News   Apr 26, 2024
 94     0 
News   Apr 25, 2024
 410     0 
News   Apr 25, 2024
 1.2K     4 

The Coming Disruption of Transport

Would you buy an EV from a Chinese OEM?

  • Yes

    Votes: 10 11.5%
  • No

    Votes: 61 70.1%
  • Maybe

    Votes: 16 18.4%

  • Total voters
    87

Highlight of issues with electric buses that likely spill over to electric trucks
 
Road capacity will remain a challenge, particularly because AEVs will result in a big increase in VMTs. They can also play a big role in making transit more relevant to more users (minibuses).
Pedestrians will still need to cross the road. Which means lights will have to change cycles so capacity would be hampered as roadway volume increases. Any form of mobility has a maximum theoretical capacity. We will still be asking the same questions, and having the same concerns about roadway volumes, despite how safe AVs will be.
 
Pedestrians will still need to cross the road. Which means lights will have to change cycles so capacity would be hampered as roadway volume increases. Any form of mobility has a maximum theoretical capacity. We will still be asking the same questions, and having the same concerns about roadway volumes, despite how safe AVs will be.
Yes, I think a lot of techno-optimists forget about pedestrians, bicycles, pets/wildlife when they come up with their optimistic visions of massively increasing capacity of existing roads. There is some room for improving overall traffic by optimizing for collective flow rather than each driver greedily optimizing their travel speed and timing intersections, etc. But the idea of weaving intersections is crazy. Pedestrians still need to cross the street.

 
Which issues specifically? Charging?
I thought the technology was further ahead as there are many fleets purchasing electric buses around the world. The TTCs test of eBuses was disappointing in that regard
Issues noted:
-short range
-supply chain / spare parts / post sales support
-mean time before defects (notably worse than hybrids)
 
Last edited:
I thought the technology was further ahead as there are many fleets purchasing electric buses around the world. The TTCs test of eBuses was disappointing in that regard
Issues noted:
-short range
-supply chain / spare parts / post sales support
-mean time before defects (higher than hybrids)
There are lots in China, but their standards are maybe not as high.

I see range as less of a problem than the ability to charge quickly between rushes. You just need a network of satellite charging locations for buses to be taken out of service for an hour or so. Near terminals would make sense.

It's a shame Tesla can't do everything. I'm sure they would make a pretty good bus, though the durability challenge buses represent is not something Tesla has really tackled yet. I think they could do very well with powertrain.
 
There are lots in China, but their standards are maybe not as high.

I see range as less of a problem than the ability to charge quickly between rushes. You just need a network of satellite charging locations for buses to be taken out of service for an hour or so. Near terminals would make sense.

It's a shame Tesla can't do everything. I'm sure they would make a pretty good bus, though the durability challenge buses represent is not something Tesla has really tackled yet. I think they could do very well with powertrain.
The powertrain development out of FormulaE may actually end up being quite good for buses. The stop-start, high power requirements of buses have different design requirements than a standard production car. Electric powertrain development is going to continue to move at a rapid pace over the next 10 years.
 
There are lots in China, but their standards are maybe not as high.

I see range as less of a problem than the ability to charge quickly between rushes. You just need a network of satellite charging locations for buses to be taken out of service for an hour or so. Near terminals would make sense.

It's a shame Tesla can't do everything. I'm sure they would make a pretty good bus, though the durability challenge buses represent is not something Tesla has really tackled yet. I think they could do very well with powertrain.
It was noted that because of the difference in available operating time between battery vs hybrid due to charging times would cause a need for more ebuses.

That's all at today's tech of course. Things will get better. I didn't realize how much room for improvement there was.
 
It was noted that because of the difference in available operating time between battery vs hybrid due to charging times would cause a need for more ebuses.

That's all at today's tech of course. Things will get better. I didn't realize how much room for improvement there was.
It's very similar for last mile delivery vans. The cost, charging speed, range is pretty disheartening compared to where passenger EVs are today.
 
Potential for what? Any throughput that a minibus vehicle can achieve is no better than the equivalent by a longer vehicle at less frequency? What benefit does reducing the vehicle size to vastly smaller pods have in increasing throughput? The only reason he is using smaller "vehicles" is because it directly ties with being able to sell individual vehicles to the general population. He has no desire to build public transportation systems, as his desire is to sell you a car. It's not like he's genuinely interested in improving signalling technology for trains or for reducing subway tunneling costs either...

The Loop design from Boring company isn't designed to replace subway lines - it's designed to be a more intermediate capacity solution with great last mile potential as vehicles theoretically wouldn't be limited to the tunnel itself, or at least it's low station costs would allow for much larger amounts of them.

A 16 passenger mini bus, running at theoretical 10 second frequencies (fairly realistic I think, notably Boring Company claims they are aiming for 5 second frequencies), can run about 6,000 pphd. That's nothing compared to a subway, which can typically handle about 30,000.

The innovation isn't there though - it's the cost. A 6,000 pphd service using loop can be constructed for almost nothing. The Las Vegas loop was built for $52 million for 2.7km of service - that's $19 million per km. And Boring Company is trying to push costs far lower than that as well.

So lets say Boring Co manages to achieve 10 second frequencies, half the target they are aiming for. The service is provided as a mix of mini-bus vehicles and private vehicles which pay a (likely quite high and profitable) toll to use the facility. Let's say 50-50 split. Assume private vehicles have an average occupancy of 1.3 people... that's 234 high toll customers an hour, and the mini-buses provide an additional throughput of 2,880 people per hour. So about 3,000 pphd overall.

Even if costs remain the same as today (unlikely), with signaling systems and other additional requirements offsetting any savings coming from their efforts to lower tunneling costs. We are looking at about $20 million / km, or $6,666 per pphd/km.

A subway provides about 10x the capacity for $500m/km, or $16,666 per pphd/km. The Boring co provides a much higher quality, faster, more direct service, for about 1/3 the cost. And this is presuming relatively conservative assumptions - a lower frequency than they are aiming for, a 50/50 mix of buses and private vehicles, etc.

It's never going to replace dense inner city transportation as it still just requires too much space, but that's not what it's trying to do.

But it's got great potential for intermediate capacity routes, in medium density environments, where demand can easily be served by 3,000-5,000 pphd services. Especially if you can build a more effective network to spread that out, or create double tunnels in busier stretches.

Lets test this in a theoretical situation and see how it plays out financially.

Lets say That Boring Co makes good progress and in 2030 is looking to enter the GTA market with it's first major project - an express tunnel from Toronto to Mississauga City Centre. This is a busy travel corridor that has no real plans for a proper mass transit connection, and has likely demand that is high, but can be accommodated with the capacity of the loop system.

Boring Co proposes a tunnel starting somewhere downtown. It follows the right of way of an existing street, well below grade, and travels under the lake over to Hurontario Street, before travelling under Hurontario to MCC. An intermediate stop is proposed at Port Credit.

That works out to about 28km of tunnels. Assuming $20m/km USD - achieved in Vegas without any additional advancements in tunneling or other construction costs since then, the project would cost about $450 million CAD.

Lets assume peak travel is at capacity of 3,000 people in the peak direction, and that total daily ridership is about 10x that, as is typical for urban transportation systems, so a total of 30,000 daily trips. Lets assume it's an all mini-bus service, and no teslas are allowed. and for the heck of it, lets say it had major cost overruns and actually cost $800 million to build.

financing costs will be about $3.5m/yr assuming 3.5% financing and 30 year period, and lets double it for operating & maintenance expenses. So profitability is anything above $7 million annual revenues. Assume 20% margins, we have an aim of about $8.5 million a year in revenue.

Then lets say that it gets half the ridership it expects, at 15,000 passengers a day on average. At an average fare of $3, is about $16 million a year. double the profitability margin.

The service has the potential to be transformational if they can keep costs contained and develop significant frequencies. We just have to see what happens.

And I haven't even discussed time savings. Boring co is already getting vehicles up to about 200km/h in their test tunnels, even without hyperloop. If they can reliably do that safely, the trip from Mississauga CC to Downtown toronto could be done about 10 minutes.

Even if they achieve a system that is half the capacity, double the cost, and half as fast as they are aiming for, it has a lot of potential to be a transformational urban transport system.
 
Last edited:
I think the idea of getting vehicles up to 200+ kph in a tunnel is possible, but unnecessarily risky. Even 100 kph is transformational in transit times. No other travel method could match a 20 minute trip MCC to downtown Toronto short of a helicopter/VTOL air taxi. I think the biggest problem is that they would have to charge a lot to manage demand, which would be quite high as it would be very attractive even to die-hard drivers who would make the same trip in 50-60 minutes today.

I think the way TBC is planning to operate the switching and stations is to have vehicles enter and leave stations in pulses/platoons. You would have queuing area before entering the station.

As I understand it, the process would look like:
  1. Passengers are assigned to berths based on destination, where a vehicle is awaiting or soon to arrive
  2. Passengers board vehicles
  3. Any vehicles ready to depart do so at the same time. Any vehicles where passengers are still boarding wait until the next departure window.
  4. Arriving vehicles queue just before the station. Vehicles in the queue advance to the berths freed up in the previous step following their departure.
  5. Passengers exit the arriving vehicle.
If the stations operated in pulses like this, I think those pulses could be recurring every 90-120 seconds or so. Not every vehicle can arrive, disgorge passengers, have new passengers board every cycle, but the system would be tolerant of that and not hold up the other vehicles. A 12 berth station with 16 passenger vehicles would have a capacity for 5760 departures per hour if each berth is turned over every 120s.

If vehicles are all headed in the same direction, they can travel as a platoon fairly safely (with reasonable follow distances 3-5s), leaving larger separations between platoons diverting or merging.
 

Tl;dw
18:50 - first comments autonomous cars, connected AVs are a huge security risk

21:30 - SAE Level numbers suck, should be names.

25:00 - Tesla is not Waymo Jr.

25:30 - Truck driving has high turnover, no kid dreams of driving for Uber. These are good jobs for robots to take over.

29:00 - Robotaxis will spread gradually, unless Tesla pulls off a miracle. Highway autonomy is easy. Pushes back on hosts who are really stuck on Boston-LA being some kind of big milestone.

34:00 - Companies are ready to do ramp-to-ramp driverless highway trucking very soon (if they can get regulatory approval). In early days Waymo considered ramp-to-ramp for personal cars. Honda Legend kind of does it now.

36:30 - 25% of energy for transport. We spend 50 billion hours/year turning steering wheels vs. 240 billion working.

37:45 - "This decade" we'll see Robotaxis spreading to many cities. Prime territories will get it in 5 years, some places not for decades.

39:30 - Starship autonomous robots have delivered a million pizzas/groceries/etc.

40:15 - No more AV stuff. EFF/digital rights (Java lawsuit, social media, etc.). Brad's new morality framework -- it's wrong to exploit bugs in human psychology. Etc.

I find the part about autonomous trucking interesting. I think people are overselling how long it will take to get autonomous trucking going especially since autonomous driving for highways is trivial.

I remember seeing on this thread that CN invested in an autonomous trucking developer and plans to abandon rail as their core business to become an autonomous trucking company. It will be interesting to see where that goes, but I have read many reports that say autonomous trucks will have a devestating impact on rail and will even be cheap enough to pull freight off of inland waterways.

Regardless though, I think a child born today won't even be able to recall that trucks on highways were driven by humans.
 

Back
Top