Dan Sperling on automation: “It completely changes the trucking industry's business model”
Dan Sperling on automation: “It completely changes the trucking industry's business model”
In the early nineties, California set up a plan to drastically cut smog as it tried to comply with a federal guideline: the state would reach over one million electric vehicles by 2010. The journey was far more tortuous than anticipated, but decades later a series of incentives have finally paid off. California has the most EV sales of all 50 US states and is leading the country when it comes to legislation to accelerate the transition to electric heavy-duty transport.
Behind the scenes, acclaimed professors have been working to provide society with research to support sustainable transportation. One of them is University of California Professor Daniel Sperling, who founded the UC Davis Institute of Transportation Studies in 1991 and has, since then, worked closely with governments, industries and the environmental community.
With several awards for his research efforts, Professor Sperling has been focusing in the last few years on autonomous, electric and shared transportation as the sustainable mobility solution.
In the interview below, Daniel Sperling discusses the importance of supporting innovation and why he compares autonomous transportation to a true industrial revolution.
Professor, scientists have been talking about climate change for almost a century. You have been studying electric vehicles for over 30 years. Now, we seem to finally be on track to sharply reduce greenhouse gas emissions, particularly in transportation. What made it possible for it to happen at this time?
Daniel Sperling: For the transportation sector, it's mostly about battery costs coming way down. They are 10% of what they were 15 years ago. And they are more energy-dense, so you get a longer range.
Today, we see countries like Norway, for instance, with 80% of new cars hitting the roads being electric. What is the impact passenger cars have on transport electrification more broadly?
The most important way is scaling up car and light-duty vehicle production, and thereby reducing costs. The impacts more broadly are large reductions in local pollution and greenhouse gas emissions. The downside is greater stress on the already-stressed electricity grid, which is under a lot of pressure to convert to renewable energy.
There is a more fundamental challenge as well: creating a new mindset and new business model for electric utilities. In most of the more industrialized countries, electric utilities have been focusing on energy efficiency for many years and not growth in load. Now, we're asking them to greatly increase their load, not just for light-duty vehicles, but for trucks and buildings. You need substations and transformers. And you need the utilities to be more forward-looking. They've been operating in a steady state mode; the regulatory processes and institutional structures are not in place to build more capacity. The short story is that utilities are lagging in anticipating the needs for vehicles and buildings. One outcome is that utilities are not prepared to build the charging infrastructure for trucks. Trucks need massive charging stations with massive capacity—similar to the demand of a small city.
So we have to change the rules so that utilities can be more forward-looking and prepared to build not only the charging stations but all the equipment to make that possible. They have to build the transmission lines and need to put in place a lot of storage capability, whether it's batteries or pump storage. All of that takes money, time and initiative. In most places, it's not happening fast enough.
When you say forward-looking, what do you mean by it?
The way it happens in California is the utilities generate a forecast for future demand. So, the forecasts are based upon data that's already a couple of years old. The forecasting exercise takes another two years, so now you have a four or five-year lag time. Then there is another couple of years' review by the regulatory body, our Public Utilities Commission. In the past, this long time lag wasn’t an issue because electricity loads didn’t change much, and we were in the mode of trying to incentivize efficiency and demand reductions. Now we’re in a new world: we want more generating capacity. If you compare 2018 to 2023, it was an entirely different world. We had no idea that we were going to be ramping up the number of electric vehicles, cars, trucks and buses at such a fast rate. All of their investments are based upon the approval of their forecasts, which are now way below the actual demand. The regulatory process has such long lag times that they limit the utilities from anticipating electricity growth.
You mentioned learnings before. What can the heavy-duty industry learn from the transition to electric passenger cars?
People who buy new vehicles tend to be more affluent – in the US, about a third of the households buy almost all of the new vehicles. They tend to have their own homes and be able to charge there. So, public charging was more of a psychological issue than a real necessity for many. But with trucking, it is a necessity. Trucks need public charging and large charging capacity at depots. Many trucks come back to their central location, where they need a dramatic expansion of charging capacity And public charging stations for trucks need large amounts of space, because charging is much slower than diesel refueling, and large capacity. Charging is a very different challenge than with passenger cars.
Drivers – both of cars and trucks – almost always prefer the electric. I've seen some delivery trucks and asked the drivers about it. They just love the electric version. In fact, there's competition in their company for who can get to be the driver for the electric trucks. Because it's quieter, smoother, they just really appreciate and enjoy it.
For the drivers of urban-based trucks, where charging is mostly done overnight at depots, it's going to be a very attractive switch. For those that need en-route charging, they do have an extra challenge: where and when they’ll charge. They need public charging infrastructure.
You conducted a study at UC Davis in 1992 about what it would take to reach the goal of 1 million electric cars by 2010. During the research, range anxiety was one of the issues raised among consumers, but their perception changed when you asked the group to keep a diary of the miles they drove every week. So, one of the main takeaways then was public education. What goes on with our minds when faced with these new technologies?
First, I want to say that we've transitioned from range anxiety to charge anxiety. Electric vehicles have a much longer range than before. For the most part, range anxiety has dissipated. Now, they have charging anxiety because the network here and in Europe is not very reliable except for the Tesla superchargers. They've done a great job of building out a network, but if you don’t have a Tesla, then you are likely to have anxiety. You might have noticed just recently, at least in North America, that many of the car companies have made an agreement with Tesla so their vehicles can be charged at their superchargers.
Behavior and attitude vary greatly over time and space. If you're just driving a car around your community, you don't care about the chargers; you just charge at home. We need to know much more about behavior and attitudes than we do now. And we need to deal with it. We've got to make sure reliable public charging is available. We need to have incentives for some of the early adopters.
The situation with heavy-duty trucks is probably four or five years behind. The difference with trucks is that they're used in so many different ways with so many different configurations that they have different needs than with light duty.
California now has a requirement in place that by 2036, all new truck sales must be zero-emissions. Some truck owners need to move faster. For example, drayage trucks (those serving ports) have an accelerated schedule, as do local and state government fleets. Europe is moving in this direction but hasn't adopted the same aggressive requirements for trucks yet.
Again, the biggest barrier is the infrastructure. It's not just public charging, but large private truck depots.
Your book “Three Revolutions: Steering Automated, Shared and Electric Vehicles to a Better Future”, published in 2018, discusses dream and nightmare scenarios regarding autonomous mobility. How do you analyze the advancements in autonomous transportation in the last few years?
Let's talk about light duty first. If light-duty vehicle passenger cars that are manually operated now are simply swapped out for an automated personal vehicle, we're going to see their mileage probably double. You could commute three hours a day and do all kinds of things in your car, which we wouldn't do now. Imagine if you could eat, sleep, text, watch a movie, work in your car? In this so-called nightmare scenario I just painted, there's much more mileage, and that means more energy use and more traffic.
This scenario of personally owned vehicles is not all bad. We find that for elderly people who are apprehensive about driving at night or any time, it greatly improves their mobility and accessibility. And then there are people with physical limitations. But this scenario means an even more car-centric society.
The real answer is that if we're going to have personal automated light-duty vehicles, they should mostly be shared vehicles, where there's pooling and they're operated commercially. In this case, people really can give up their personal cars because they're getting this high-quality service. It's good because it reduces the cost of travel for people. It's good environmentally because you have less mileage. And it's good from an equity perspective because now we can provide mobility to a lot more people at a much lower cost than we do now—though this is more true in the US than in Europe where there is already better public transportation. The cost per passenger with shared AVs--assuming multiple riders per vehicle—is much less than for an Uber ride and even less than transit in most situations outside dense cities.
The truck story is different. Until very recently, most experts were saying that trucks were going to be the first major application of automation because there's an economic motivation. If the driver accounts for a large percentage of the total cost of providing truck services, automation greatly reduces the costs. There's a great business model. But it's been a lot harder than we thought. There are some applications that I think will happen in the not-so-far-off future. But it'll probably be limited for quite a while. There will be automated trucks on major limited-access highways, with platoons of these trucks and a special lane devoted to them.
One reason for my cautiousness is just the optics of it, how people perceive it. You're driving along a highway, then you look up, and there's this huge truck right next to you with no one in the cab. It's going to be a while before people are comfortable sharing a road with a driverless truck.
It goes back to the public education we mentioned in the 1992 research, right? What message do we have to convey to society to show that, in reality, it is actually safer?
Let me take a little piece of that. One of the challenges is that the labor unions – in the US, the Teamsters union – are apprehensive, if not opposed to driverless trucks. A bill passed by the legislature in California bans driverless trucks unless they have a safety driver in it. But just some time ago, Cruise, which is the automated vehicle part of General Motors, came to an agreement with some unions in San Francisco. They said that as part of offering driverless cars as a robotaxi service, they would commit to hiring a large number of people who would work in all their facilities, as remote operators, managers, maintenance workers, and so on, convincing the unions that there would not be a loss in the total number of jobs. If you take away the concern about the loss of jobs, then in the policy arena it'll ease the introduction and use of driverless vehicles, both light-duty and trucks. Of course, it will still be disruptive to individuals, since those who lose their jobs are not necessarily the ones who gain the new jobs.
Public education is a tricky thing. At the end of the day, regulators and policymakers are responding to voters. With automated vehicles, it's an issue of how safe they have to be before the public accepts them. We have no way of accurately measuring it. Until we have massive numbers of these vehicles on the road with lots of experience, we won’t really know. And it's going to vary between different companies and different products. People want companies to be held to a higher standard than they hold themselves. That's a factor that has to be dealt with.
There’s also the liability issue and how that works. If a crash happens, is it the software at fault, the hardware, or is it the vehicle owner? Who's to blame? None of that has been worked out yet. It's all part of the public and the policy discussion. The transition, both for passenger and commercial vehicles, will be an incremental process, with constant feedback and testing. They are not going to flood the market all of a sudden as happened with scooters and Uber-type services. In this case, consumers will be more cautious, the service providers more concerned about liability and more sensitive to local concerns, and local governments will be more prepared (and usually more conservative).
And how do you balance that? Because you have to test the new technologies, but then you also have to allow them to be tested.
That's why it's going to be incremental. For instance, Cruise and Waymo, the Google company, were mostly operating in only two cities as of a year ago. Both of them are now expanding in the US to a lot of different cities. They’re introducing themselves to the public and gaining acceptance, but they're also still testing the technology and the consumer responses. I think they're largely comfortable that the technology works, though with the realization that many small fixes will continue to be needed—sometimes referred to as edge cases One glitch in San Francisco happened when the internet went down in one location, causing vehicles to stop where they were. The companies are using a geofencing approach, where they map the hell out of a geographic area and prepare the local population, gain experience and acceptance, and then gradually expand that geofenced area. For trucks, that area could be a highway corridor.
What's your take on AB 316, the California bill requiring a driver on board?
It’s political. The unions are doing it because they're trying to protect jobs. So, it's a pretty heated debate right now. I don't know what's going to happen. But it'll be very unfortunate, in my mind, if they do pass that law because it will greatly slow innovation*.
Truck and technology companies will just move to another area, state or country, where there's a more hospitable regulatory environment. It's kind of ironic because California is the place where many of these companies are headquartered, and a lot of the technology is being developed, including the AI. There is a movement at the federal level in the US to allow companies to operate a much larger fleet of automated vehicles. This is going to be an incremental process, and it's going to evolve differently in different areas.
*California Governor Gavin Newsom vetoed the bill last Friday, September 22.
In your “Three Revolutions” book, you compared vehicle automation to a true industrial revolution. What sets it apart from other innovations?
Automation really will be transformational, even more so than electric vehicles. Electric vehicles are a technology fix. You're mostly just changing the propulsion system. With automation, you’re creating the potential to really transform so many industries, businesses and activities. Instead of owning the vehicle, now you're being taxi-ed around all the time.
If vehicles are automated, it completely changes the trucking industry's business model because you dramatically reduce your driver cost, and you have other kinds of opportunities. As with smartphones, one can’t know how this might play out. It will be a revolution.
In some places where it happens faster, it will be disruptive. And it'll be disruptive even in the long term as well. The role of policy is to smooth those disruptions, everything from jobs to safety and how businesses operate, to ports and intermodalism. It boggles the mind what this could be like in the future.
This interview has been edited and condensed for clarity.