Published on February 16th, 2020 | by Michael Barnard0
Tesla Semis Are Cheaper Than Rail Enough Of The Time To Reshape Ground Freight
February 16th, 2020 by Michael Barnard
Ground freight choices will shift with the Tesla Semi, and rail will be one of the losers
Later this year, the first Tesla Semis will be rolling out of a gigafactory. As with many of Musk’s plans, schedule is the least firm thing about them, as the trucks were originally going to be rolling in 2019. But the question that this article deals with is: “What dynamics will shift in the rail vs. road shipping equation?”
To be clear, the Tesla Semi isn’t a “rail killer.” That would be silly. But it will shift the distribution of what mode of transportation gets chosen enough to make a dent. And it’s not alone. Electric freight trucks from other vendors will be coming to Interstates and the Autobahn in the coming years as well, and with many of the same attributes.
Some bona fides: In addition to regularly publishing assessments of the transformation of transportation, including Tesla’s key role, on CleanTechnica and other sites, I did a two-day freight rail operations course for vendors to Canadian National Railway a few years ago, have clambered over a diesel electric freight engine, and have shaken hands with conductors and engineers. I helped build a North American asset mapping and KPI tracking system for them. I even published an article on innovation in freight rail in an industry-oriented publication a few years ago: “Working on the railroad — Inside Logistics.”
For every logistical assessment of shipping, the questions arise: Which is cheaper? Which is faster? Which is more certain to hit delivery dates? The combination means that Tesla Semis will be favored over rail immediately upon availability in many circumstances, and more so in coming years with obvious and known autonomous capabilities.
With modern transshipment of containers, the fewer touch points at transshipment terminals, the lower the cost. It’s about $100 per lift of a container, so if it gets lifted off four fewer times because there’s no train in the middle, then you can save $400 bucks per container delivery. Putting a container on a truck at a port and having that delivered directly to a client saves some money that way, but right now the miles per ton delivered is higher per truck with diesel tractors, so it gets done a bit less often than it will when economics change.
The lower operating costs of Tesla Semis due to less maintenance, the much harder to break glass and the lower cost of fuel means that the marginal costs of operation are lower than for diesels. That means that they will break even more often in the truck vs rail question, which means more loads on roads. Most of those were covered in Musks’ presentation, but now that we’re closer to actual delivery, it’s worth looking at more aspects.
The lower breakage of glass is important, too. In at least North America (I haven’t looked at European or Asian regulations on this), a cracked windscreen means that the truck is on the side of the road and not operating until a replacement windscreen is in place. That can put a big dent in a delivery schedule, something less important with a single container than a fleet of containers moving, but it increases the certainty of delivering on time, which has an economic value.
But that’s just out-of-the-box savings, which will shift the distribution to favor trucks a bit more often. Up to mile-long (1.6 km) trains have only two staff members on board, a conductor and an engineer. That’s a key point, and it’s where autonomous driving and trucks will be making more points in favor of Tesla Semis over freight rail more of the time.
Tesla, along with Daimler and others, have already demonstrated on-road autonomous convoying of freight trucks. Multiple trucks act as a single truck, communicating with each other not with brake lights but via internet-enabled messaging. The same autonomous features that work so well in Tesla’s cars work just as well in Tesla Semis, but with the added capability of convoy mode.
Why is this important? Two ways. You can start to see that with the mesh of roads being much broader than the mesh of rails and with no physical connection between trucks, they can self-assemble into convoys for subsets of the journeys, especially for longer hauls where rail has advantages today. This isn’t dissimilar to patterns of two to five 18-wheelers on highways running grill to bumper today, and convoying creates a couple of cost savings.
The first is just physics. Drafting saves gas, having the first truck in the convoy push the air out of the way and having the rest in its slipstream saving money on fuel for all of them. Autonomous vehicles will do it better and be able to achieve better savings than human drivers can. With the ability to have individual “cars” in the train pull out of the train and deliver off of an offramp along the way, you get some of the cost savings of freight rail without the rails, which are expensive. This has been projected for years, but at least one major vendor has backed away from it. Daimler has said that its pilots haven’t shown significant enough savings to make it worthwhile. If physics were the only advantage, then this wouldn’t be a strong enough statement, but it’s on top of the out-of-the-box value proposition of electric semi tractors and is followed by other advantages.
And the next advantage is crew costs. Right now trucks all require drivers, and drivers have limits on how many hours that they can drive in a day and week.
“You may be expected to work up to 70 hours over an eight-day period. After you’ve worked for 70 hours, you cannot drive again until you take a full 34 hours off duty. The 70-hour limit could be reached by working 14-hour days, but you cannot drive for more than 11 hours in a day.”
What does this mean in the context of autonomous convoys in short and interim durations? In the short term, not much. Initially, drivers will just be better rested, which in addition to autonomous features will lead to fewer collisions per million miles. But you can see regulations shifting to allow convoys where drivers can book off and get a couple of hours of sleep if they are third in a row of five trucks fairly quickly.
In the medium term, you can see drivers delivering trucks to a point where they can pick up with a convoy, a big parking lot not a freight rail transshipment point, and then head home while the truck convoy drives a thousand kilometers to another big parking lot with additional local drivers.
And when fully autonomous trucks are allowed on the road, you can see individual trucks peeling off of convoys without any drivers, while the convoy might have a conductor and an engineer.
The next point is an interesting one. I couldn’t persuade Canadian National Railway to be interested in an approach to putting air carbon capture on its freight trains using Global Thermostat’s technology. It would have been cheap air carbon capture because it was just adding a car to an existing train, using the movement of the train to funnel air into the sorbents, using waste heat from dynamic braking to power the process of removing the CO2 from the sorbents, then having the captured CO2 being on a low-carbon distribution network right off the bat. Why wasn’t Canadian National Railway interested? Various reasons, but one was that it was already the lowest-CO2-emissions form of land transportation available except for pipelines. But with convoys of Tesla Semis running on renewable electricity and the increasing likelihood of carbon pricing on diesel, the rail majors will have to adapt to being behind freight trucking in that measure, and having additional costs related to their diesel emissions.
The last point is a conceptual one that I use as part of a framework to assess what technologies are likely to win in the long term. The concept I use is loose coupling, something from software systems architecture. Tightly coupled solutions, like wine bottles with corks and corkscrews, or freight rail trains and rail tracks, are less flexible, requiring both components to be in good shape, available, and working well in order to deliver the value. More loosely coupled systems, like wine bottles with screw tops or trucks running on a broad network of big and small roads, have advantages. Tightly coupled solutions have inherent limitations that inhibit their competitiveness in some ways while enhancing it in others, but innovation in loosely coupled systems tends to win in the long run. That’s why the lack of physical connections between conveying trucks over a network of roads, with self-assembling road trains, is such a powerful innovative and competitive force. It plays directly into that premise.
And that loose coupling is inherent to the fundamental innovation of transshipment containers in any event. Standardized containers that can be stacked on ships, trains, or trucks have transformed transportation already.
Imagine a road train of ten independent trucks with only two people on board, no emissions, low maintenance, and low fuel costs. That’s the future of road freight. That’s balanced by still having more staff than trains and limits on load weights, but the distribution will have changed substantially.
In the future, a lot more freight tonnage will end up on roads in Tesla Semis and similar vehicles from Daimler and other vendors. It’s the nature of the beast.
Freight rail has to innovate more to deal with this competitive threat. In Europe, they have already shifted powering freight trains to external power lines, running the electric traction engines not off of electricity from the diesel generators but from remote wind and solar farms. That’s increasingly going to be done in the rest of the world. In China, high-speed rail is all electric with externally supplied electricity. This isn’t hard, it’s just not the freight rail standard yet. The engines are all diesel electric hybrids as it is, so displacing the diesel part is relatively straightforward. And freight rail companies are experimenting with adding batteries as well, not only charging them at terminals to supplement diesel, but also filling them en route with dynamic braking being turned into regenerative braking and with externally supplied electricity when available.
All transportation is electrifying. All transportation is moving to lower labor costs. All transportation is becoming more loosely coupled. All transportation is moving to lower carbon models. But the Tesla Semi is going to have a substantial advantage once it’s on the roads, and for a decade after that. No other manufacturer has anywhere near the experience with electrification, en route provision of high-speed charging and autonomy.
Tesla’s Semi will be very profitable for Tesla. And for the people who buy the electric, high-tech semi trucks.
Have a tip for CleanTechnica, want to advertise, or want to suggest a guest for our CleanTech Talk podcast? Contact us here.