talkdirectly, or maybe you want to
come to one of our labs atElectrical Equipment Company,
which we would love to have youcome do that with us, reach out.
We will have links in the shownotes for you to connect with
us.
Uh, if this conversation, thisjust insights is if you found
(12:09):
helpful and you want to sharewith others, we would be
extremely grateful for that.
Again, electrical equipmentcompany were 100 years.
We were just big, super excitedfor this night uh uh 2026, just
think about 1926.
I don't even know if this was onthe radar uh for those guys that
put that founded the companythat we would still be here 100
(12:29):
years later, but here we are,stronger than ever, out here
serving and growing, and we wantto walk with you.
So reach out to us atecoonline.com.
Uh again, share this podcast outand give us a direct review.
Uh, look forward to seeing younext time.
And remember, just keep askingwhy.
Thank you for listening to EcoAsk Why.
This show is supported ad-freeby Electrical Equipment Company.
(12:52):
Eco is redefining theexpectations of an electrical
distributor by placing peopleand ideas before product.
Please subscribe and share withyour colleagues and friends.
Also leave comments, feedback,and any new topics that you
would like to hear.
To learn more or to share yourinsights, visit ecosy.com e co.
(13:13):
A S A S W H Y dot com.
Welcome to Eco Ask Why, a podcast that dives into
industrial manufacturing topicsand spotlights the heroes that
keep America running.
I'm your host, Chris Granger,and on this podcast, we do not
cover the latest features andbenefits on products that come
to market.
Instead, we focus on advice andinsight from the top minds of
(00:20):
industry because people andideas will be how America
remains number one inmanufacturing in the world.
Welcome to Eco Ask Why.
I'm your host, Chris Granger.
I look forward to spending sometime with you today.
And we're going to be havingsome fun on this episode, diving
into a topic that is it's at thedoorstep of anyone in industrial
(00:45):
manufacturing.
Uh and it's data-drivenmanufacturing.
And what does that look like?
And how can we move beyond AIhype and basic control and
really make sense of this?
And we're super excited for thishere at Electrical Equipment
Company.
Again, we're we're coming intoour 100th year of being in the
business, and it's justsomething we're celebrating big
(01:06):
time.
Looking forward to that bigcelebration.
It's gonna be happening laterthis year, but you'll be hearing
more and more about that,hopefully on EcoSY as we talk
about it.
And uh something that that thatrecently we recognize uh within
industrial transformationconversations have evolved
around two things, okay.
And this is not gonna be earthshattering, but this is
(01:27):
important automation and now AI.
But here's the reality, here'ssome things that we recognize to
be true.
Most plants out there,industrial manufacturers, don't
need more algorithms, right?
What we need is better datadiscipline, okay, because the
true frontier of modernindustrial operations isn't just
(01:47):
smarter machines, it's to shifttowards a fully data-driven
manufacturing ecosystem wherethat operational data becomes a
strategic asset and notsomething like this as an
overthought, okay?
Because AI is powerful.
I mean, we we all can admit thatAI is powerful, but without
clean, contextualized data, it'sjust noise at scale, and that's
(02:12):
not going to serve us at all, nomatter where we're at in
manufacturing.
And the real shift we need tothink about is from reactive
control to operationalintelligence.
Because traditional automationsystems were reactive by design.
Let's just think about that.
Like PLC logic executedpre-programmed rules.
When a signal changes state,this is gonna happen, right?
(02:34):
This alarm's gonna be triggered,uh, and and um alarms will go
off when limits were crossed.
And that system worked verywell, but it really didn't
learn, and that's thedifferentiator because it's not
whether or not your plant'sautomated, most likely it is.
It's whether your plantunderstands itself.
And when manufacturers begincapturing granular operational
(02:58):
data, like machine speeds andtor profiles or any fluctuations
in voltage or or operationaloperator interventions, they
move from reactive to proactive,right?
So we're not just reacting tocontrol, we're proactively being
intelligent, and that's a shiftthat's gonna change everything.
Because the old model, thinkabout that, they really rely
(03:19):
heavily on clipboards, on pens,and just tribal knowledge, all
that stuff, right?
And the new model capturestime-stamped, contextualized
data streams that allow you andyour teams to identify the
bottlenecks invisible to justthe human eye.
Okay, and and you can'toptimize.
Think about this, you reallycan't optimize what you don't
(03:41):
measure, and you definitelycan't simulate it if you can't
capture it.
So you need to be able tosimulate accurate data.
Okay, so data quality is soimportant, it's the foundational
of digital twins and predictivesystems out there.
And there's a there's a really agrowing fascination with this
idea of digital twins.
(04:01):
We spent some time on EcoAs Wildtalking about this because
you're virtually representingphysical systems, and that
allows you to simulateperformance on what that's gonna
look like before you make a realworld change.
But here's what's missed sooften when we think about this
that the digital twin is only asgood as the fidelity of the
(04:22):
input data.
That's it.
So if you don't know exactlylike when a machine is stopped,
or why it's stopped, or underwhich low conditions forced it
to stop, or what upstream ordownstream events influenced it,
then you don't have a digitaltwin, right?
You have a digital guess, andthat's not gonna serve you well.
And AI agents requirestructured, high-resolution,
(04:47):
real-time data to build usablerelationships between these
variables.
And without that foundation, youreally can't optimize, right?
And things are gonna collapseunder ambiguity.
So data-driven manufacturingstarts really at the component
level.
This we're gonna talk about justfor a little bit.
The components, smart componentsare no longer passive devices,
(05:10):
and you have devices out therelike the E300 or C445, these are
smart motor overload relays, andthat's a fundamental shift in
data architecture and how thisworks.
And these are not justproductive devices anymore,
right?
These are data hubs.
I mean, this is really unrealthe type of data you can get.
So instead of just tripping orduring a fall condition, which
they're gonna do, they'recontinuously sampling and
(05:32):
reporting back on voltageprofiles, temperature trends,
vibration, average peak current,current imbalance, you got
torque estimations, you gotenergy consumption.
The list goes on and on.
And that transformation is gonnamake it possible to go from
reactive to predictivemaintenance.
Okay, instead of waiting for acatastrophic failure, you're
you're gonna be able to receivealerts when performance is sort
(05:55):
of starting to drift fromanything that's from the
baseline.
And motors can be servicedduring scheduled downtime.
I have a big lot of a bighistory in being able to do
this.
Uh, and let me tell you what,when you service motors on your
downtime versus an emergencyshutdown, it's so much cheaper,
so much less stress foreveryone.
Damn, that downtime becomesplanned, not chaotic.
(06:19):
And that difference shows upreally big time on your PL and
your balance sheet at the end ofthe day.
So let's talk about OEE for asecond, okay?
Because OEE in the past, it waslike a lagging indicator, right?
But it could be a predictivelever, and and it's usually
thought about overall equipmenteffectiveness is like a
performance metric.
(06:40):
But when you're doing adata-driven ecosystem, this it
changes, it shifts fromhistorical scoreboard to a
predictive lever.
Because when you have hundredsof variables being analyzed and
thousands of variablessimultaneously, like conductor
speed or filler speed or torqueor operator type of
interventions, optimization ofopportunities start to come to
(07:05):
the surface, right?
And this is what so think aboutshifting the the this question.
What was our OEE yesterday?
The question could be whatvariables can we adjust
virtually to increase OEEtomorrow, right?
Without increasing any risk.
And that's the power ofsimulation when you layer it on
clean data, and it's howmanufacturers find those really
(07:28):
sweet spots without breaking anyphysical equipment.
So you really want to move fromguessing to root cause precision
and data that doesn't justoptimize output, it clarifies
accountability.
Uh we we've had instances at anelectrical equipment company
where we saw a railroadmanufacturing facility, they had
(07:49):
repeat motor failures, and thatled to assumptions of electrical
defects, and the costs were thisextravagant costs.
But we were able to put in someof these smart motor relays and
told a different story.
It showed how these littlesnapshots and load history
revealed how the how theequipment was being operated and
how it was being overfed, right?
(08:10):
And that really that uhoverfeeding of the equipment
caused the issues, okay.
The issue wasn't hardware, theissue was our behavior.
But without the data, wewouldn't have known it, right?
We just kept replacingequipment.
But with the data, we couldaddress the root cause, and
that's the difference betweenguessing and knowing.
And the human advantage is is isdata, right?
(08:34):
Uh in the data-driven culture welive in.
And the greatest, like if youstart thinking about a
misconception out there about AIand automation, the big
misconception is that it's goingto replace people.
Like, no, bro, it's not going toreplace people.
They elevate people, right?
Because when you have automatedsystems that are that are
handling repetitive monitoringand data storing, engineers and
(08:55):
technicians are freed up at thatpoint to focus on strategic
optimization or cross-functionalsystem thinking or creative
problem solving for things thatare just popping up, you know,
that that's newly risen to thesurface or continually
improving, right?
That's what it's all about.
Talented people, here it is.
Talented people don't want tofight daily fires, they want to
(09:19):
solve meaningful problems.
And data-driven environmentscreate space to do that, and
they help build, uh, bridgethese generational transitions
because you have youngertechnicians and you have uh
these guys are you know theygrew up on smartphones and
laptops and all that, right?
Uh and dashboards, and nowthey're able to come together
(09:39):
and learn so much about thesystems that they're integrating
with and that they're workingwith.
And this is not about replacingexperience.
We're about we need to amplifyit.
Amplify it.
That's really what it's allabout.
And becoming data-driven is notlike just something you just go
buy some software.
Like it's it starts withculture, it's a cultural
commitment.
(10:00):
It requires discipline, really,it's a discipline and data
collection.
Because if we're if we're gonnaget good out, we've got to put
good in.
But it also means we need tomake investment because you need
to have sensors that can canhandle and provide you the data
that you need to make the rightdecisions.
You need clean architecturebetween components and
analytics, right?
You got to have that cleanarchitecture to be able to
(10:20):
actually uh work with the data.
You gotta have a leadershipcommitment, right?
And you have to be willing toconfront uncomfortable truths
when the metrics show you that,right?
You can't just hide from it andthen run from it.
And the plants that thrive inthe next decade are not the ones
that just gonna have AI, they'llhave data integrity coupled with
(10:40):
it, they'll have systems thatlearn, and they'll have teams
that empower to act on insightinstead of instinct.
Now that instinct is bad, butinstinct, when you couple it
with insight, that's a powerfulcombination.
And the future of manufacturingisn't just about smarter
machines alone, it's about theoverall smarter ecosystem.
(11:03):
When raw operational databecomes strategic intelligence,
right?
That's a big deal because theybecause you'll, as the
manufacturer, become morepredictive, more optimized,
you'll be reducing any type ofwaste, you'll have more engaged
teams, and you'll create, andeverybody's gonna love this, a
substantial competitiveadvantage.
(11:25):
So think about that.
AI, when used correctly as atool, changes everything.
But data, clean, accurate datais the fuel.
Culture's the engine.
All right, so now manufacturerswho line and all three of these
are going to define the next eraof industrial performance.
And we're here to help you,we're here to serve it to serve
(11:47):
you to walk along with you.
If you need help with this, ifyou'd like to talk about it, if
you'd like to get one of ourexperts to come to your
manufacturing facility and talkdirectly, or maybe you want to
come to one of our labs atElectrical Equipment Company,
which we would love to have youcome do that with us, reach out.
We will have links in the shownotes for you to connect with
us.
Uh, if this conversation, thisjust insights is if you found
(12:09):
helpful and you want to sharewith others, we would be
extremely grateful for that.
Again, electrical equipmentcompany were 100 years.
We were just big, super excitedfor this night uh uh 2026, just
think about 1926.
I don't even know if this was onthe radar uh for those guys that
put that founded the companythat we would still be here 100
(12:29):
years later, but here we are,stronger than ever, out here
serving and growing, and we wantto walk with you.
So reach out to us atecoonline.com.
Uh again, share this podcast outand give us a direct review.
Uh, look forward to seeing younext time.
And remember, just keep askingwhy.
Thank you for listening to EcoAsk Why.
This show is supported ad-freeby Electrical Equipment Company.
(12:52):
Eco is redefining theexpectations of an electrical
distributor by placing peopleand ideas before product.
Please subscribe and share withyour colleagues and friends.
Also leave comments, feedback,and any new topics that you
would like to hear.
To learn more or to share yourinsights, visit ecosy.com e co.
(13:13):
A S A S W H Y dot com.
Popular Podcasts
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
The Breakfast Club
The World's Most Dangerous Morning Show, The Breakfast Club, With DJ Envy, Jess Hilarious, And Charlamagne Tha God!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.