How can commercial fleets benefit from AI? | with Lawrence Bader (Retired UPS)

Recorded live at ACT Expo 2026, host John Heinlein interviews former UPS fleet manager Lawrence Bader on the evolution of fleet technology, including keyless systems, AI-driven predictive maintenance, and autonomous trucking. Bader emphasizes that to remain competitive, fleets must start small, iterate quickly, and integrate technologies across all operational departments.

Creators and Guests

Host

John Heinlein, Ph.D.

An experienced technology and marketing leader, John brings his background from startups and established companies to Sonatus. He worked for 14 years at Arm, most recently leading Automotive Partnerships for North America where he engaged OEMs, Tier-1s, and others to deploy Arm-based solutions into automotive applications, including autonomous vehicles. His team was integral to launching the SOAFEE industry initiative for software-defined vehicles of which Sonatus is a member. Earlier he served as VP and Chief of Staff to the CEO and led a group responsible for competitive strategy. For three years, he was VP of Corporate Marketing where he led centralized outbound marketing, spanning marketing campaigns, press, events, web, digital marketing, ecosystem programs, and working closely with investor relations. Prior to Arm, John had an 11-year tenure at microprocessor startup Transmeta where he held several senior roles spanning business development, marketing, and customer success, among others. John earned his B.S. in Computer Engineering from Carnegie-Mellon University and an M.S. and Ph.D., both in Electrical Engineering, from Stanford University.

What is The Garage by Sonatus?

The Garage Podcast from Sonatus brings you conversations with thought leaders from around the vehicle technology space discussing far-reaching topics about software innovation in vehicles.

These episodes will include industry experts, whether from Sonatus, or from our amazing partners across the industry, all of whom will share their Ideas and their outlook for the most important topics in vehicle technology. Episodes span from more technical topics, to business evolution, and market trends. Join us and learn about the future of vehicle in The Garage.

Today in The Garage,

we're recording live at
ACT Expo 2026 in Las Vegas,

Nevada.

Around the show floor,

there's an incredible diversity
of commercial vehicles of all

types, from heavy
tractors, trailers, buses,

lots of electric
charging equipments,

specialty purpose
vehicles of all kinds .

And one of the things that we
recognize is the importance of

how technology can benefit
fleets and benefit their

different missions.

Whether it's for
preventative maintenance,

ensuring uptime and
reliability and safety,

all the way through to
autonomous trucking and

optimized operations.

Technology has a role to play
in commercial vehicles and

we're beginning to see that.

So many of the exhibits here
are looking at how fleets are

using technology in new ways to
solve a diverse range of problems.

To dig into that topic, we have
a special guest with us today.

Our guest is Lawrence Bader,

who spent thirty two years at
UPS running and managing their

fleet operations and now is
an independent consultant.

And Lawrence is going to share
with us in today's interview a

wide range of problems
that he encountered

and fleets have to solve
and how he's seeing

technology solving a range of
those issues now and in the future.

It's a very interesting
conversation on commercial vehicles.

Let's go.

Welcome to The Garage.

I'm John Heinlein, Chief
Marketing Officer with Sonatus.

We're here recording live at
ACT Expo in Las Vegas 2026.

I'm so well so excited to welcome
Lawrence Bader to the podcast.

Lawrence, welcome to The
Garage. Thank you, John.

Appreciate it.

We always have our guests begin
by telling us about themselves

and their background.

Please introduce yourself.
Tell us about you.

Sure. Well, my name
is Lawrence Bader.

I recently retired from
UPS after thirty two years.

Thirty two years.

That's great.

Thirty two years is a long time,

where I had the opportunity to run
what we call global fleet systems.

And I spent my entire career in
really the IT discipline, you will.

But, you know, working for UPS
gave me lots of opportunities to,

to work in the airline space,

to work in the the
automated hubs.

And in the last
decade of my career,

I actually focused on what we
call ground transportation.

And that was basically
the fleet of about 275,000

pieces of rolling equipment.

That's a lot of that's
a lot of equipment.

That kind of insight is
exactly what we're looking for.

We're looking for to learn
about you and what you what you

learned from those that time.

But begin first by telling us a
fun fact about you as we always

do on the podcast.

Well, you know, actually, John,

I really enjoy building
and racing cars.

I didn't start that really
until later in life.

As a matter of fact,

I couldn't even change a set of brakes
until I was forty five years old.

But, you know, I was an an IT background
for such a long time that when you

build software, you don't
get your really hands on it.

You don't feel it and get you
really overly overly experienced it.

So, you know, I learned
to to turn a wrench

and, you know, be able to
put that power to the ground,

so to speak, and really just
enjoy racing racing cars.

That's that's fantastic.
I love that discipline.

I must say it's not
an expertise I have,

but I like building
physical things.

I do a little carpentry. I
like to build stuff at home.

So Yeah.

I think that's in touch
with you what you build.

Right? Yeah. You see
could see it tangible.

I think it's great.

So now tell us about your what
you do now and your company.

Yeah.

So, you know, after thirty two
years of being in the industry,

I didn't really want to
leave the industry just yet.

So I made a lot of friends,

and and you have to get a
lot of opportunities here.

So really, I'm just an independent
consultant where I like to help

fleets adopt technologies.

I had the luxury of really
exercising a lot of products

while I was at UPS.

And if I can bring that
to other fleets who don't

necessarily have the background
or the time to kind of press

into those things, you know,

it's just a good
thing for me to do.

But I also like
to help, you know,

vendors and solution providers
who are developing products

and, you know, help
them to, you know,

to determine really
what's gonna stick,

what's gonna sell to fleets.

But really, all roads
for me really lean

towards autonomous vehicles.

You know, I'm trying to help accelerate
the adoption of autonomous operations.

Fantastic.

If someone wants to find you,

what's the name of your company?

Advanced Transportation
Technology. Oh, great.

And you could just
find me on LinkedIn.

Okay.

Well and we we're happy to put
a link in the show notes as

well for people to be able to

find you.

So based on your
long career at UPS,

I'm wondering what are the
lessons you learned there that

can apply to fleets today
that people can learn from?

You know, I think the biggest
overarching learning that I've had

throughout my entire career is
you you really just gotta start.

You know?

You really have got to to
to really move the needle by

getting going.

Oftentimes, we
over-analyze solutions.

We often over-analyze
where we wanna be,

where what you wanna
get accomplished.

But start.

Start small, gain
some confidence,

and then you can build on top of
those, of of those wins, if you will.

So it's like iterative
development is something in the

IT industry we did a lot of.

And and if you don't start,
you're never gonna get there.

Can you share an example of
how you see people using technology

in a fresh and interesting way?

Yes.

As as a matter of fact,

something that I recently had
the opportunity to to help

deploy before I
before I left UPS was,

something we called
keyless tractor.

We actually physically removed the
keys from the operation of a tractor.

No keys?

No keys. Yes.

We actually, you know,
did an encrypted digital

key on the driver's device,

which allowed us to to
issue electronic keys at the right

time to the right driver
for the right vehicle.

So the the concept of having to
manage keys or having the key

board up there or keys in in
coffee cans, we eliminated.

And that helped us
in in numerous ways.

Where one of them was,
just driver time alone.

The driver would, when they
cut to came to the facility,

they would go to
the guard shack,

and then they had to walk
into the dispatch office.

They would get
assigned their tractor.

And sometimes the driver
did not necessarily,

want that particular tractor
that was, scheduled for him.

So he would maybe sweet talk to
the person behind the counter

to, give him a different
tractor for the day. It

just threw everything out
of sorts, right?

Here, now the driver can not have
to walk through the dispatch office.

It's automatically
on their device.

They walk through the unit.

They can immediately open the
doors when they get into range.

It starts to pre-trip
cycle, and off they go.

And like maintenance
personnel, oftentimes,

they couldn't find the keys
when they had to move the units

to bring them into the shop.

But now it's actually on
the device for the mechanic,

and they can unlock the vehicle,

move it into the place,

and and begin their work without
having to go to dispatch office.

I'm sure that allows you to
provision keys and also revoke

rights later when
they're not needed.

Yep.

And that mean, that's actually an
interesting technology that could play in a

lot of different spaces.

Like, it's with the rental
market or the leasing market.

But I think any fleet who
has a key management problem,

that's just a just an interesting
solution that's out there.

Right? Just just one of many.

Yeah. Just just using
technology in a fresh way.

In a fresh way.

We're finding and I think
when we talk to customers,

much times the perfect
is the enemy of the good.

Absolutely.

And people are trying to
arrange the perfect solution

and what they end
up with is nothing.

You're never gonna get there.

It's so much better to start
with a solution that's that

works, that's solid,
and add to it.

One of the things we're our
solutions provide is the

ability to to
change and iterate.

You might begin using
it for one application,

and you realize, well, I
have this other problem,

but you can grow into
that and you can expand.

Right.

And I think we're all
discovering that today with def

definitely with AI.

You know, just collecting the data
has been a tremendous amount of

time we've all spent in in in in
getting the data in our hands.

But we didn't know what
to do with it. Right?

We didn't we couldn't
put action to it.

So so now we can iterate on
top of all the data we gathered

over all those years.

It's a perfect transition
because AI is such an important

topic for us, an important topic
we cover on the podcast now.

So what are you seeing in AI
as moving the needle today,

and where do you see
the opportunities,

and where would you suggest people
be investing in AI in the future?

Because I think people think
about AI and they immediately

think about autonomous,
which is important.

I think they think about
voice assistants and so on.

But for me, that's just the beginning of the
scratching the surface of what's possible.

What do you see
as opportunities?

Well, you know, to me,

leveraging AI and leveraging
all the data that you've

collected is is really how
can we begin to automate those

mundane tasks?

You know, let's remove the human
element out of the mundane tasks that

are just simply repetitive.

And then that's just where
we need to start. Right?

Well, you know, we we started back
in our early days with with capturing

telematics data,
capturing GPS location.

But, you know, that's
just the start of it.

And and how can we take that data
and get to true predictive analytics?

How can we determine component
failures before they occur?

How can we change our
maintenance programs from away

from, you know, time and miles
based to really condition

based, based upon, you know,

what's going on with the
actual piece of equipment?

I think that's such an
important point you mentioned

because when we think
about maintenance

today and we think
about diagnostics,

it's relatively blunt
instrument people are using.

And it means that you're
probably doing excess

maintenance you don't need.

Absolutely. And you're
doing it sometimes too late.

If we can use a better
intelligence to both understand

how the vehicle is actually
being used instead of just as

you say, the mileage schedule,

and then look more specifically
at at what your indications of

problems are, I think
that's incredibly valuable.

Yeah.

I think oftentimes, you
know, the respectable fleets,

those who need to deliver
on time will will tend to

over-maintain their equipment.

And over to maintaining
equipment's expensive. Right?

I mean, you know,
we would you know,

as I've seen fleets out there
who would actually replace

tires before they're
they need to be replaced.

They replace starters before
they need to be replaced based

upon just time and miles
because they had to protect

their reputation.

Right.

And, you know, if we
can move towards this,

actually look at the
signals that are coming off,

let's just look at the
electrical system as an example.

You know?

You could absolutely change
three or four major parts when

you're trying to trying
to prevent a failure,

whether it's the battery or the
cabling or the starter itself.

Let's actually go in and dig
into the the signature patterns

of the electrical
circuit to see, you know,

we can have AI
actually predict, well,

when is that starter
actually going to fail?

You know, instead of
replacing it every year,

let's replace it on
an as needed basis.

Right.

We're we're showing a
demonstration just a few feet

away from us in in
the booth showing something similar to

that, how you can use, you know,

RP129 protocols to better
understand the actual

situation of the vehicle,

help understand is it the starter
or is it some is it a wiring issue?

Is it some other issue?

So instead of just brute force
replacing things that aren't

broken, which gives you higher

maintenance costs,
higher downtime,

You replace what's broken
when it needs to be fixed

Now you mentioned tires as well.

I think it's an interesting
application because, you know,

tires are so crucial where the
rubber meets the road literally.

Right? Yeah.

But we have a collaboration
with Michelin that we've shown

recently using AI where they're
doing exactly what you said.

They're looking at specifically
how the the vehicle is being

used, how is the load
varying over time,

what's the the weight,

all those different things
factoring in and and giving a

better indication of how tires
are actually wearing versus

versus how they might be
theoretically wearing.

And that's allowing the
extended use of of tires for a

longer time and again, less
downtime and less unnecessary

maintenance.

It's using AI.

And we're we feel that by
putting some of the AI like

that closer to the vehicle,

you have the opportunity to
do the kinds of calculations that

you can't do in the cloud.

It's too slow.
It's too far away.

It requires too much data.

So that's kind of the one of
the visions we're looking for.

Do you do you see
that as viable?

Well, absolutely.

I mean, what you can compute on
the edge and and being able to get

the processing done done
away from the cloud,

can you makes you, you
know, far much more nimble.

And also, you know, I've I've
come across the the ideas,

and maybe you guys
have talked about this,

with virtual sensors.

You're actually creating
sensors from two or three other

inputs and being able to gain
deeper insight into that.

I've actually seen that in a
couple different locations.

It's true.

And one of the we're showing
some examples of of how that

works in the past.

We've got some collaborations
with a vendor called

COMPREDICT, for example,

and others who are using virtual
sensors to calculate things.

Actually, there's plenty of information
to calculate a sensor result

instead of putting
in a physical sensor.

And by doing that,

they can avoid cost and
simplify the vehicle.

But what's really exciting,

and you and I were
chatting about this,

is imagine you realize
that I wish I had a sensor

after the vehicle was shipped
and I have all the data,

but darn I didn't
put the sensor in.

You could put a virtual
sensor in after production.

That's that's pretty cool.

And so it and one of the things
we we chatted about is imagine

a virtual hazard light.

You realize that, well,
when these combinations of

situations happen,

when I see this indication
on this subsystem and that

indication on that subsystem,

it's an indication that there's
going to be a problem coming soon.

You could then raise a virtual
hazard light that says, hey,

look, you need to take this
in for service, say, urgently,

but that could be deployed as a kind of
a virtual hazard light after production

You know, I actually see that as
we look forward down the road to

autonomous trucking.

When we have autonomous
trucks running up and down our

highways, which we will
indeed have, you know,

there's not a
driver in the seat.

Right?

And being able to detect
an issue before it occurs,

being able to articulate
how far can I go with that

autonomous vehicle?

Do I have to get off
immediately at this next exit?

Can I complete my mission?

Or do I need to kind of plan
my route to get to the next,

you know, certified,

repair facility that may be one
hundred and twenty five miles away?

I think that that's
an interesting play.

Yeah, so much of driver
expertise, you know,

we we respect drivers and their their
their knowledge and their know how.

You know, I think myself and and you
and certainly professional drivers

can see and feel that
something's not quite right.

Right. Absolutely.

And they want to adjust it.

But as you go towards an
autonomous world where the

driver's not driving or maybe
not driving all the time,

you might have a situation
where you really need to be

monitoring that in a new way.

Yes. Yep.

And you have to mount it to both
the tractor and the trailer.

It's true.

You know, the trailer is
also a very important aspect,

which I'm now seeing more
and more fleets adopt sensor

technologies on those trailers.

Yes. Smart trailers
is a big topic.

And there's nothing
exclusive about

any of these things
we're covering.

I think they both apply to to
the tractors and trailers both.

Yep.

As I mentioned
earlier, you know,

encouraging fleets
to just get started,

and there's a great
example there.

We're just looking at TPMS or automatic
tire inflation system on trailers.

You know, they're not
everywhere right now, but,

you know, the tires is one
of the most important things.

And actually, I came across
a fleet here recently who,

who adopted tire pressure monitoring
systems on their on their equipment,

and they are now as they
as you look at the roadside inspection

data from, the states, they
have zero tire violations.

And that's phenomenal
in our industry.

And that's just an example of
where fleets actually picked up

the ball, started simple,
something that they knew knew

about, and applied it.

And now they're being able
to reap benefits from that

financially, but also
from a brand image.

So much of what we're
talking about relies on data.

Think fleets have been
collecting data for years,

but a lot of that data I think
has probably gone into a pile

that's never looked at.

You know, what do you see as
the opportunity from using data

in smart ways to do some of
these things we're talking about?

Yeah. I mean, that that
is an industry challenge.

As I talk to a lot of fleets,

there a lot of data
is being collected,

but it is collected in silos,

and it's too much data for
any fleet to really put to action.

So that's really the key piece.

Now that we have AI,

it's gonna actually come in
and create these, you know,

these actionable insights.

And we have to bring those
actionable insights not to some

email chain that goes
out to everybody,

not into a vendor dashboard
where it's buried only in in in

certain vendors' platforms.

It needs to meet the operation
where they are. Right.

Let's roll that into
their fleet management system or

their work dispatch for the day
or for the drivers, you know,

the drivers panel
inside the vehicle.

Right.

You gotta meet the
operation where they are,

not layer on another piece of
technology, another screen,

another portal that
they have to log into.

That simply does not work.

Right.

So it's such a smart point
because some of the data needs

to go to the central fleet
manager to understand perhaps

collective
maintenance...scheduling it.

Some of it wants to
go to the driver,

but the answer is it's both.

It's not just, okay.

If I just have an
app for the driver,

I've solved all the problems.

Or if I just throw it
into some, console a mile away.

It needs to be a mixture
of all of those things.

Well, and that's
a key piece here,

which we really
haven't pressed into.

But as I talked to
a lot of fleets,

their their challenge with
adopting technology or AI in

general is they're very siloed.

You know, it has to be in
across the entire organization.

The chief operating officer
is the one who really needs to

kinda get in and make sure
that all of their business units are

playing together,

collaborating on bringing
this technology into play.

It's gotta be the
maintenance department,

the dispatch department,
health and safety, and, yes,

the IT organization because
the IT organization one that

stitches it all together
and makes it available.

Obviously, commercial fleets
and commercial vehicles are

different than, than passenger vehicles
because downtime is is critical.

Downtime is just money
lost, directly money lost.

So how do you see
maintenance changing that?

Do see the ability for us
to do smarter maintenance,

smarter predictive maintenance
materially improving downtime?

Oh, absolutely. Yes.

I mean, you know, today,

we have to wait until
it's an after the fact.

The failures already occurred.

It already throws the diagnostic
trouble code, the DTC.

We need to be in front of that,

and we can actually pick up
data from various sensors in a

normal course of the operation,

have a digital twin of
that particular platform,

and know when it has
a deviation from that.

So I can predict before the
actual DTC code is ever thrown.

That's where the magic happens.

That's where I can
actually get in,

pull the equipment off the road
before it's a breakdown on road.

I can't stand
breakdown on roads.

Right.

And now now that's looking
at an individual vehicle.

I think the the opportunity,
and you mentioned AI as well,

is if we're able to
look at across a fleet,

now we're seeing patterns
across a fleet that says, okay,

I have this problem
on this vehicle,

but now I realize that
it's affecting other vehicles or

other vehicles are going
to have that problem soon.

Well, if they're using the
same components. Right?

Or the same the the, you know,

the the same tier one supplier
has an issue across the

different fleets.

But, you know, even pressing
into a specific make and model,

we have issues with a
specific make and model.

And being able to take those
earlier on before I continue to

increase the number of of
units that I bring on board,

that that's a gold mine.

So here at ACT, the the show is full
of diverse vehicles of all kinds.

There's heavy construction.
There's buses.

There's there's trucks. There's
trailers.. tractors, trailers.

What

are you some of the things
that you see as interesting,

here at the show,

and what are you what are
you focusing on looking at today?

Well, for me, it is is getting
back to automation. Alright?

So a lot of companies are
leaning into automation,

putting that data to use,

and then I'm taking automation
a little bit further with

autonomous vehicles,

whether it's over the road
or on property shifting,

especially in our
class eight space.

And so here at ACT,

you do see a lot of the
autonomous vehicle vendors here

who are partnering up with
some OEM suppliers here, and,

I'm looking forward to
seeing those on the road,

and growing in numbers.

So when you think
about autonomous,

do you look at that as end to,
you know, last mile autonomous?

Do you see it as
primarily highway?

How do you see the
industry adopting this?

Well, I think the industry
is gonna adopt, first of all,

at the distribution
centers or the hubs.

So on-property shifting is
what I'm referring to there.

So moving the trailers around
from a staging area to the dock

door, pulling off the dock
door, you know, outbound,

and so on and so forth,
and then over the road.

So basically, you
know, your long hauls,

those that are going
multiple states, you know,

exceeding the the ELD hours
of service rules, you know,

going coast to coast.

Right?

A lot of times, we'll
put trailers on trains,

and we'll train
them over the rail,

which is not always
reliable and timely.

But if we can pull those
off the road and put them in

autonomous trucking and run
for thirty two hours without

refueling, that'll
be a great day.

I see.

So it's really kind of a
mixture of from the the

distribution center,

which maybe people don't
think of as an important area.

There's lots of
personnel-intensive work to

move move things around there.

Oh, on the property? Yes.

Yeah.

And that because it's
an enclosed space,

it gives you opportunities to
do things differently than you

could do on roads.

Yeah. It's much more controlled.

I don't have to worry about
public perception. Right?

I can really, you know,

manage how the yard
behaves and operates.

And to be quite honest with you,

that's a that's a tough job.

You know, if you've never ridden
a terminal tractor or a shifter

and you're slamming into the
back of that trailer loading

that thing up, it's it's
physically demanding.

To do it all day long,
moving one after the other.

It's not it's not exactly
a sought-after job. Right?

People rather be on
the road. Yes. Local.

And when you talk about
over the road, you know,

a lot of our drivers,

they don't wanna be gone for four
or five days away from their family.

So that's just another great
application that can really

help the the livelihood
of our drivers today.

So really looking for solving
the right problems using

different technologies because
those may be different.

Distribution center can avail
itself of high resolution

knowledge of the center,

perhaps additional beacons and so
on versus over the road has

different issues,

needs to have a broader highway
map system and and so on like that.

Well, AI is certainly
accelerating that,

it's been basically
physics-based AI.

Physics AI is what's
really, really,

doubling down and accelerating
the growth of autonomous and

automated operations.

It's an exciting future to
think about all of those things

working together.

I think the the fleet management
job is a is a hard job.

You've seen that firsthand.

Absolutely.

But I think if you
if you do it right,

it can create incredible
results from the industry.

And, bottom line, you need
to lean into technology.

It's there. It works.

And if your company doesn't
lean into it, you know,

you're gonna be missing out a
lot of opportunity and there's

gonna be other companies who
are gonna surface and could

potentially just, you
know, simply run past you.

Well, Lawrence, we've had a wide
ranging discussion and your experience

in the industry is so valuable.

I appreciate you coming by
and spending time with us.

My pleasure. Surely
enjoyed it, John.

If you like what you're
seeing in the this episode,

please like and subscribe
to see more like it,

both here from ACT Expo and our
other shows all around the world.

Thanks for watching.
We'll see you again soon.