Pains, Trains and Automobiles – Margins of Error

We are in the center of the great island of Manhattan at Penn Station, the Moynihan part of Penn Station, which was just built up a couple of years ago. The sunlight is beaming down. Amtrak trains come through here. And you might be asking yourself a question, why the heck am I at Penn Station? The real reason I’m here is because I came across a stat a few days ago that I found absolutely extraordinary. It turns out that less than 50% of Americans.

Proceed to Track 4.

In any event, this statistic about.

Attention please.

See, this is what happens when you try to record on location, folks. But as I was trying to say, for all the attention that air travel gets, less than 50% of Americans flying a plane every year. When I read that, I was shocked. So that’s why I took a trip down to Penn Station to better connect with a more grounded form of transportation.

We pay so much attention to plane travel is the way to go long distances. And I want to explore that maybe there are other ways that we can really go long distances, ways that we can enjoy it, ways to improve upon what could otherwise be a miserable experience.

Now, I’ve got to be honest with you, I’m not really the biggest travel aficionado in the world. Heck, it’s probably fair to say that I just do not like traveling at all. So why am I doing an episode about travel? It’s a fair question. Traveling can be really cool. After all, who doesn’t like a good vacation? But it’s the act of traveling, itself, that can cause so much pain, be it from cramped leg room, mysterious service fees, or just the stress of having a run through an airport to make a connecting flight. But I want to explore that pain. And maybe, just maybe, by the end of today’s episode, we’ll find some ways to make traveling in this country just a little bit better. So fasten your seat belts and keep your arms and legs inside the vehicle. Today, we’re tackling the world of travel, from the most efficient way to board a plane to the possible biofuel of the future. Plus, how to make your dream road trip or train ride come true. I’m Harry Enten and this is Margins of Error.

All right. It’s time for another confession. And as embarrassing as it may be to reveal, I don’t have a driver’s license. Of course, in New York, it’s not really a big deal. I can walk to work, and if I need to get out of the city, there are plenty of local train lines nearby, which is good because out of all the possible ways to travel, I happen to like trains the most. They’re easy to board, not super cramped. And if you want to get up and walk around, well, you can do that pretty easily. So let’s start here today. Let’s talk trains.

Train travel, really is such a beautiful way to get your time back, while getting from point A to point B. So that was a big reason why I also wanted to get on the train, is because there is just such a unique opportunity to get a new perspective of not just traveling across the country, but a new perspective of just traveling in general.

This is Lisa Dougherty. These days, she’s a digital content director living in Portland, Oregon. But back in 2015, she quit her advertising job and decided to see what else America had to offer.

So I came up with a kind of a haphazard plan and I was like, I’m going to take three months and I’m going to go around the country. And I had a very loose idea of like all the places that I wanted to go. And I bought my first ticket and then just got on and went.

How many total cities do you think you went to? Obviously I don’t think you could list the exact number, but approximately speaking?

I was trying to do the math the other day and I think I stopped in 12 different cities and I took about 15 to 18 trains, Amtrak only. So that doesn’t include like New Jersey Transit or like some of those like smaller routes. So yeah, I think I estimated around 12 different cities, 15 to 18 different routes.

Okay. So there’s both good and the bad when it comes to train travel. What were some of your favorite parts?

Meeting people! I’m a pretty social person. I have no problem going up and just talking to people. I love hearing other people’s stories and perspectives. You know, like some people that I met, like the guy who worked on the train, he was actually a rapper. And like we listened to some of his tracks together and it was just cool to, like, learn a little bit about his story. And one guy that I met, he was really active and like, really liked to work out. And so during like one of the quick, like the smoke break stops that they make during these like long hauls, you can go outside for about 10, 15 minutes. We both went outside and like jump roped together and it was just a really cool way to interact with each other, like share stories and like also find common ground and do that together. And those are some of my favorite parts about train travel, is definitely the people and the stories that I got to hear from everyone.

It sounds like what sets train travel apart from, let’s say, flying, going on a bus, certainly driving by yourself is that train travel is a much more social activity if you want it to be.

Yeah, you can be as social as you want to be. If you don’t like to be social, you can definitely just go to the lounge car and just look out the window for hours at a time. You can just read a book. So there is definitely some variety and creativity in the type of experience you can have on the train. And you always have like this moving scene that’s happening while you’re traveling. Like the landscape around the U.S. Is just so beautiful and incredible and that’s always available to you. You can just stare out the window for hours and it’s very nice.

So what were some of your less or not so favorite parts of the train traveling experience?

I mean, train travel is really, is romantic to an extent. You know, like some of the routes were really well kept and others weren’t. So, for example, going from Los Angeles to Albuquerque, beautiful route, the landscape is like, amazing. But the train itself, wasn’t really well kept or like, maintained, so it was like, just not as nice as some of the other routes. So just not as clean and very packed. The whole time, the whole three months that I took the train around the country, I never slept in a sleeper car. And the reason for that was the price barrier. So I wasn’t working at the time and I was very conscious of my spending. So a one-way, could be anywhere from, you know, like $75 to you know, $86, maybe a little over $100, but a sleeper car was like, you know, $400 to $500 a night. And I’m sure it’s a price barrier for a lot of people. So it will be great if that was a little bit more accessible. So I understand that there is some challenges that come with some of these long haul or just being on the railways in general. But travel in general is really remarkable. So if you can remember that, sometimes I think it can help put things into perspective while traveling.

Lisa’s right. Trains aren’t perfect, even though she and I are big fans of them. But let’s say that you don’t want to take a train ride. You want to see the country from the comfort of your own car. Well, don’t worry. We found an expert in the art of creating the perfect road trip.

I’m Dr. Randy Olson. I’m a computer scientist, artificial intelligence researcher, broadly, technologist. And also, like, just generally a giant geek.

And are you on a road trip right now or no?

Yeah, yeah, we’re we’re actually right in the middle of a big road trip that we that we actually use surprise, surprise, we used artificial intelligence to optimize it for us.

So why do you think that the road trip is the superior method of travel?

Road trips speak to like the spirit of every red-blooded American, right? You know, it’s freedom, open road, music blaring on the radio. We have so many good playlists we put together for this trip. And you have the absolute freedom to, you know, be going in a general direction, have a general idea of where you want to go. But, you know what, if you see this really cool place alongside the road and you want to stop there for an hour or for a day, you’re totally free to. We actually have fixed up our car so we can sleep in the car, itself. It’s a Honda Element, so we’re not packed in like sardines. And so, you know, we can pull in at any campground whenever we need to camp for the night and then hit the road again.

Back in 2015, Randy took two of his biggest passions computer science and road trips, and worked together with science writer Tracy Staedter, then at Discovery News, to create the algorithmically verified, ultimate road trip across America.

So Tracy went off and she agreed to do the research on like what are the most interesting places to visit in the US? Limiting to one per state. And she ended up picking like amazing places like, you know, national parks like Yellowstone, historically significant places like Mt. Vernon, and some seemingly oddball places like the C.W. Parker Carousel Museum. You have to experience it for yourself. I’m not going to make a comment on it here. My, my job was to find the road trip that stopped at each of these locations with the shortest drive possible, right? And this is actually like a harder problem than you might think, because if we have 50 destinations that we want to stop at, if you get into the math, the combinatorics of it, there are 3 times 10 to the 64, possible ways to arrange those 50 destinations. Right? And to put that ludicrous number into context, if you asked your laptop to measure every possible, like to even just to generate every possible combination of those 50 destinations, your computer would find the optimal route by trying every single one in about 9.64 times ten to the 52 years.

Yeah, I like to play on my trips, you know, maybe at most a year in advance. Like millions of years into the future is just a little too far off, you know, too far out to think about a road a little. So the concept of this route optimization algorithm, is essentially we start with a random arrangement of these destinations, right? We’re going to go to Yosemite first and then we’re going to go to Glacier National Park next, etc.. You know, for all 50 of them, what’s what’s the order that we’re going to visit them in? And then essentially, we take two of the destinations and we swap their position in this list. And then we say, okay, is this new road trip shorter in terms of distance? Right? And we’re plugging this into Google Maps every time to measure those driving distances every single time. And essentially, we just repeated this process, right? Randomly swap two of the destinations, measure the new road trip. Is it shorter? If yes, keep it. If not, throw it out and just keep trying, trying, trying. It takes only a few minutes on my MacBook to find the optimal driving route, right? Which is really cool.

We’ll link to Randy’s website in the show notes for this episode so you can see the perfect road trip for yourself. The map looks, well, chaotic at best, as it veers from state to state. If you really blitzed through the whole thing, you probably could knock it out and I don’t know, say a couple of weeks. But Randy recommends giving it a month or two to really take in the sights. Like, here’s one leg of the trip. You drive north from the Grand Canyon up through Utah and Idaho before arriving in Yellowstone, in Wyoming. Then you lurch back south through Colorado and New Mexico until you get to the Alamo in Texas.

For me. You know, I want to visit as many interesting and beautiful places as possible while driving the least amount of time between them. And that’s why I designed this road trip and I designed this algorithm.

Let me ask you, as, you know, we have people listening here who, you know, are thinking about the way that they may spend their next vacation time or get from place to place. What do people need to do, you know, just to get that first road trip out and actually do it?

I would say start small, start local, right? Like start with a trip where you can leave early in the morning, if you’re a morning person, and then be back home by night. So, yeah, you know, start local. Hit up a bunch of local parks. I think that’s a great way. And then you could build up, you know, to overnight trips that maybe last the weekend. Try doing a theme for your road trip, right? It can be parks. Gosh, I met someone who wanted to eat like a hot dog in every state, you know, like, just think about what you love and how you can combine travel with that, and then, you know, maybe, maybe someday you can build up to, like, sleeping in a Honda Element, you know?

Of course, there are more ways to get around than trains or Honda Elements. One thing we wanted to talk about in this episode was buses. So a while back, I put a call on my Twitter hoping to hear from a listener or two, who really enjoyed long distance bus rides. But the general response to my question could be summed up by one reply I got: crickets. Turns out you’d be hard pressed to find anyone who really loves bus travel, which may be bad news for some airline customers moving forward. Indeed, one consequence of the pandemic is that we now find ourselves in a pilot shortage with regional airlines and shorter routes being hit the hardest. As a result, some airlines are implementing buses to ferry passengers from smaller airports to their regional hubs. But we all know that air travel isn’t perfect either. And coming up after the break, we’ll talk with an astrophysicist about how to fix the nightmare of boarding an airplane.

If it took half an hour to load an airplane, than you can do it in 10 minutes with my method.

Hey, folks, welcome back. So before the break, we were talking about airlines and their pivot to, of all things, buses, which it turns out everyone hates. But I’ll be honest with you, I don’t care for flying either. Airports are stressful. Planes are cramped. And there are just so many, so many lines.

I arrived at the airport and waited in line to go through security. And then I arrived at the gate and waited in line there, and then walked down the jetway to the airplane and waited in line there as well. And when I was waiting in the jetway, the thought occurred to me that there had to be a better way to get people onto the airplane once they got through the gate.

I’m an associate professor of physics at UNLV. Most of my research is in the field of exoplanets, so those are planets that orbit distant stars. But I have done some work in particle astrophysics and cosmology in the past.

But spurred by personal experience waiting in line, after line at airports, Jason has also done some work on answering life’s biggest question: What’s the best way for passengers to board an airplane? The most common method is called block boarding, though it’s also referred to as group boarding, front to back, or back to front, depending on where you get on the plane.

Oh, I was completely sold and entirely confident in the fact that you would board the airplane from the back of the airplane to the front. It was just so obvious going in that that was going to be the right answer. It was the first thing that I tried is, “okay, what’s the fastest way to put passengers onto an airplane? Well, it’s obviously to board from the back to the front.” So I loaded that up, ran it, and then thought to myself, okay, so the worst possible way must be boarding from the front to the back where everybody gets in everybody’s way, no matter what. And so I lined the passengers up in reverse order and ran the code and it ended up taking almost exactly the same amount of time. And that was when I started thinking that the problem might actually be a bit more interesting than what I’d originally thought.

So on to less common ways to board a plane. First, there’s the WILMA method or window, middle, aisle, which you can probably guess, boards passengers by their seat location within their row. So window seats first, then middle seats, then aisle. It’s faster than block boarding and people like it more. But for whatever reason, it’s not used all that often. Maybe it’s that acronym. But that brings us to the boarding scheme that you hear about pretty frequently: Southwest Airlines and their unique brand of anarchy. Okay, that’s not quite fair. I actually like Southwest and took it a few months ago and it worked pretty well. But anarchy is kind of what it is.

Southwest Airlines’ approach is known to be among the best, if not the best in the industry in terms of getting people on board quickly. Mainly, because it lets people make their own decisions. They get on board and they say, “wow, that’s crowded. I’m going to sit here instead,” where most other methods don’t allow that kind of liberty.

Think about that. Of all the options you’ll find when boarding a commercial plane, believe it or not, this is usually the fastest way to do it. Which brings us back to Jason Steffen and his quest to find an even better way to board a plane.

The initial step was to come up with some kind of model for passengers walking, like working their way down an airplane. And it turned out that I thought that that was going to take quite a while to develop that piece of software, but it ended up only taking a couple of hours. So it was a little embarrassing that it was so straightforward to implement once I sat down and worked on it. But I had spent so much time thinking about it and not enough time just doing it.

Of course, it’s a little more complicated than that, because you have to accommodate for the fact that airplanes are cramped. So Jason needed to figure out the right way for his computer model to actually simulate the boarding process from things like how long do people take to stow their luggage, to how much space they’d have to move around.

It was a fairly small number of parameters that every individual passenger had. But that was enough to reproduce what you commonly see in lines. Like, for example, when you come to stop at a stoplight, you can see how everyone bunches up when they’re stopped at the stoplight, and then when the light turns green, then the cars spread out as they pick up speed and move on. And this piece of code that will just, you know, wait until there’s a space in front of you before you start moving, reproduce that kind of behavior that you observe in all sorts of other places where you’re waiting in line. So that made me pretty confident that I was reproducing the bulk of the behavior that was important. People take time to put their luggage away and they need personal space to do it.

And voila! The real problem here, the bottleneck had been identified. Think about it. When you’re boarding a plane, the only thing you really end up waiting for is every single person ahead of you to store their luggage. Once they’re in their seats, it’s all fine. So once Jason figured that out, and he had his model, it was time to see how to actually improve boarding.

So you want adjacent passengers in line, to have their seat assignments spread all throughout the airplane, so that when one person comes to stop at their row, the next person behind them is able to stop at their row. In this case, it was two rows away and then both of them could put their luggage away at the same time without getting in each other’s way, and they could sit down at the same time. The moral of the story was that you wanted to have adjacent passengers in line be as close together as possible, and yet still allowing them to comfortably put their luggage away so that they could do it simultaneously. If it took half an hour to load an airplane, then you could do it in 10 minutes with my method.

And over the course of a few years. Jason’s results gained more and more traction until he was invited to appear on TV to test his boarding method, which had come to be known as the “Steffen method,” against the airline with the best boarding times: Southwest.

Pitting my method against Southwest Airlines was really kind of the two best options in the industry, in terms of what are the fastest things that are on the market at the moment. And it ended up that my method was about 30% faster than Southwest Airlines, as well.

So why don’t we see the Steffen method in use now? After all, he first published his findings back in 2008.

In order to get my method to work, you have to have pretty strict controls over where people stand in line, and who they stand with and what’s between them and the next person where, you know, getting real people to behave that way, it’s like herding cats. And so if you had the military and you needed to board people on an airplane and you had the military’s ability to control where people stand and how they act, then my method would undoubtedly be, what would be adopted if that’s what they cared about. So, you know, could mine be adapted for the industry? Yeah, I don’t have any doubt that it could be. But, you know, that’s up for the executives to decide, what’s the best way to incorporate this while still balance all of the competing demands for cargo space and where passengers want to sit and how much they paid, and if they have gold, platinum, elite status, or if they’re just in the lead paint class like I am. These are things that are above my pay grade to determine.

So there you have it, folks. When it comes to air travel, it seems like the lines are here to stay. Coming up after the break, this probably won’t surprise you, but the transportation sector in America is a real energy hog. So I’ll talk with a biological oceanographer about something that could totally revolutionize the environmental impact of travel. That’s coming up right after the break.

Hey, y’all, welcome back. So we’ve been talking a lot about travel. Indeed, we’ve hit planes, trains and automobiles. Great movie, by the way. But there’s an important side of transportation that we haven’t talked about yet: it’s environmental impact. According to the EPA, transportation accounted for 27% of U.S. greenhouse gas emissions in 2020. And, quote, “over 90% of the fuel used for transportation is petroleum based.” I’m sure I don’t need to remind you, but there is a finite amount of petroleum on our planet. So it’s no surprise that scientists have been researching alternative fuel sources. Surely you’ve heard about corn being used to produce ethanol or soybeans and diesel. But there’s a new kid on the biofuel block these days, one that may change the future of travel for good.

Since 2017, I’ve been studying giant kelp.

This is Diane Kim. She’s a senior scientist at the USC Wrigley Institute for Environmental Studies. And yes, you heard her right: giant kelp, which for a variety of reasons may be the fuel source of the future. We’re talking once again about planes, trains and automobiles. Even good ol 18 wheelers.

We think there’s a lot of potential in giant kelp because, one, there are very few resources, natural resources that go into growing kelp. Right? You don’t need land. You don’t need fertilizer. You don’t need fresh water, which is becoming increasingly scarce. But kelp is also one of the fastest growing organisms on the planet, so akin to bamboo on land. Kelp can grow extremely fast under ideal conditions. You know, you’re talking about growth rates of over one foot per day. And so you can generate a ton of biomass which you need to convert into bioenergy. You’re talking about a more carbon neutral source of energy. The big objective of this project and many others that are focused on giant kelp for biofuel, is to move away from fossil fuel sources because it’s much more carbon neutral. So you’re talking about creating an oil from giant kelp that you do burn, and so you’re releasing carbon into the atmosphere. But that carbon is, you know, comparable to the carbon that the kelp are sequestering to build more biomass in the first place. So it’s much more circular, regenerative, right? And you’re not talking about burning something that took tens to hundreds of millions of years to form.

So let’s break it down. How do you get from kelp to airplane fuel? What is the equation? What’s happening there?

Take this with a grain of salt, because I’m not a chemical engineer. But the nice thing about the process that our collaborators at the Pacific Northwest National Laboratory are focused on, is something called hydrothermal liquefaction. Yes, I know. I made that same face when I first heard that term. But it’s basically a pressure cooker, high pressure, high temperature. You’re squeezing the oil out of this biomass. And what’s nice about this process is you don’t have to dry the kelp. So you can put in wet biomass into this reactor and you generate a few different products. But most of the carbon turns into a bio crude that you can feed into existing refineries. So you don’t need to create a whole new infrastructure for this to create drop in fuels, including jet fuel.

Which honestly sounds pretty good. Except there’s one problem. How do we produce enough kelp for it to become a feasible source of biofuel? Well, that’s where Diane comes in with something she’s working on. It’s as simple as an elevator, but for kelp. Or a kelp elevator.

Yes, there is a lot of kelp. But even if we harvested all the help that exists naturally, which we would never do, we don’t want to do that, there still would never be enough kelp that grows naturally. And so we have to create the infrastructure to farm kelp, essentially, like we farm corn. And one way that we see this being feasible is, is to do it out in the open ocean where you have plenty of space. The problem there, though, is you’re talking about a really deep environment and you’re talking about an environment where usually you have very, very low nutrients at the surface. You actually have a lot more nutrients as you get deeper in the ocean. But the problem is, of course, the deeper you get, the less sunlight you have. And in fact, most of the ocean is a dark environment. So you have sunlight at the top. You have nutrients at the bottom. How do you get the nutrients and the sunlight to the kelp all year long? And so we designed this concept that would depth-cycle them. So taking them down to nutrient-rich deep depths and bathing them, right, in nutrients so that they can soak it up, and then bringing them back up to where the sunlight is so that they can photosynthesize. Hence the design and development of the kelp elevator, which is meant to be a prototype to test that concept.

So I must admit, when I hear all of this, it almost sounds like something out of I want to say.

Almost like The Jetsons. Like I don’t you know, it sounds almost out of this world in some sense. I get exactly what you’re talking about.

First off, is it true, you’re saying we’re not sure if this will actually work? We’re still in that preliminary stage, right?

We actually completed a 100-day experiment. So under depth-cycling conditions, we actually know that kelp will not only survive, but they will grow, they will thrive. It’s really more about optimization, more than proof of concept at this point.

How much could we harvest from a kelp farm at a time?

I mean, it depends on how big the farm is, but what we’re aiming for is, you know, we did some back of the envelope calculations and we’re talking about kelp farms, you know, that when put together, so I’m not talking about a single farm, but many farms when put together, if you can imagine the size of the state of Utah?

It’s pretty big. But when you look at just the Pacific Ocean, you know, I think you can fit like over 700 Utahs in there. But if you can grow ten Utah’s worth of kelp in the open ocean, then, boy, I can’t remember the exact percentage. I think it can help supply or provide energy for the transportation sector.

That’s not bad. And that’s the transportation sector in the United States or the transportation in the world?

Just in the United States. So about a third of our energy consumption in the United States. So, yeah, you would need a lot of kelp and you would take up a lot of ocean space. But there is a lot of open ocean space. You know, when you compare it to fossil fuels, I mean, it’s a much better alternative. But not saying that kelp biofuel will be the panacea or anything. Right? It’s going to be part of the portfolio of solutions that help us get to a more sustainable planet.

So there you have it, folks. It won’t happen immediately. In fact, it may not happen for a while, but some years down the line when hopefully we’ve also figured out how to board planes faster, your flight might be powered by kelp. What a world, huh? When I was a kid, I usually hated it when my family went on vacation. Looking back, I think it was mostly because I hated the traveling part of it. What kid wants to spend hours getting to the airport, getting through security, and then having to sit still on a plane for even more hours? When it came to car vacations. I certainly didn’t like spending hours having to deal with a scrunched back seat. That’s why this episode hit close to home. I really wanted to find out what are the best ways we can get around and have fun doing so. It turns out there are a few different ideas where you can get the biggest bang for your buck. Maybe the answer for some of you is to take the train. The rail system in this country is far from perfect, but there’s enough of it so that those of you looking to get away should at least try it. Maybe you want to stick with the car. I will admit sitting in the front seat has helped me enjoy it more as an adult. If so, maybe you’ll plug your road trip into the algorithm to get the most out of it. And for the love of God, can some airline try out the Steffen method for boarding, please? I know I wouldn’t be the only one to sign up for an airline that promised shorter boarding times. The truth is that there’s no perfect way to travel. My hope, though, is that this episode will have inspired some of you to get out and enjoy our great country. Now, I’m off for a bit. Turns out this has been a season of love for me, and she is forcing me to go on a vacation for the first time in ten years. I am upbeat about that — and about seeing you around the bend. Margins of Error is a production of CNN Audio and Western Sound. Our showrunner is Cameron Kell. Our producer is Savannah Wright. Production assistance and fact checking by Nicole McNulty. Misha Stanton is our mix engineer. Additional support from Tameeka Ballance-Kolasny, Dan Dzula, Allison Park and Alex McCall. Our executive producers are Ben Adair and Megan Marcus. And me? Well, I’m Harry Enten.