May 6, 2026 • 37 mins
Stuck in traffic? There's someone to blame, but it's not who you think it is. Jorge cruises to an answer with two traffic science experts.
See omnystudio.com/listener for privacy information.
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:01):
Hey, Welcome to Sign Stuff, the production of iHeartRadio More. Hey,
cham and today we're answering the question what really makes
traffic so bad? It turns out it's not careless drivers
or accidents or roadwork. It's something much more surprising. We're
going to talk to two experts on traffic engineering. We're
(00:22):
going to step us through the history of traffic, the
real causes of it, and what we can do to
make commuting easier. To hit the accelerator because we are
clearing the road to answer the question what really makes
traffic that enjoy? Hey everyone? So, like most people, I
(00:48):
hate traffic. I always feel like I'm wasting my life
away sitting in my car, hitching my way across the
highway or a busy road. And the worst part is
that sometimes there's no clear reason for it. You've probably
had the experience of sitting in traffic for a while
and then all of a sudden, the traffic clears up,
and you think was it all for nothing? What's going on? Well,
(01:12):
to find out why that happens and to answer all
my traffic related questions, I reached out to two traffic
engineering experts. The first one is doctor Maria Lauda de
la Monice, a professor of civil and environmental engineering at
the University of California at Berkeley who works at the
no pun intended intersection between traffic science and math. She's
(01:35):
going to tell us why these random traffic jams happen,
but first I wanted to know a little bit about
the history of traffic. So here's my conversation with doctor
Maria Lauda de la Monique. Well, thank you, doctor de
le Monique for joining us.
Speaker 2 (01:51):
Thank you, it's a pleasure to a here.
Speaker 1 (01:52):
Well, I kind of have to apologize for being laid.
I was stuck in traffic.
Speaker 3 (01:57):
That seems to be a problem for everybody these days,
especially in the Bay Area where we are.
Speaker 1 (02:03):
Actually, I'm just kidding. I work from home, so I
haven't had to deal with traffic in a long time.
Speaker 3 (02:09):
I bike to avoid the traffic. I have an electric bike.
That's my commute.
Speaker 1 (02:13):
Oh that's smart. Wow, at least for now, until everybody
gets an electric bike, Yes, then there might be bike traffic.
Speaker 3 (02:19):
Yeah, there's still is some places in Europe. Alreadia experience
in traffic with bikes in the Netherlands. In Paris, it
was a problem for a bit.
Speaker 1 (02:27):
Oh wow, all right, we'll get into that, but I
thought way we could start by asking you to tell
us a little bit of the history of traffic, Like
when did traffic become a problem for humans.
Speaker 3 (02:36):
Well, people tend to think that traffic was born, let's say,
with cars, but in reality it's something that all does
humans and cities. So we can go as far back
as ancient Rome and we still have traffic and congestion.
There's Julius sister, even have the regulatory intervention to avoid
that commercial trucks or commercial cars will go on through
(02:58):
the city during daytai because they would cause congestion and
they would make too much noise. So it's a very
whole problem and we're still here trying to solve it.
Speaker 1 (03:07):
It's oldest Roman empire, yes, yes, And then a city
started to grow, we also started to experience more and
more traffic and different type of traffic.
Speaker 3 (03:20):
Obviously, with the advent of cars is when the problem
became very big because a lot of cars that are
in the road, and also became apparent that the road
cannot be shared anymore between pedestrians and cars, right where
before it was the case everybody was in their pedestrian, bicycle,
animal power vehicles, and then since the cars were there,
(03:41):
and then it became a very chronic problem. And now
it became a problem of everybody going to work in
the morning at the same time and being stuck in traffic.
Speaker 2 (03:50):
When we built freeways.
Speaker 3 (03:51):
In the fifties, now people didn't live anymore in the city,
but now they had to take the freeway to go
to the world, and now freeway became congested. And so
it's a problem that we've been dealing for a long time.
Speaker 1 (04:02):
Who it seems like any time you try to do
something about it, it still manages to come back.
Speaker 3 (04:07):
Yes, because it seems the more we make it better,
the more we make people want to use the cars,
and so instead like we build more roads and then
people use more cars instead of like fixing what we
had before.
Speaker 1 (04:20):
And so that seems like a very human problem, is
it is, Well, can you take me back a little
bit on the history of I guess, trying to manage
traffic or trying to deal with traffic, Like what did
the Romans do?
Speaker 3 (04:33):
So before the cars, what people were trying to do,
it's mostly intervened like on the human behavior. So it's
like forbid the use of cars during daytime, right, and
so you can only go in the city during nighttime
when there was animal power vehicles. So there may be
a road becomes a one way road, or you had
a policeman at an intersection that decides, like you know,
(04:57):
now it's animal power flow that goes through the intersection,
then it's pedestrian.
Speaker 2 (05:01):
Then as whatever else.
Speaker 3 (05:03):
As technology has gotten better and better, we're also getting
better and better and managing traffic. So as car came on,
also traffic light came on because now we were suddenly
able to measure traffic. And the way to measure traffic
at the beginning of the nineteen hundred was a man
standing on the side of the road and literally counting
car as time.
Speaker 1 (05:24):
Passed, right, And so graduate student, I imagine.
Speaker 3 (05:27):
At the time it was actually scientists, like there's this
famous picture of Bruce Greenshield as one of the fathers
of traffic flow theory, that's standing on the side of
the road with a camera and counting cars in the
nineteen thirties or something. And so the idea it became
managed traffic. The same way as we can measure traffic,
we place traffic lights and traffic lights allow us to
(05:48):
have a certain amount of cars per hours. And if
they are in the cities, you have traffic light and intersection.
If we are in freeway, then you're going to have
a traffic light at the entrance of an intersection.
Speaker 2 (06:00):
And it becomes what is known as from metering.
Speaker 3 (06:03):
As we've gotten better and getting data and we have
more information. So now we have, for example, data on
sensors on roadsides, or for example, we have vehicles that
give us information about the GPS or the phone. Then
we've gotten also better and managing traffic, and so now
you have city wide traffic lights plans, or you had
(06:24):
for example, what is variable to p limit, which I'm
sure everybody at some point you saw this message panel
on the freeway that says slow down and go at forty.
Speaker 2 (06:34):
Miles an hour.
Speaker 3 (06:35):
And so how we manage traffic has evolved throughout the year,
and it's going to evolve as we have different types
of measurements and different ways of estimating traffic.
Speaker 1 (06:45):
It seems like the traffic light was a big innovation.
Do we know anything about the history of the traffic light?
Speaker 3 (06:51):
So the first traffic light that I know of dates
back to the beginning of the nineteen hundred and I
think it's in Cleveland. At the beginning was exactly that,
like a single traffic light that would just manage traffic
at a single intersection. And then as things have evolved
nowadays we're able also to have traffic lights that are
(07:11):
synchronized and have the so called the green waves in
which you hit green at first traffic lights and then
you're able to go through all the intersections with that
green light and all of that.
Speaker 1 (07:22):
I see. So that's a little bit of the history
of traffic science and engineering. But here's the big question,
why is this problem so hard to fix? Can we
just build wider roads? Well, it turns out there is
an answer to that question, and it's something that totally shocked.
I learned it while talking to the next expert on
(07:44):
our show today, Callin Mees. Well, thank you so much
for joining us, Miss mess.
Speaker 4 (07:50):
Thanks for having me. I'm excited to be here.
Speaker 1 (07:52):
Can you please tell U who you are and what
you do.
Speaker 4 (07:54):
Sure. My name is Colin Mess and I am a
PhD candidate at the University of Delaware in the Civil,
Environmental and Construction Engineering Department. My research focuses on traffic
congestion and applying new methods such as artificial intelligence to
enhance traffic management and response to incidents.
Speaker 1 (08:14):
Very cool, bite the start us off. He tell us
why is traffic so bad? Sometimes?
Speaker 4 (08:20):
Sure, traffic congestion is bad generally as a cause of,
you know, at the highest level, a mismatch between supply
and demand. So we can't build our way out of congestion,
so to speak, as a phrase, you hear a lot
in traffic engineering, and we try to design the roads
in a way that we can accommodate a large part
(08:41):
of the demand we would see even during rush hour periods.
But we can't design the road for you know, the
busiest day of the year economically or feasibly. And so
depending on things you know that are going on, whether
there's a special event or an incident, or the weather
is really nice and people are traveling due to the season,
(09:02):
we will always inevitably have some congestion as a result
of the fact that we just can't build an infinitely
sized road.
Speaker 1 (09:09):
Oh that's interesting. I hadn't heard of that idea before.
So what do you mean by economically we can't build
the roads beginnough.
Speaker 4 (09:17):
On the one hand, I think the biggest concern regarding
that would be space, in that we we want to
build roads, hopefully in a way that has a minimal
impact on everyday life and people, that minimizes the environmental impact.
And so it would be nice if we had you know,
infinite space and could build floating roads that would really
minimize the impact on the local environment in the region.
(09:41):
But until we can do that, sadly, we have to
make a compromise in some respect to how much capacity
we can build into the road given the you know,
constraint from the land perspective and the environmental perspective.
Speaker 1 (09:53):
Uh. It's like, even if we know that in a
certain road or a certain section of a city or
a town there's going to be two hundred cars going
through a peak rush hour, we don't typically design the
road to be able to take two hundred cars an hour. Yes, exactly.
Speaker 4 (10:10):
We would try to get as close to that as
we possibly could, and in some cases we can build
to accommodate the expected rush hour congestion.
Speaker 1 (10:20):
Interesting, it's like, even if we let's say, had infinite
land to build a twenty lane road a, it may
not make sense to design it so that it can
take peak rush hour traffic because most of the time
is not peak rush hour traffic.
Speaker 4 (10:36):
Yes, exactly, And similar to how we build parking lots,
we try to build for the thirtieth busiest day of
the year, so hopefully we're encapsulating the vast majority of
the regular traffic on that road through a model like that.
But unfortunately, if we built for the you know, for
the one percent or busiest day of the year, we
likely it would not be feasible from some perspective, whether
(10:59):
that's you know, environmental due to the space constraints, or
maybe it's just an economic cost that would be too
high to be supported with the resources we have.
Speaker 1 (11:08):
Oh, the parking lot is a really good analogy. What
do you mean that you might design it for the
thirty if you said.
Speaker 4 (11:14):
Yeah, So if we look at the entire year, cross
all seasons and estimate what the demand or usage might
be of the road or the parking lot, a lot
of times the planners will pick the thirtieth busiest day
of the year as a good benchmark. And that's usually
the reason for that is because the thirtieth busiest day
is if you look at the pattern in traffic graph,
(11:34):
it would be where things start to really level off,
and then you could almost consider the twenty nine other
days as a rare incident or a rare case. For example,
by me, we have a lot of travel down to
the beach for during the summer. A lot of people
live down at the beach for three or four months
of the year, and that's a great example of a
place where we can't really design the road for just
(11:56):
those two months because the utilization outside of those few
months it's significantly lower, maybe eighty percent lower, as just
a really rough estimate, so it wouldn't make sense from
an investment perspective, especially if we have to divide a
limited amount of money across many projects throughout the state,
not just in the beach area.
Speaker 1 (12:15):
Wow, that's fascinating. He just kind of blew my mind, Colin.
Speaker 4 (12:20):
Yeah. You know, a lot of times you'll see, like
I know, I'll travel down a highway and it'll be
congested and I'll just see completely empty open space on
either side while I'm you know, on the road congested,
and I feel like a lot of people see that
and think, well, why didn't they just build the road
with more lanes. Yeah, I wish we could do that
every time. It would definitely save a lot of headache.
But yeah, just the real world challenges always make things
(12:44):
a little more difficult than expected.
Speaker 1 (12:46):
It's kind of like an almost effective life.
Speaker 4 (12:49):
Yeah, I would agree with that.
Speaker 1 (12:51):
Yeah, Okay, this really blew my mind. What Colin you're
saying is that the real reason we have traffic jams
is because we want to have them. Yes, the general
cost of traffic is that there are more cars trying
to get through a road than the road has space for.
But then why don't we just build bigger roads? Well,
the real reason is that we choose not to. It
(13:14):
doesn't make sense economically to build roads for the absolute
worst traffic days of the year, because then most of
the time the roads would be mostly empty. Instead, we
make a compromise. We accept that there are going to
be traffic jams some of the time to minimize the
cost and the burden of building bigger roads. So living
(13:34):
with traffic is a choice. Okay, So that's kind of
the real reason traffic exists. There are sometimes too many cars,
and we choose not to build roads that can take
all those cars. But here's the thing. As annoying as
it is to send in your car in a congested road.
There is something even worse about the problem of traffic,
(13:55):
and it's something that we definitely can't predict, but that
AI could potentially help us solve in the future. So
when we come back, we'll talk about what that is
and something called phantom traffic jams. So don't get off
the freeway just yet. Stay with us. We'll be right back. Hey,
(14:26):
we'll come back. We're talking about the real reasons traffic
is bad, and so far we talked about how it's
a basic supply and demand problem. There are sometimes too
many cars and not enough road to fit them all.
But here's a shocker. Roads are usually not designed to
handle peak traffic demand. It would be too expensive and
take up too much space to make roads and highways
(14:49):
big enough to handle the worst case traffic scenarios. So
in a way, traffic jams are sort of built into
our society, and according to the experts we're talking to
you today, part of the is that it's hard to
even know what the worst traffic scenario is going to be.
Here's traffic science researcher Colin Meess.
Speaker 4 (15:10):
Another issue there is that the roads take time to build,
and we find that travel demand patterns change very quickly
and suddenly, and so one issue would be that we
might design the road for the geek rush hour at
the time of building, and we'll project out based on
how we see the population changing and try to get
it as close as possible. But some road projects might
(15:32):
take twenty years to build. For example, there's a road
near me called the four seventy six where they basically
ran into that issue. It took about twenty years to
build this project, and by the time the project was built,
the population had increased to a point where the road
was instantly congested. It's almost overnight, No, it's open to
(15:52):
the public, and there's already a congestion on that road
from day one.
Speaker 1 (15:56):
Oh wow, it was sort of like a it was
obsolete from the get gode that it was opened exactly.
Speaker 4 (16:03):
And I'm sure the people working on that project, you know,
projected out thirty years, use the best population models they
could to try to really estimate how many lanes they
could build or how many lanes they should build, and
what the throughput of that road should be. But unfortunately
the models are not always correct, especially when it comes
to modeling human traffic.
Speaker 1 (16:23):
Wow, Yeah, sometimes we don't know how much traffic there's
going to be in the future exactly.
Speaker 4 (16:28):
So in the case of that road, that road was
built to help specifically commuters coming from a residential area
that was about forty five minutes away from the central
business district where they worked, and because the traffic was
so bad, a lot of people over time changed how
they got to work, whether that was using maybe some
kind of carpooling system, or maybe they were using the
(16:48):
public transit like the metro or trains to get there,
or they move closer to the city and use they
bike or something like that. And then when that new
road opens up, word spreads that hey, this new route
is great, you can take your car into the city
again with no problem, and very quickly people will then
shift their travel patterns from these alternate modes that they
(17:10):
may have been using. And so when that option becomes available,
you can see a shift in demand that may not
have been able to be predicted just looking at things
like population change and the amount of vehicles being registered
in area.
Speaker 1 (17:24):
WHOA, it's super dynamic exactly.
Speaker 4 (17:28):
That's a great way to say accounting for human behavior
can be so difficult.
Speaker 1 (17:32):
Right, That's kind of the problem is trying to predict
what people are going to do exactly.
Speaker 4 (17:36):
I feel like, well, we will always have this challenge
to some extent.
Speaker 1 (17:42):
So traffic seems almost like a fact of life. But
here's the thing. Traffic by itself is not considered the
real problem. Most people don't mind a little bit of traffic.
The problem, according to our experts, is something else.
Speaker 4 (17:57):
The interesting thing about congestion to me is that I
feel like everybody expects congestion, right Like, if you're traveling
a route during rush hour, you have an expectation that
you will be sitting in some level of congestion. But
I think what really frustrates people for good reason is uncertainty.
And I think if you asked a lot of people,
would they rather have a trip that they know to
(18:18):
work their commute's always going to be around thirty five minutes,
give or take, or some days it's twenty minutes and
some days it's fifty five. I think they would almost
always rather have the certainty that they can plan around
that I know that I'm going to my route's going
to take thirty five minutes, even if it's a little
longer than it could take. As long as it's consistent.
That's really important for people to be able to, you know,
(18:39):
plan and go about their lives and plan their travel.
Speaker 1 (18:42):
Oh, that's another super fascinating point. The problem with traffic
is not that there is traffic, because people are to
some degree okay with planning around it. It's the uncertainty
that's really frustrating.
Speaker 4 (18:55):
Yeah, I would say from the people that I've talked
to and just my lived experience, that's what frustrates the
majority of people, in myself the most. I remember when
I used to commute into the city. On a good day,
it would take thirty five minutes to get to work,
but on a bad day, it could take an hour
and a half. And you know, when you have it,
there's an expectation that I'm arriving at a certain time,
(19:15):
and so when the travel time is that unreliable, it
becomes unfortunate because I basically have to plan for the
worst case scenario, and then I may arrive, you know,
an hour earlier than I needed. And that's you know,
nobody wants to arrive to work an hour earlier if
they don't need to, So I feel like that is
one of the most frustrating pain points of traffic congestion.
Speaker 1 (19:36):
Right right, or the opposite where you know, you plan
for the average, but then you're late because there's more
traffic than.
Speaker 4 (19:43):
Usual exactly and maybe nine at a ten times you
leave with the expectation that it's going to take thirty
five minutes, and nine out of ten times it does,
but one out of ten times it's three times as long.
And I mean, you can only explain that to your
boss so many times before there's no you know, excuse,
and you just have to leave earlier and plan for
the worst case, which is unfortunate.
Speaker 1 (20:05):
Okay, I guess then the question is what causes that
uncertainty in traffic congestion.
Speaker 4 (20:10):
So I think traffic incidents are one of the biggest
drivers of cong unexpected congestion. And by a traffic incident,
in my line of work, we call them non recurrent events.
So we would call like rush hour recurring traffic, or
like a seasonal change due to like we talked about
people going down to the beach, that would be something
that you could call recurrent traffic because we expect it,
(20:31):
we can plan for it, and we kind of know
it's happening. But with an incident we would call that
non recurring congestion, and that could be anything from unexpected
inclement weather that reduces visibility and overall would lower the
average speed traveled on the road, which would then effectively
lower our capacity temporarily. It could also be, of course,
things like a traffic accident or a work zone where
(20:54):
the capacity is essentially temporarily altered. So a vehicle gets
in an accident and we need time for that vehicle
to get moved out of the lane that it's blocking.
And while that's happening, say it's a three lane road
and one lane's blocked, we're effectively reducing our capacity by
at least thirty three percent. We can't really plan for
an unexpected incident, right, These are things where we can't
(21:15):
plan around them and prevent them.
Speaker 1 (21:18):
I see. It's sort of like there's two sources of
traffic or congestion, Like there's a base level of traffic
that's just due to the like you said, the recurring causes,
which are you know, just not enough supply to meet.
Speaker 4 (21:32):
The demand exactly.
Speaker 3 (21:34):
Yeah.
Speaker 4 (21:34):
I think that's a good way to think about it,
almost as like two different types of congestion, two distinct
different types and the non recurring congestion I feel like
is the most frustrating for people because it really does
play into the travel time reliability. So you know, it
really impacts reliability in a way that you're just not
expecting at all.
Speaker 1 (21:55):
Okay, so there's general traffic just from having too many
cars in the road, but then that you can have
extra unexpected traffic because of accidents or cars breaking down,
or construction or weather. Those can be frustrating, but maybe
not as frustrating as the congestion that someplace seems to
happen for no reason at all. You know, when there's
(22:15):
a big slowdown but there isn't an accident or a
car broken down, or whether someplace cars just seem to
jam up, and that is actually a well studied phenomenon
in traffic science called phantom traffic. To explain this, here's
doctor Maria Lauda di la Monarky.
Speaker 2 (22:37):
For me, it's much more interesting is what we call
phantom traffic jams.
Speaker 1 (22:41):
Phantom traffic jams, yes, like a ghost like.
Speaker 2 (22:44):
Ghosts, because they appear out of nowhere.
Speaker 3 (22:47):
I'm sure everybody that is driven in the freeway have
lived this. At some points, you're stuck in traffic, and
you start accelerating and breaking, accelerating and breaking constantly, and
an all a sudden we are out of it and
there's nothing inside that could have caused us. There's no accident,
there's no worksite constructions, not even run shower right, And
(23:11):
this happened just because we assume intend to get distracted. Well,
we're driving, and then all of a sudden we realized
that OOPS were too close to the Kara had, so
we slightly it up on the brake, and that creates
a ripple effect that fifteen minutes later, someone else is
crossing the same freeway and alives at that point is
gonna be stuck in stop and go.
Speaker 1 (23:31):
Traffic fifteen minutes later, even up.
Speaker 2 (23:35):
To fifteen it can be even longer than that.
Speaker 1 (23:36):
Yes, whoa one simple distraction I got a text message
or I'm scrolling through Instagram and instead of paying attention
can ruin someone today twenty minutes later.
Speaker 3 (23:48):
Yes, we call it either stopping away more fun on
traffic jam because they are not caused by any physical rism,
and there's no physical rism for them to appear, not
physical obstacle.
Speaker 1 (23:58):
A list in the road, I see, So it's just
human sloppiness. Yes, let's say traffic gets flowing smoothly. What's
going to cause it to suddenly not be smooth?
Speaker 3 (24:09):
It's as simple as one driver slightly tapping on the brake.
He taps on the brake because for whatever reason, and
let's say it sees that is too close with the
car ahad, But it can simply be just not come
from the speed is going, so slightly up on the
brakes and then like the driver slightly tapping on the
brake creates basically a little disturbance, and that disturbance can templify.
(24:31):
That means that the driver behind the first driver is
gonna have to slightly up on the brake a bit more.
The third one is gonna tap on the brake harder
and harder, and harder and harder, until the last one
is gonna have to basically stop, almost stop or brake
so hard that the car stopped or you end up stopping,
and that basically creates the stopping.
Speaker 1 (24:56):
Why does it get worth each time because it's a delay.
Speaker 2 (24:59):
Because there's a delay, there's the reaction time of the drivers.
Speaker 3 (25:03):
There's delay that comes from the kra had to a
car behind, So by the time you realize that the
kara had break, you have to break as well, and
you're still going in the speed of before, so now
you have to break a bit harder respect the car had.
Speaker 1 (25:18):
Oh it's no balls, it's no balls exactly, yes, And
is it also the same for accelerating.
Speaker 3 (25:25):
It's the same with accelerating, like if you try to
catch up with the person in front of you. So
this is what happened in the phantom traffic jam, right,
So you go into this period of breaking where everybody
breaks and the last vehicle has to break the hardest.
But then everybody tries to catch up with the vehicle
in front of it. So everybody accelerate, accelerate, and the
last vehicle acces accelerate the most because it is to
(25:46):
catch up to the previous vehicle, which is counterintuitive from
what you're supposed to do. You're supposed to like just
stay at your constant speed and if it's safe to
do so.
Speaker 1 (25:59):
It just low everybody down, yes or no, good exactly exactly.
It's getting distracted or it's just human reaction time.
Speaker 2 (26:08):
It's a mike, sure of both. Sometimes it's because you're distracted.
Sometimes it's just utilized.
Speaker 1 (26:12):
Too late makes me think we should just let the
robots drive. Everyone should let the robot car drive. This
brings us to the next big question about all of this.
Could robots or AI solve the traffic problem? What if
everyone drove self driving cars with that get rid of
traffic jams? Or what if we put an AI in
(26:33):
charge of our traffic light system? Would that create a
perfect road system. When we come back, I'll ast car
experts all these questions and we'll see if it puts
them in a jam. So stay with us and we'll
be right back. Hey, welcome back. We're talking about the
(27:03):
science of traffic, and so far we've talked about what
really causes traffic and about how the real problem with
it is the uncertainty of it. We also talked about
something called phantom traffic, which is a big source of
unexpected congestion in our roads. Now the question is what
can we do about all this uncertainty. What are ways
(27:23):
we can make traffic or tolerable. As it turns out,
one of our experts today has been at the head
of a project to use robots to make the flow
of cars more efficient. Here's Professor Mardy allowed a dile monarking.
So the idea is that you planned robot cars along traffic, right, yes,
(27:46):
and somehow they're able to make things better. Can you
explain that idea a little more.
Speaker 3 (27:51):
Yeah, So this is what we actually do in my live.
So we try to exploit vehicles of technology to get
transportation better for everybody and not just for the owner
of the self driving car. And so the idea is
that when we're talking, for example, about phantom traffic jams there,
it happens because we get distracted. So if we instead
of we assum and getting distructed, we have a self
(28:13):
driving car, that self driving car doesn't get distracted and
so doesn't have to slightly tap on the brake. And
this is in the extreme case where everybody has a
self driving car, but even the mixed case where we
have I don't know, five percent self driving cars and
ninety five percent humans, we can insert those self crime
car or smart cars anyway in the midst of the humans,
(28:36):
and what they would do is simply dissipate or smooth
out traffic so that the disturbance caused by that slight
up on the brake doesn't amplify it throughout the freeway.
Speaker 1 (28:48):
So then if there's a wave of breaking, having an
automated car there to not break suddenly can fix it.
Speaker 4 (28:56):
Yeah.
Speaker 3 (28:57):
So the idea there is basically you look at the
overall traffic. You estimate what would be the best speed
for that amount of traffic for everybody to go, let's say,
at a constant speed and not have any breakdown, not
having any traffic jam. And then you tell this to
whatever smart vehicle you have there, and basically the goal
(29:18):
of that vehicle will try is to make sure that
he does not overreact to what other people are doing.
So he's not gonna catch up with the vehicle in
front of it. His skin is gonna keep on going
at that speed that the vehicle assumed to be the best.
One kind of try and keep everybody at a constant speed.
(29:39):
Oh I see, And so then all of a sudden,
fifteen minutes later, you don't have anymore these huge traffic jams.
And we did this both in close courts with one
ab and twenty human drivers, but we did this also
in freeway. So we had the largest ever experiment in
which we had one hundred smart cars Floyd and rush hour.
(30:01):
This was done in Tennessee on nine twenty four. And
the idea was exactly that can we tell those one
hundred vehicles to be like a pacer, like going the road,
maintain a certain speed that we will give it given
the current traffic condition, and see if they can smooth
out traffic or at least avoid the creation.
Speaker 1 (30:17):
Of like plant some perfect drivers. Kind of yes, and
it worked.
Speaker 3 (30:24):
Yeah, it showed that we can with a reasonable penetration rate.
So then does need to be that big, so it
can be like up to five percent. You can smooth
out traffic. And the other thing that you can do,
which is really cool, it's also reduced energy footprint of traffic.
Speaker 2 (30:40):
Because there's phantom traffic jams.
Speaker 3 (30:42):
What they do is you're currently accelerating and braking, so
that is what caused the highest farce consumption on your
vehicle or let's use an electric vehicle, so that caust
also the highest pollution. And so if you smooth out traffic,
you're able also to reduce this behavior and so reduce
U summation caused by transportation.
Speaker 1 (31:02):
I see, well that's fascinating. So what would it take
to do that, Like the city would have to buy
some automated cars and for programming, or some people would
have to volunteer too, like I want to be a
pacer card.
Speaker 3 (31:13):
I think it's more the second, Like, I think you'll
need to have some people that are willing to offer,
like their vehicles, in.
Speaker 2 (31:22):
Order to do this.
Speaker 3 (31:23):
Oh yeah, there are different ways to do this. It
can be incentive, or it can be simply volunteers, or
it can be simply some good samaritan sides that they
want to make life better for everybody else.
Speaker 1 (31:35):
Interesting, it's interesting that you can put robots to kind
of guide people to behave better Be a car that
doesn't make the common human mistake yes, overreacting.
Speaker 3 (31:45):
Yes, And also you don't want this car to we
have two different than in humans, because otherwise, if you
see some car in the road that behaves fundamentally different
and how you would expect, then you get scared while
you're driving, and then then it comes unsafe. So you
want to have something that does this without the other
human driving surrounding realizing that is doing this.
Speaker 1 (32:06):
I see.
Speaker 2 (32:07):
So that's the siddle thing to that.
Speaker 1 (32:09):
Okay, well that's fascinating. Okay, So that's one way to
make traffic flow more efficiently and reduce uncertainty. You can
have self driving cars sprinkled throughout that talk to each
other and coordinate to basically hurt everyone along and not
overreact or get distracted to break those phantom traffic waves.
Another idea to improve traffic is to basically let computers
(32:33):
take control of everything. Let's say a car breaks down
in the middle of a freeway, or if there's a
car accident that closes down the lane. A central AI
could take real time data from traffic sensors on the
road and predict what would be the best way to
reduce the impact of that unexpected incident. Then that AI
could tell people's Google Maps or ways where to go,
(32:55):
and it could change the pattern of traffic lights at
intersections to optimize flow of cars. Here's how traffic engineer
Colin Mees puts it. Okay, so let me see if
I can paint the perfect scenario here, Colin, So we
have perfect knowledge of everything that's happening on the road, Like,
(33:15):
first of all, we have historical data of like what
the baseline congestion is going to be. We have perfect
data of when accidents happen, and we can use that
right away to predict the ripple effects of that event
and then be able to coordinate with the apps people
are using to divert them and change the traffic lights
to accommodate those changes to ease that traffic.
Speaker 4 (33:38):
Yeah. I think that that was a great way to
summarize the perfect scenario. That's really what we're working towards.
And I hope that once these systems are you know,
tested more and develop more, and we really have a
reliable AI driven for example, system, then we can start
seeing things like an AI agent changes the traffic light
in response to the incident for us. I think we're
(33:58):
pretty far away from an area where the AI would
just fully be controlling the signal system, for example, without
human interaction. But I think the AI can recommend things,
simulate things, and then maybe a human says, yes, that
looks good and will change the light dynamically to accommodate
this for the next thirty minutes. And one thing I
(34:20):
guess I will add to that that we didn't talk
about would be just having better travel options too. So
really improving the multimodal infrastructure in our nation is something
that would go a long way. I know there's a
lot of people that drive that don't even really want
to drive, Like if they could take a reliable train
from A to B every day instead of having to
drive on the roads, then they would prefer to do that.
(34:43):
If we think about investing more in alternate modes like
we see in Europe, where the infrastructure exists and it's
good and it's reliable and it's affordable, people will take
those alternate modes, and that can also be another way
that we can relieve some of the stress on the
existing roadway system.
Speaker 1 (34:57):
I see, But then let me update my optimal optimal scenario.
It seems like there's an optimal optimal optimal scenario, which
is that people ride their bicycles and take the train more.
Speaker 4 (35:08):
Yeah, exactly.
Speaker 1 (35:09):
I mean not that sitting in traffic listening to say
a science podcast is necessarily a.
Speaker 4 (35:14):
Bad No, definitely not. I think if you were sitting
in traffic, that's probably one of the best ways you
could be spending your time.
Speaker 1 (35:20):
There, you go. I feel like maybe traffic also points
to a fundamental human need to come together to gather,
Like if we didn't have that need and nobody wanted
to ever align with anybody else's schedule, maybe there would
be less traffic. But because you know, schools start at
the same same time, work starts, the business day starts
(35:41):
at the same time. Everyone wants to work together. We
have that fundamental human need. Then that just naturally causes traffic.
Speaker 3 (35:49):
Yes, I think that's also part of the I wouldn't
say a problem because it's not a problem like that.
Speaker 2 (35:55):
We want to be all together. This is part of
just the situation as it is.
Speaker 3 (35:58):
And so I think it's just we need to be
better at building public transits so that people will want
to use public transit stuff taking the car or other
sorts of transportation.
Speaker 1 (36:10):
Or maybe flying cars. I think that would solve everything.
Speaker 3 (36:12):
What do you think then we're gonna have congestion in
the skies on the roads.
Speaker 1 (36:17):
But yes, works because of three D problem. Yes, yes,
it's never gonna ends it.
Speaker 3 (36:25):
I don't know, Like I think it's gonna probably get
batter us. We are getting better as time passes in theory.
I hope it's gonna be a batter as time passes.
Speaker 1 (36:36):
All right, Well, if you're sitting in traffic right now
listening to this, we really appreciate you spending your commute
with us. I'm sure traffic will clear up for you.
Pretty soon and hopefully it'll clear up for all of
us in the future. Thanks for joining us, and hey,
next week is the season finale of sign Stuff and
it's going to be a pretty dramatic one, so be
(36:56):
sure to tune in next week as we tackle one
of the biggest signs questions facing us today, Isn't AI
going to kills all? See you? Then you've been listening
to Science Stuff, the production of iHeartRadio, written and produced
(37:18):
by me or hitch Ham candidate by Rose Seguda, executive
producer Jerry Rowland, and audio engineer and mixer Kasey Peckram,
and you can follow me on social media. Just search
for PhD Comics and the name of your favorite platform.
Be sure to subscribe to sign Stuff on the iHeartRadio app,
Apple Podcasts or wherever you get your podcasts, and please
tell your friends we'll be back next Wednesday with another episode. Hey,
(37:50):
please take a second and leave us a review on
Apple Podcasts, Spotify, or wherever you listen to the podcast.
Thanks a lot,
Hey, Welcome to Sign Stuff, the production of iHeartRadio More. Hey,
cham and today we're answering the question what really makes
traffic so bad? It turns out it's not careless drivers
or accidents or roadwork. It's something much more surprising. We're
going to talk to two experts on traffic engineering. We're
(00:22):
going to step us through the history of traffic, the
real causes of it, and what we can do to
make commuting easier. To hit the accelerator because we are
clearing the road to answer the question what really makes
traffic that enjoy? Hey everyone? So, like most people, I
(00:48):
hate traffic. I always feel like I'm wasting my life
away sitting in my car, hitching my way across the
highway or a busy road. And the worst part is
that sometimes there's no clear reason for it. You've probably
had the experience of sitting in traffic for a while
and then all of a sudden, the traffic clears up,
and you think was it all for nothing? What's going on? Well,
(01:12):
to find out why that happens and to answer all
my traffic related questions, I reached out to two traffic
engineering experts. The first one is doctor Maria Lauda de
la Monice, a professor of civil and environmental engineering at
the University of California at Berkeley who works at the
no pun intended intersection between traffic science and math. She's
(01:35):
going to tell us why these random traffic jams happen,
but first I wanted to know a little bit about
the history of traffic. So here's my conversation with doctor
Maria Lauda de la Monique. Well, thank you, doctor de
le Monique for joining us.
Speaker 2 (01:51):
Thank you, it's a pleasure to a here.
Speaker 1 (01:52):
Well, I kind of have to apologize for being laid.
I was stuck in traffic.
Speaker 3 (01:57):
That seems to be a problem for everybody these days,
especially in the Bay Area where we are.
Speaker 1 (02:03):
Actually, I'm just kidding. I work from home, so I
haven't had to deal with traffic in a long time.
Speaker 3 (02:09):
I bike to avoid the traffic. I have an electric bike.
That's my commute.
Speaker 1 (02:13):
Oh that's smart. Wow, at least for now, until everybody
gets an electric bike, Yes, then there might be bike traffic.
Speaker 3 (02:19):
Yeah, there's still is some places in Europe. Alreadia experience
in traffic with bikes in the Netherlands. In Paris, it
was a problem for a bit.
Speaker 1 (02:27):
Oh wow, all right, we'll get into that, but I
thought way we could start by asking you to tell
us a little bit of the history of traffic, Like
when did traffic become a problem for humans.
Speaker 3 (02:36):
Well, people tend to think that traffic was born, let's say,
with cars, but in reality it's something that all does
humans and cities. So we can go as far back
as ancient Rome and we still have traffic and congestion.
There's Julius sister, even have the regulatory intervention to avoid
that commercial trucks or commercial cars will go on through
(02:58):
the city during daytai because they would cause congestion and
they would make too much noise. So it's a very
whole problem and we're still here trying to solve it.
Speaker 1 (03:07):
It's oldest Roman empire, yes, yes, And then a city
started to grow, we also started to experience more and
more traffic and different type of traffic.
Speaker 3 (03:20):
Obviously, with the advent of cars is when the problem
became very big because a lot of cars that are
in the road, and also became apparent that the road
cannot be shared anymore between pedestrians and cars, right where
before it was the case everybody was in their pedestrian, bicycle,
animal power vehicles, and then since the cars were there,
(03:41):
and then it became a very chronic problem. And now
it became a problem of everybody going to work in
the morning at the same time and being stuck in traffic.
Speaker 2 (03:50):
When we built freeways.
Speaker 3 (03:51):
In the fifties, now people didn't live anymore in the city,
but now they had to take the freeway to go
to the world, and now freeway became congested. And so
it's a problem that we've been dealing for a long time.
Speaker 1 (04:02):
Who it seems like any time you try to do
something about it, it still manages to come back.
Speaker 3 (04:07):
Yes, because it seems the more we make it better,
the more we make people want to use the cars,
and so instead like we build more roads and then
people use more cars instead of like fixing what we
had before.
Speaker 1 (04:20):
And so that seems like a very human problem, is
it is, Well, can you take me back a little
bit on the history of I guess, trying to manage
traffic or trying to deal with traffic, Like what did
the Romans do?
Speaker 3 (04:33):
So before the cars, what people were trying to do,
it's mostly intervened like on the human behavior. So it's
like forbid the use of cars during daytime, right, and
so you can only go in the city during nighttime
when there was animal power vehicles. So there may be
a road becomes a one way road, or you had
a policeman at an intersection that decides, like you know,
(04:57):
now it's animal power flow that goes through the intersection,
then it's pedestrian.
Speaker 2 (05:01):
Then as whatever else.
Speaker 3 (05:03):
As technology has gotten better and better, we're also getting
better and better and managing traffic. So as car came on,
also traffic light came on because now we were suddenly
able to measure traffic. And the way to measure traffic
at the beginning of the nineteen hundred was a man
standing on the side of the road and literally counting
car as time.
Speaker 1 (05:24):
Passed, right, And so graduate student, I imagine.
Speaker 3 (05:27):
At the time it was actually scientists, like there's this
famous picture of Bruce Greenshield as one of the fathers
of traffic flow theory, that's standing on the side of
the road with a camera and counting cars in the
nineteen thirties or something. And so the idea it became
managed traffic. The same way as we can measure traffic,
we place traffic lights and traffic lights allow us to
(05:48):
have a certain amount of cars per hours. And if
they are in the cities, you have traffic light and intersection.
If we are in freeway, then you're going to have
a traffic light at the entrance of an intersection.
Speaker 2 (06:00):
And it becomes what is known as from metering.
Speaker 3 (06:03):
As we've gotten better and getting data and we have
more information. So now we have, for example, data on
sensors on roadsides, or for example, we have vehicles that
give us information about the GPS or the phone. Then
we've gotten also better and managing traffic, and so now
you have city wide traffic lights plans, or you had
(06:24):
for example, what is variable to p limit, which I'm
sure everybody at some point you saw this message panel
on the freeway that says slow down and go at forty.
Speaker 2 (06:34):
Miles an hour.
Speaker 3 (06:35):
And so how we manage traffic has evolved throughout the year,
and it's going to evolve as we have different types
of measurements and different ways of estimating traffic.
Speaker 1 (06:45):
It seems like the traffic light was a big innovation.
Do we know anything about the history of the traffic light?
Speaker 3 (06:51):
So the first traffic light that I know of dates
back to the beginning of the nineteen hundred and I
think it's in Cleveland. At the beginning was exactly that,
like a single traffic light that would just manage traffic
at a single intersection. And then as things have evolved
nowadays we're able also to have traffic lights that are
(07:11):
synchronized and have the so called the green waves in
which you hit green at first traffic lights and then
you're able to go through all the intersections with that
green light and all of that.
Speaker 1 (07:22):
I see. So that's a little bit of the history
of traffic science and engineering. But here's the big question,
why is this problem so hard to fix? Can we
just build wider roads? Well, it turns out there is
an answer to that question, and it's something that totally shocked.
I learned it while talking to the next expert on
(07:44):
our show today, Callin Mees. Well, thank you so much
for joining us, Miss mess.
Speaker 4 (07:50):
Thanks for having me. I'm excited to be here.
Speaker 1 (07:52):
Can you please tell U who you are and what
you do.
Speaker 4 (07:54):
Sure. My name is Colin Mess and I am a
PhD candidate at the University of Delaware in the Civil,
Environmental and Construction Engineering Department. My research focuses on traffic
congestion and applying new methods such as artificial intelligence to
enhance traffic management and response to incidents.
Speaker 1 (08:14):
Very cool, bite the start us off. He tell us
why is traffic so bad? Sometimes?
Speaker 4 (08:20):
Sure, traffic congestion is bad generally as a cause of,
you know, at the highest level, a mismatch between supply
and demand. So we can't build our way out of congestion,
so to speak, as a phrase, you hear a lot
in traffic engineering, and we try to design the roads
in a way that we can accommodate a large part
(08:41):
of the demand we would see even during rush hour periods.
But we can't design the road for you know, the
busiest day of the year economically or feasibly. And so
depending on things you know that are going on, whether
there's a special event or an incident, or the weather
is really nice and people are traveling due to the season,
(09:02):
we will always inevitably have some congestion as a result
of the fact that we just can't build an infinitely
sized road.
Speaker 1 (09:09):
Oh that's interesting. I hadn't heard of that idea before.
So what do you mean by economically we can't build
the roads beginnough.
Speaker 4 (09:17):
On the one hand, I think the biggest concern regarding
that would be space, in that we we want to
build roads, hopefully in a way that has a minimal
impact on everyday life and people, that minimizes the environmental impact.
And so it would be nice if we had you know,
infinite space and could build floating roads that would really
minimize the impact on the local environment in the region.
(09:41):
But until we can do that, sadly, we have to
make a compromise in some respect to how much capacity
we can build into the road given the you know,
constraint from the land perspective and the environmental perspective.
Speaker 1 (09:53):
Uh. It's like, even if we know that in a
certain road or a certain section of a city or
a town there's going to be two hundred cars going
through a peak rush hour, we don't typically design the
road to be able to take two hundred cars an hour. Yes, exactly.
Speaker 4 (10:10):
We would try to get as close to that as
we possibly could, and in some cases we can build
to accommodate the expected rush hour congestion.
Speaker 1 (10:20):
Interesting, it's like, even if we let's say, had infinite
land to build a twenty lane road a, it may
not make sense to design it so that it can
take peak rush hour traffic because most of the time
is not peak rush hour traffic.
Speaker 4 (10:36):
Yes, exactly, And similar to how we build parking lots,
we try to build for the thirtieth busiest day of
the year, so hopefully we're encapsulating the vast majority of
the regular traffic on that road through a model like that.
But unfortunately, if we built for the you know, for
the one percent or busiest day of the year, we
likely it would not be feasible from some perspective, whether
(10:59):
that's you know, environmental due to the space constraints, or
maybe it's just an economic cost that would be too
high to be supported with the resources we have.
Speaker 1 (11:08):
Oh, the parking lot is a really good analogy. What
do you mean that you might design it for the
thirty if you said.
Speaker 4 (11:14):
Yeah, So if we look at the entire year, cross
all seasons and estimate what the demand or usage might
be of the road or the parking lot, a lot
of times the planners will pick the thirtieth busiest day
of the year as a good benchmark. And that's usually
the reason for that is because the thirtieth busiest day
is if you look at the pattern in traffic graph,
(11:34):
it would be where things start to really level off,
and then you could almost consider the twenty nine other
days as a rare incident or a rare case. For example,
by me, we have a lot of travel down to
the beach for during the summer. A lot of people
live down at the beach for three or four months
of the year, and that's a great example of a
place where we can't really design the road for just
(11:56):
those two months because the utilization outside of those few
months it's significantly lower, maybe eighty percent lower, as just
a really rough estimate, so it wouldn't make sense from
an investment perspective, especially if we have to divide a
limited amount of money across many projects throughout the state,
not just in the beach area.
Speaker 1 (12:15):
Wow, that's fascinating. He just kind of blew my mind, Colin.
Speaker 4 (12:20):
Yeah. You know, a lot of times you'll see, like
I know, I'll travel down a highway and it'll be
congested and I'll just see completely empty open space on
either side while I'm you know, on the road congested,
and I feel like a lot of people see that
and think, well, why didn't they just build the road
with more lanes. Yeah, I wish we could do that
every time. It would definitely save a lot of headache.
But yeah, just the real world challenges always make things
(12:44):
a little more difficult than expected.
Speaker 1 (12:46):
It's kind of like an almost effective life.
Speaker 4 (12:49):
Yeah, I would agree with that.
Speaker 1 (12:51):
Yeah, Okay, this really blew my mind. What Colin you're
saying is that the real reason we have traffic jams
is because we want to have them. Yes, the general
cost of traffic is that there are more cars trying
to get through a road than the road has space for.
But then why don't we just build bigger roads? Well,
the real reason is that we choose not to. It
(13:14):
doesn't make sense economically to build roads for the absolute
worst traffic days of the year, because then most of
the time the roads would be mostly empty. Instead, we
make a compromise. We accept that there are going to
be traffic jams some of the time to minimize the
cost and the burden of building bigger roads. So living
(13:34):
with traffic is a choice. Okay, So that's kind of
the real reason traffic exists. There are sometimes too many cars,
and we choose not to build roads that can take
all those cars. But here's the thing. As annoying as
it is to send in your car in a congested road.
There is something even worse about the problem of traffic,
(13:55):
and it's something that we definitely can't predict, but that
AI could potentially help us solve in the future. So
when we come back, we'll talk about what that is
and something called phantom traffic jams. So don't get off
the freeway just yet. Stay with us. We'll be right back. Hey,
(14:26):
we'll come back. We're talking about the real reasons traffic
is bad, and so far we talked about how it's
a basic supply and demand problem. There are sometimes too
many cars and not enough road to fit them all.
But here's a shocker. Roads are usually not designed to
handle peak traffic demand. It would be too expensive and
take up too much space to make roads and highways
(14:49):
big enough to handle the worst case traffic scenarios. So
in a way, traffic jams are sort of built into
our society, and according to the experts we're talking to
you today, part of the is that it's hard to
even know what the worst traffic scenario is going to be.
Here's traffic science researcher Colin Meess.
Speaker 4 (15:10):
Another issue there is that the roads take time to build,
and we find that travel demand patterns change very quickly
and suddenly, and so one issue would be that we
might design the road for the geek rush hour at
the time of building, and we'll project out based on
how we see the population changing and try to get
it as close as possible. But some road projects might
(15:32):
take twenty years to build. For example, there's a road
near me called the four seventy six where they basically
ran into that issue. It took about twenty years to
build this project, and by the time the project was built,
the population had increased to a point where the road
was instantly congested. It's almost overnight, No, it's open to
(15:52):
the public, and there's already a congestion on that road
from day one.
Speaker 1 (15:56):
Oh wow, it was sort of like a it was
obsolete from the get gode that it was opened exactly.
Speaker 4 (16:03):
And I'm sure the people working on that project, you know,
projected out thirty years, use the best population models they
could to try to really estimate how many lanes they
could build or how many lanes they should build, and
what the throughput of that road should be. But unfortunately
the models are not always correct, especially when it comes
to modeling human traffic.
Speaker 1 (16:23):
Wow, Yeah, sometimes we don't know how much traffic there's
going to be in the future exactly.
Speaker 4 (16:28):
So in the case of that road, that road was
built to help specifically commuters coming from a residential area
that was about forty five minutes away from the central
business district where they worked, and because the traffic was
so bad, a lot of people over time changed how
they got to work, whether that was using maybe some
kind of carpooling system, or maybe they were using the
(16:48):
public transit like the metro or trains to get there,
or they move closer to the city and use they
bike or something like that. And then when that new
road opens up, word spreads that hey, this new route
is great, you can take your car into the city
again with no problem, and very quickly people will then
shift their travel patterns from these alternate modes that they
(17:10):
may have been using. And so when that option becomes available,
you can see a shift in demand that may not
have been able to be predicted just looking at things
like population change and the amount of vehicles being registered
in area.
Speaker 1 (17:24):
WHOA, it's super dynamic exactly.
Speaker 4 (17:28):
That's a great way to say accounting for human behavior
can be so difficult.
Speaker 1 (17:32):
Right, That's kind of the problem is trying to predict
what people are going to do exactly.
Speaker 4 (17:36):
I feel like, well, we will always have this challenge
to some extent.
Speaker 1 (17:42):
So traffic seems almost like a fact of life. But
here's the thing. Traffic by itself is not considered the
real problem. Most people don't mind a little bit of traffic.
The problem, according to our experts, is something else.
Speaker 4 (17:57):
The interesting thing about congestion to me is that I
feel like everybody expects congestion, right Like, if you're traveling
a route during rush hour, you have an expectation that
you will be sitting in some level of congestion. But
I think what really frustrates people for good reason is uncertainty.
And I think if you asked a lot of people,
would they rather have a trip that they know to
(18:18):
work their commute's always going to be around thirty five minutes,
give or take, or some days it's twenty minutes and
some days it's fifty five. I think they would almost
always rather have the certainty that they can plan around
that I know that I'm going to my route's going
to take thirty five minutes, even if it's a little
longer than it could take. As long as it's consistent.
That's really important for people to be able to, you know,
(18:39):
plan and go about their lives and plan their travel.
Speaker 1 (18:42):
Oh, that's another super fascinating point. The problem with traffic
is not that there is traffic, because people are to
some degree okay with planning around it. It's the uncertainty
that's really frustrating.
Speaker 4 (18:55):
Yeah, I would say from the people that I've talked
to and just my lived experience, that's what frustrates the
majority of people, in myself the most. I remember when
I used to commute into the city. On a good day,
it would take thirty five minutes to get to work,
but on a bad day, it could take an hour
and a half. And you know, when you have it,
there's an expectation that I'm arriving at a certain time,
(19:15):
and so when the travel time is that unreliable, it
becomes unfortunate because I basically have to plan for the
worst case scenario, and then I may arrive, you know,
an hour earlier than I needed. And that's you know,
nobody wants to arrive to work an hour earlier if
they don't need to, So I feel like that is
one of the most frustrating pain points of traffic congestion.
Speaker 1 (19:36):
Right right, or the opposite where you know, you plan
for the average, but then you're late because there's more
traffic than.
Speaker 4 (19:43):
Usual exactly and maybe nine at a ten times you
leave with the expectation that it's going to take thirty
five minutes, and nine out of ten times it does,
but one out of ten times it's three times as long.
And I mean, you can only explain that to your
boss so many times before there's no you know, excuse,
and you just have to leave earlier and plan for
the worst case, which is unfortunate.
Speaker 1 (20:05):
Okay, I guess then the question is what causes that
uncertainty in traffic congestion.
Speaker 4 (20:10):
So I think traffic incidents are one of the biggest
drivers of cong unexpected congestion. And by a traffic incident,
in my line of work, we call them non recurrent events.
So we would call like rush hour recurring traffic, or
like a seasonal change due to like we talked about
people going down to the beach, that would be something
that you could call recurrent traffic because we expect it,
(20:31):
we can plan for it, and we kind of know
it's happening. But with an incident we would call that
non recurring congestion, and that could be anything from unexpected
inclement weather that reduces visibility and overall would lower the
average speed traveled on the road, which would then effectively
lower our capacity temporarily. It could also be, of course,
things like a traffic accident or a work zone where
(20:54):
the capacity is essentially temporarily altered. So a vehicle gets
in an accident and we need time for that vehicle
to get moved out of the lane that it's blocking.
And while that's happening, say it's a three lane road
and one lane's blocked, we're effectively reducing our capacity by
at least thirty three percent. We can't really plan for
an unexpected incident, right, These are things where we can't
(21:15):
plan around them and prevent them.
Speaker 1 (21:18):
I see. It's sort of like there's two sources of
traffic or congestion, Like there's a base level of traffic
that's just due to the like you said, the recurring causes,
which are you know, just not enough supply to meet.
Speaker 4 (21:32):
The demand exactly.
Speaker 3 (21:34):
Yeah.
Speaker 4 (21:34):
I think that's a good way to think about it,
almost as like two different types of congestion, two distinct
different types and the non recurring congestion I feel like
is the most frustrating for people because it really does
play into the travel time reliability. So you know, it
really impacts reliability in a way that you're just not
expecting at all.
Speaker 1 (21:55):
Okay, so there's general traffic just from having too many
cars in the road, but then that you can have
extra unexpected traffic because of accidents or cars breaking down,
or construction or weather. Those can be frustrating, but maybe
not as frustrating as the congestion that someplace seems to
happen for no reason at all. You know, when there's
(22:15):
a big slowdown but there isn't an accident or a
car broken down, or whether someplace cars just seem to
jam up, and that is actually a well studied phenomenon
in traffic science called phantom traffic. To explain this, here's
doctor Maria Lauda di la Monarky.
Speaker 2 (22:37):
For me, it's much more interesting is what we call
phantom traffic jams.
Speaker 1 (22:41):
Phantom traffic jams, yes, like a ghost like.
Speaker 2 (22:44):
Ghosts, because they appear out of nowhere.
Speaker 3 (22:47):
I'm sure everybody that is driven in the freeway have
lived this. At some points, you're stuck in traffic, and
you start accelerating and breaking, accelerating and breaking constantly, and
an all a sudden we are out of it and
there's nothing inside that could have caused us. There's no accident,
there's no worksite constructions, not even run shower right, And
(23:11):
this happened just because we assume intend to get distracted. Well,
we're driving, and then all of a sudden we realized
that OOPS were too close to the Kara had, so
we slightly it up on the brake, and that creates
a ripple effect that fifteen minutes later, someone else is
crossing the same freeway and alives at that point is
gonna be stuck in stop and go.
Speaker 1 (23:31):
Traffic fifteen minutes later, even up.
Speaker 2 (23:35):
To fifteen it can be even longer than that.
Speaker 1 (23:36):
Yes, whoa one simple distraction I got a text message
or I'm scrolling through Instagram and instead of paying attention
can ruin someone today twenty minutes later.
Speaker 3 (23:48):
Yes, we call it either stopping away more fun on
traffic jam because they are not caused by any physical rism,
and there's no physical rism for them to appear, not
physical obstacle.
Speaker 1 (23:58):
A list in the road, I see, So it's just
human sloppiness. Yes, let's say traffic gets flowing smoothly. What's
going to cause it to suddenly not be smooth?
Speaker 3 (24:09):
It's as simple as one driver slightly tapping on the brake.
He taps on the brake because for whatever reason, and
let's say it sees that is too close with the
car ahad, But it can simply be just not come
from the speed is going, so slightly up on the
brakes and then like the driver slightly tapping on the
brake creates basically a little disturbance, and that disturbance can templify.
(24:31):
That means that the driver behind the first driver is
gonna have to slightly up on the brake a bit more.
The third one is gonna tap on the brake harder
and harder, and harder and harder, until the last one
is gonna have to basically stop, almost stop or brake
so hard that the car stopped or you end up stopping,
and that basically creates the stopping.
Speaker 1 (24:56):
Why does it get worth each time because it's a delay.
Speaker 2 (24:59):
Because there's a delay, there's the reaction time of the drivers.
Speaker 3 (25:03):
There's delay that comes from the kra had to a
car behind, So by the time you realize that the
kara had break, you have to break as well, and
you're still going in the speed of before, so now
you have to break a bit harder respect the car had.
Speaker 1 (25:18):
Oh it's no balls, it's no balls exactly, yes, And
is it also the same for accelerating.
Speaker 3 (25:25):
It's the same with accelerating, like if you try to
catch up with the person in front of you. So
this is what happened in the phantom traffic jam, right,
So you go into this period of breaking where everybody
breaks and the last vehicle has to break the hardest.
But then everybody tries to catch up with the vehicle
in front of it. So everybody accelerate, accelerate, and the
last vehicle acces accelerate the most because it is to
(25:46):
catch up to the previous vehicle, which is counterintuitive from
what you're supposed to do. You're supposed to like just
stay at your constant speed and if it's safe to
do so.
Speaker 1 (25:59):
It just low everybody down, yes or no, good exactly exactly.
It's getting distracted or it's just human reaction time.
Speaker 2 (26:08):
It's a mike, sure of both. Sometimes it's because you're distracted.
Sometimes it's just utilized.
Speaker 1 (26:12):
Too late makes me think we should just let the
robots drive. Everyone should let the robot car drive. This
brings us to the next big question about all of this.
Could robots or AI solve the traffic problem? What if
everyone drove self driving cars with that get rid of
traffic jams? Or what if we put an AI in
(26:33):
charge of our traffic light system? Would that create a
perfect road system. When we come back, I'll ast car
experts all these questions and we'll see if it puts
them in a jam. So stay with us and we'll
be right back. Hey, welcome back. We're talking about the
(27:03):
science of traffic, and so far we've talked about what
really causes traffic and about how the real problem with
it is the uncertainty of it. We also talked about
something called phantom traffic, which is a big source of
unexpected congestion in our roads. Now the question is what
can we do about all this uncertainty. What are ways
(27:23):
we can make traffic or tolerable. As it turns out,
one of our experts today has been at the head
of a project to use robots to make the flow
of cars more efficient. Here's Professor Mardy allowed a dile monarking.
So the idea is that you planned robot cars along traffic, right, yes,
(27:46):
and somehow they're able to make things better. Can you
explain that idea a little more.
Speaker 3 (27:51):
Yeah, So this is what we actually do in my live.
So we try to exploit vehicles of technology to get
transportation better for everybody and not just for the owner
of the self driving car. And so the idea is
that when we're talking, for example, about phantom traffic jams there,
it happens because we get distracted. So if we instead
of we assum and getting distructed, we have a self
(28:13):
driving car, that self driving car doesn't get distracted and
so doesn't have to slightly tap on the brake. And
this is in the extreme case where everybody has a
self driving car, but even the mixed case where we
have I don't know, five percent self driving cars and
ninety five percent humans, we can insert those self crime
car or smart cars anyway in the midst of the humans,
(28:36):
and what they would do is simply dissipate or smooth
out traffic so that the disturbance caused by that slight
up on the brake doesn't amplify it throughout the freeway.
Speaker 1 (28:48):
So then if there's a wave of breaking, having an
automated car there to not break suddenly can fix it.
Speaker 4 (28:56):
Yeah.
Speaker 3 (28:57):
So the idea there is basically you look at the
overall traffic. You estimate what would be the best speed
for that amount of traffic for everybody to go, let's say,
at a constant speed and not have any breakdown, not
having any traffic jam. And then you tell this to
whatever smart vehicle you have there, and basically the goal
(29:18):
of that vehicle will try is to make sure that
he does not overreact to what other people are doing.
So he's not gonna catch up with the vehicle in
front of it. His skin is gonna keep on going
at that speed that the vehicle assumed to be the best.
One kind of try and keep everybody at a constant speed.
(29:39):
Oh I see, And so then all of a sudden,
fifteen minutes later, you don't have anymore these huge traffic jams.
And we did this both in close courts with one
ab and twenty human drivers, but we did this also
in freeway. So we had the largest ever experiment in
which we had one hundred smart cars Floyd and rush hour.
(30:01):
This was done in Tennessee on nine twenty four. And
the idea was exactly that can we tell those one
hundred vehicles to be like a pacer, like going the road,
maintain a certain speed that we will give it given
the current traffic condition, and see if they can smooth
out traffic or at least avoid the creation.
Speaker 1 (30:17):
Of like plant some perfect drivers. Kind of yes, and
it worked.
Speaker 3 (30:24):
Yeah, it showed that we can with a reasonable penetration rate.
So then does need to be that big, so it
can be like up to five percent. You can smooth
out traffic. And the other thing that you can do,
which is really cool, it's also reduced energy footprint of traffic.
Speaker 2 (30:40):
Because there's phantom traffic jams.
Speaker 3 (30:42):
What they do is you're currently accelerating and braking, so
that is what caused the highest farce consumption on your
vehicle or let's use an electric vehicle, so that caust
also the highest pollution. And so if you smooth out traffic,
you're able also to reduce this behavior and so reduce
U summation caused by transportation.
Speaker 1 (31:02):
I see, well that's fascinating. So what would it take
to do that, Like the city would have to buy
some automated cars and for programming, or some people would
have to volunteer too, like I want to be a
pacer card.
Speaker 3 (31:13):
I think it's more the second, Like, I think you'll
need to have some people that are willing to offer,
like their vehicles, in.
Speaker 2 (31:22):
Order to do this.
Speaker 3 (31:23):
Oh yeah, there are different ways to do this. It
can be incentive, or it can be simply volunteers, or
it can be simply some good samaritan sides that they
want to make life better for everybody else.
Speaker 1 (31:35):
Interesting, it's interesting that you can put robots to kind
of guide people to behave better Be a car that
doesn't make the common human mistake yes, overreacting.
Speaker 3 (31:45):
Yes, And also you don't want this car to we
have two different than in humans, because otherwise, if you
see some car in the road that behaves fundamentally different
and how you would expect, then you get scared while
you're driving, and then then it comes unsafe. So you
want to have something that does this without the other
human driving surrounding realizing that is doing this.
Speaker 1 (32:06):
I see.
Speaker 2 (32:07):
So that's the siddle thing to that.
Speaker 1 (32:09):
Okay, well that's fascinating. Okay, So that's one way to
make traffic flow more efficiently and reduce uncertainty. You can
have self driving cars sprinkled throughout that talk to each
other and coordinate to basically hurt everyone along and not
overreact or get distracted to break those phantom traffic waves.
Another idea to improve traffic is to basically let computers
(32:33):
take control of everything. Let's say a car breaks down
in the middle of a freeway, or if there's a
car accident that closes down the lane. A central AI
could take real time data from traffic sensors on the
road and predict what would be the best way to
reduce the impact of that unexpected incident. Then that AI
could tell people's Google Maps or ways where to go,
(32:55):
and it could change the pattern of traffic lights at
intersections to optimize flow of cars. Here's how traffic engineer
Colin Mees puts it. Okay, so let me see if
I can paint the perfect scenario here, Colin, So we
have perfect knowledge of everything that's happening on the road, Like,
(33:15):
first of all, we have historical data of like what
the baseline congestion is going to be. We have perfect
data of when accidents happen, and we can use that
right away to predict the ripple effects of that event
and then be able to coordinate with the apps people
are using to divert them and change the traffic lights
to accommodate those changes to ease that traffic.
Speaker 4 (33:38):
Yeah. I think that that was a great way to
summarize the perfect scenario. That's really what we're working towards.
And I hope that once these systems are you know,
tested more and develop more, and we really have a
reliable AI driven for example, system, then we can start
seeing things like an AI agent changes the traffic light
in response to the incident for us. I think we're
(33:58):
pretty far away from an area where the AI would
just fully be controlling the signal system, for example, without
human interaction. But I think the AI can recommend things,
simulate things, and then maybe a human says, yes, that
looks good and will change the light dynamically to accommodate
this for the next thirty minutes. And one thing I
(34:20):
guess I will add to that that we didn't talk
about would be just having better travel options too. So
really improving the multimodal infrastructure in our nation is something
that would go a long way. I know there's a
lot of people that drive that don't even really want
to drive, Like if they could take a reliable train
from A to B every day instead of having to
drive on the roads, then they would prefer to do that.
(34:43):
If we think about investing more in alternate modes like
we see in Europe, where the infrastructure exists and it's
good and it's reliable and it's affordable, people will take
those alternate modes, and that can also be another way
that we can relieve some of the stress on the
existing roadway system.
Speaker 1 (34:57):
I see, But then let me update my optimal optimal scenario.
It seems like there's an optimal optimal optimal scenario, which
is that people ride their bicycles and take the train more.
Speaker 4 (35:08):
Yeah, exactly.
Speaker 1 (35:09):
I mean not that sitting in traffic listening to say
a science podcast is necessarily a.
Speaker 4 (35:14):
Bad No, definitely not. I think if you were sitting
in traffic, that's probably one of the best ways you
could be spending your time.
Speaker 1 (35:20):
There, you go. I feel like maybe traffic also points
to a fundamental human need to come together to gather,
Like if we didn't have that need and nobody wanted
to ever align with anybody else's schedule, maybe there would
be less traffic. But because you know, schools start at
the same same time, work starts, the business day starts
(35:41):
at the same time. Everyone wants to work together. We
have that fundamental human need. Then that just naturally causes traffic.
Speaker 3 (35:49):
Yes, I think that's also part of the I wouldn't
say a problem because it's not a problem like that.
Speaker 2 (35:55):
We want to be all together. This is part of
just the situation as it is.
Speaker 3 (35:58):
And so I think it's just we need to be
better at building public transits so that people will want
to use public transit stuff taking the car or other
sorts of transportation.
Speaker 1 (36:10):
Or maybe flying cars. I think that would solve everything.
Speaker 3 (36:12):
What do you think then we're gonna have congestion in
the skies on the roads.
Speaker 1 (36:17):
But yes, works because of three D problem. Yes, yes,
it's never gonna ends it.
Speaker 3 (36:25):
I don't know, Like I think it's gonna probably get
batter us. We are getting better as time passes in theory.
I hope it's gonna be a batter as time passes.
Speaker 1 (36:36):
All right, Well, if you're sitting in traffic right now
listening to this, we really appreciate you spending your commute
with us. I'm sure traffic will clear up for you.
Pretty soon and hopefully it'll clear up for all of
us in the future. Thanks for joining us, and hey,
next week is the season finale of sign Stuff and
it's going to be a pretty dramatic one, so be
(36:56):
sure to tune in next week as we tackle one
of the biggest signs questions facing us today, Isn't AI
going to kills all? See you? Then you've been listening
to Science Stuff, the production of iHeartRadio, written and produced
(37:18):
by me or hitch Ham candidate by Rose Seguda, executive
producer Jerry Rowland, and audio engineer and mixer Kasey Peckram,
and you can follow me on social media. Just search
for PhD Comics and the name of your favorite platform.
Be sure to subscribe to sign Stuff on the iHeartRadio app,
Apple Podcasts or wherever you get your podcasts, and please
tell your friends we'll be back next Wednesday with another episode. Hey,
(37:50):
please take a second and leave us a review on
Apple Podcasts, Spotify, or wherever you listen to the podcast.
Thanks a lot,
Popular Podcasts
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.
Hey Jonas!
Hey Jonas! The official Jonas Brothers podcast. Hosted by Kevin, Joe, and Nick Jonas. It’s the Jonas Brothers you know... musicians, actors, and well, yes, brothers. Now, they’re sharing another side of themselves in the playful, intimate, and irreverent way only they can. Spend time with the Jonas Brothers here and stay a little bit longer for deep conversations like never before.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by Audiochuck Media Company.