The future of the city and Robocar Oriented Development

It's been a while since I've done a major new article on long-term consequences of Robocars. For some time I've been puzzling over just how our urban spaces will change because of robocars. There are a lot of unanswered questions, and many things could go both ways. I have been calling for urban planners to start researching the consequences of robocars and modifying their own plans based on this.

While we don't know enough to be sure, there are some possible speculations about potential outcomes. In particular, I am interested in the future of the city and suburb as robocars make having ample parking less and less important. Today, city planners are very interested in high-density development around transit stops, known as "transit oriented development" or TOD I now forecast a different trend I will call ROD, or robocar oriented development.

Read the essay Robocar Oriented Development

For a view of how the future of the city might be quite interesting, in contrast to the WALL-E car-dominant vision we often see.

Earlier I wrote an essay on robocar changes affecting urban planning which outlined various changes and posed questions about what they meant. In this new essay, I propose answers for some of those questions. This is a somewhat optimistic essay, but I'm not saying this is a certain outcome by any means.

As always, while I do consult for Google's project, they don't pay me enough to be their spokesman. This long-term vision is a result of the external work found on this web site, and should not be taken to imply any plans for that project.


A variety of things that may not have immediately come to mind:

Downtowns become more attractive to visit via robocar, by both suburbanites and "uptowners," when you don't need to find and/or pay for parking.

A tourist who arrives via robocar to a downtown hotel can eliminate the daily parking fee needed for storing their car. A city center hotel might then increase in attractiveness over a hotel on the outskirts of a city that offers cheaper or free parking. (See: WorldCons that are in driving distance, and then having to pay expensive hotel parking over five days for a car that you might not use at all during that time.)

Newly developed city center hotels may have fewer city regulations regarding the amount of available parking they need to provide. This could result in cost savings, more hotel capacity, and/or other side effects that could result in lower room prices for customers.

A trip on current mass transit bears many similarities to a trip in a self-driving car: You don't need to pay attention to the road, you can read, work, have conversations with others, etc.

Some people don't find travel time in general to be very productive time, so being able to get work done while being driven or transported is not universal. There are also many arguments against doing or continuing work while on the way to or from your place of employment (decompression/transition, not being paid because you're not "on the clock," etc.).

Many advancements in self-driving car tech can be applied to mass transit, too. (Driverless busses and trains lower the cost of operation, no driver fatigue, etc. Plus: Drivers won't be reporting late to work or a shift change, or taking too long a break; stop requests will always and automatically be fulfilled; late night services will be more reliable; vehicles can communicate without being distracted from driving and ensure that they're spaced out evenly, etc.)

Since there's no driver taking a cut or paycheck, robotaxis and/or shared car services may wind up being cheaper than current human-driven taxis, but due to economies of scale robobusses are sure to still have a big cost advantage for riders over robocars.

There are going to remain large masses of people who are not affluent enough to afford robotaxis and/or shared car services.

There are no one-way streets in an area that is restricted to robocars. There isn't even necessarily a "correct" side of the road to drive on. Even a narrow alley that's only the width of a single car is a two-way street for robocars. For a short alley it becomes temporarily one-way in the direction of travel of the first robocar that enters it, until it's clear of vehicles again. Alternately, and/or for longer alleys, robocars can enter and travel in both directions simultaneously as long as there are cutouts that allow them to pass each other at certain points.

Areas that are robocar-only will not have stop lights. They won't have specific pedestrian crossings on smaller streets, either. They may, however, still have marked pedestrian crossings on higher-capacity arteries, to minimize vehicle flow disruption.

People don't walk only for transportation. There are many people who walk or jog on streets for exercise. Dogs need to be walked. People with small children often like to put the kid(s) in a stoller and just "go out for some fresh air." People often "go out for a walk" to calm down, or clear their minds, or just for "a change or scenery" from the inside of their home/office/whatever.

People don't always have a single specific destination that they want transportation to. Often you want to get to a general area and then walk around (especially for tourism and downtown shopping).

In many places the percentage of foot traffic from people walking by due to having to park away from their final destination is actually very low. The point-to-point abilities of robocars will have negligible effect, there.

Personally, if robocars are available but I'm going only up to four or five blocks I wouldn't even think of calling a robocar, unless I needed to carry something. In fact, if I knew robocars were available to get me back to my origin quickly and easily, I might not think twice about taking a much longer, one-way walk. It's quite likely that I might want to initially go on foot somewhere (good exercise/fresh air/scenery/etc.) to get something, and then use a robocar to get me and my cargo (like groceries) back.

As the prevalence of robocars goes up, walking from place to place alongside and across roads with vehicles will become safer.

City dwellers who start having children will have less pressure to move out and into suburbs whose streets are "safer" from traffic accidents.

Having children in a city currently generally means the children will be taking mass transit to school, which parents regard with suspicion in regards to safety. Robocars suitable for use by children can eliminate this perceived risk in a city.

High housing density in a city provides a large nearby potential pool for both good teachers and talented students. With robocars suitable for use by children it'd be as easy and safe to send kids to a high quality magnet school across town as to a "shopped for" school many miles from home in the suburbs. A benefit of this is that the kids are physically closer to home, which tends to be reassuring to parents.

Just as now, property developers will have great infuence on how land is apportioned in robocar oriented developments. Without plotspace taken up by garages and driveways, it's more likely that developers will favor putting more plots into the same total area (thus, more profit) than giving people the same size plots with larger yards and more space between houses. A plot in a robocar oriented development area that's the same size as a current plot with a garage and/or driveway on it will probably be a "larger," more luxurious, and thus more expensive plot.

When one or the other end of your journey is near a rapid transit stop, journey time on transit is frequently about the same over the same distance as using a car. (I just used Google Maps to compare mid-afternoon driving vs. transit on routes between the San Francisco Civic Center and no specific street address in each of San Bruno/Daly City/Oakland/Fremont/Walnut Creek. I'm not sure if Google factored in traffic, since it's late and I manually adjusted the time to mid-afternoon. Google may have based the calculations on all road travel being exactly at the speed limit.)

Rapid transit will almost invariably remain cheaper than a robotaxi (economies of scale again). First and last mile robotaxis with rapid transit in the middle is likely to be the cheapest point-to-point method of travel where rapid transit exists and aligns with a major portion of your trip. During rush hours it avoids bottlenecks and chokepoints on roads, so there's a high chance that this will win on journey time at those times.

Large park-and-ride lots at transit stations can shrink once there are a lot of robocars. Shops/offices/homes built on the land freed up will automatically become Transit Oriented Development, and most likely will be robocar oriented development, as well.

Yes, a number of things I have addressed in other essays.

Initially, parking will still cost plenty in CBDs. As robocars start parking outside the CBD to save money, the market will adjust prices to maintain an equilibrium, but soon lots will decide they can make more money as something else (or the city might buy them for parkland or push this by limiting rezoning.) While the cost of parking at CBD hotels is annoying, the reason they can charge so much is that it's not a big driver of the decision to stay there.

I actually think mass transit gets a big overhaul, perhaps even before the driverless bus (with the unions.) The robot van is probably the best vehicle for mass transit -- seats enough to be energy efficient, does not seat too many (which makes the bus inefficient) off-peak, is cheaper and smaller and thus is more frequent. But this is only for those who can't afford a robotaxi, or at rush hour.

I don't think that's going to be too many people though. I predict the cost of the robotaxi should be cheaper than the cost of today's cars, which is about 50 to 80 cents/mile. That may be too much for a long trip (like commuter rail) but is cheaper than both the real cost of many of today's transit trips and even cheaper than the subsidized cost. So while the transit gets cheaper I don't think it will be a lot cheaper -- only the very poor would find the savings compelling.

Robocar-only is quite some time away -- that generates a huge number of changes. In this article I wanted to paint the path to that with incremental changes. Even when it's 50-50 human vs. robocar, the parking lots start shrinking.

I do agree that the ability to do one-way trips strong encourages the use of other modes (like walking, biking, transit) because you don't have to get back to where you left your car. In fact, several years ago I even proposed the bikebot, a small robot that can deliver bicycles -- using the bike's wheels as part of the structure to be small and light. With the robot that delivers a bike, quick one-way bike trips become super convenient aned healthy.

The safety of streets from accidents is a factor but not a large one, I think, in the decision to suburbanize.

The plots without garages are nothing new, many cities were planned that way before cities starting insisting you have a parking space for every dwelling.

In less urban areas robot vans for transit might make more sense than in urban areas. In San Francisco very few bus lines don't see full busses at several points during the day. This applies not just to the busses that run between neighborhoods and downtown along Market street, but also the many north/south arterial lines: 19th Ave in the Sunset, Divisadero and Castro Sts, Fillmore St, Van Ness Ave., the 43 between the Presideo and the south edge of SF (through my neighborhood in the Haight), and north from the Caltrain station.

It might seem to make sense in urban areas from one standpoint to introduce airport shuttle-style vans to some lower ridership "neighborhood access" routes. However, from another standpoint those vehicles can't be "load balanced" if necessary onto the other routes that make up the vast majority of the overall vehicle trips. Plus they can't be easily swapped due to maintenence considerations. I'm sure the flexibility of keeping the entire vehicle fleet as full size busses in an urban area makes up for the slight loss of efficiency in running a few busses that have oversized capacity for the routes they're on.

As for your low cost predictions for robotaxis, it's a very rosy picture, but I just can't buy in to that.

Besides, heavy users of transit get monthly passes, which drop the per-mile rider cost for transit way down. It's unlikely there'd be the same possible kinds of fare savings for robotaxis.

There are also huge road capacity considerations if robotaxis take a lot of riders away from transit: Imagine even half of the passengers of the current crowded busses during peak demand switching to robotaxis. An SF Muni bus at peak can easily be carrying 60+ or 100+ people (offical standard bus capacity is 83 passengers, or 124 in articulated busses). Putting just half of that into robotaxis means that that one vehicle on the road turns into 30+ or 50+.

It's highly questionable whether there'd even be nearly enough robotaxis available to serve so many people at times of peak capacity.

Then, what's the efficiency difference at off peak between one bus taking a route with maybe only 5 passengers versus those 5 people in individual robotaxis and 25 more robotaxis sitting idle somewhere waiting for the next period of peak capacity?

So it's not just pricing concerns that's going to keep transit systems around and worthwhile for more than "only the very poor" in and between urban areas.

Of course if robotaxis use variable, demand-based pricing then their costs will also go up significantly during peak hours, which is when a higher percentage of lower-income passengers will be travelling. Peak pricing on transit systems is uncommon, and its variability tends to be low (e.g. the London Underground has peak pricing which is only in effect on weekday mornings before 9:30am).

Further, the larger the amount of sprawl, as you suspect will continue and/or increase, the more attractive dedicated right-of-way transit (surface and subway trains, plus some bus rapid transit) is versus robotaxis to get into urban areas, for both cost and speed reasons. It will continue to be significantly cheaper per mile over longer trips, and faster due to avoiding surface streets and highway(/bridge/tunnel) bottlenecks.

The same road capacity and vehicle availability reasons also reduce the viability of using robocars for last-mile travel inside a city for people who use first-mile robocar-to-rapid transit. Imagine the demand and local traffic spikes at transit stations as a train arrives and a significant fraction of its passengers all make requests for robotaxis at the same time.

So, transit for urban intra-city passengers is still not just highly desirable but actually essential for both road capacity and vehicle availability. For travel into and between cities it would appear that again for the same road capacity and vehicle availability reasons taking a robocar for last-mile travel inside a city has more drawbacks than I first thought. The likliest scenario for commuting into a city then appears to be first-mile robocar to rapid transit, and then robobusses and walking plus a smaller amount of robotaxis for last-mile point-to-point travel inside the city. During the day robotaxis then fill in for any need of having a personal vehicle. On the return trip overall demand might be spread out enough for robotaxis to be available enough and cheap enough for a first mile return trip to a transit station. Then it's rapid transit again and a robocar from the local station near your house to home.

I'm writing this up in more detail elsewhere, but...

  • As we move to higher robocar density we can get a lot more traffic capacity on the roads, especially with half-width 1-2 person vehicles. At rush hour I'm talking 15x more capacity! Even more if you use vans, but would you use vans in that case?
  • Again, in that future world where you can make a small light car, a 500lb car that doesn't need to worry about being crushed, and the sensors get cheap due to Moore's law, you can make that vehicle really cheap. Though I agree, it's hard to compete with a transit pass. Though that pass only exists because the whole transit system is massively subsidized. But it's not hard to compete with a regular fare, or with the real cost of a fare.
  • For rush hour, I can see the use of vans and buses. But the problem is there is no desire for them outside rush hour. If the private trips are so cheap, and the roads not that loaded, who will wait for a van to save a tiny amount of money? And in that case, what pays for the vans and buses just at rush hour? Perhaps a very high congestion charge.
  • Turns out studies show the peak is not nearly as high a peak as you think. Rush hour isn't even the peak travel time, it's lunch hour!
  • Robotaxi cost is by the mile. Sitting around costs almost nothing, just the parking cost -- if there is a parking cost for something that can go anywhere to wait, even in front of a fire hydrant. So yes, 5 small robotaxis cost a lot less in terms of equipment and fuel than a city bus with 5 people on it.
  • Dedicated ROW is useful but quite wasteful today. We put in a rail line and there's one train on it every 15 to 30 minutes. It sits empty most of the time.
  • If you want to do the train, I think it works fairly well, actually, if you design your train station for it. The train pulls up with 5 60' cars. Lined up on the platform are 22 12' long robocars. Or possibly 44 on both sides, or if you do this right, 88 because they are 4' wide and you get two abreast. The first 120 passengers (some are in pairs) get into the cars, takes about 45 seconds, and off they go, in parallel sort of like another train, but this train immediately splits apart to go to all the 88 destinations. As they leave, 88 new cars follow them in and stop, and the next round of passengers moves in.
  • Don't like that? Well, have the people move off the platform to special robocar platforms one level below. The robocar platforms are packed densely with cars. Here we have extra lanes so cars don't have to wait for the car in front to move so it's a constant flow.

"As we move to higher robocar density"

There's a chasm that seems to keep getting jumped without looking at the distance between the edges.

A lot of your replies are very pie-in-the-sky: "half-width 1-2 person vehicles" providing "15x more capacity," "a 500lb car that doesn’t need to worry about being crushed" that's also "really cheap," that "it’s not hard to compete with a regular [transit] fare," "the private trips are so cheap, and the roads not that loaded," "small robotaxis," "Lined up on the [train] platform are 22 12’ long robocars. Or possibly 44 on both sides, or if you do this right, 88 because they are 4’ wide and you get two abreast," "special robocar platforms one level below...packed densely with cars."

So, here I am talking about realistic considerations over a timeframe of the next 5-10 years when robocars should conceivably become a reality, and you're replying with "solutions" that are maybe 15-30 years out (and that's being optimistic). There's a 10 to 20 year sliding window gap or chasm, here. Now it's fun to be speculative that far out, but I don't think it serves the promotion and advancement of self-driving cars in the near-term too well.

As for dedicated ROW only being used "every 15 to 30 minutes," again that may be the case in suburbia, but many stretches of dedicated ROW in an urban area are used by a vehicle (usually a train) about every 5 minutes. e.g. both BART and the Muni Metro under Market St. in SF, and most of the central section of the London Underground. At even only 75 to 100 passengers in each vehicle (at the LOW end of capacity), that's 900 to 1200 passengers per hour, per system, in each direction. So in SF since two systems are stacked that's actually 1800 to 2400 pasengers per hour, or 3600 to 4800 when both directions are combined. Handling that on the above surface road with robocars, plus the existing trafic that's also a lot of transit vehicles, is not something that looks feasible. It's also not credible to suggest re-purposing that undergound ROW for robocars, since there's nowhere to put all the ramps that would be needed to transition them to and from surface streets.

Outside of the Market St. corridor the Muni Metro shares most of its right of way with vehicular traffic. I can easily imagine a near future where the lanes that the Metro runs in, as well as the lanes that BRT will be using on Geary St. and Van Ness Ave, will become dedicated to self-driving vehicles as well as those transit systems. The self-driving vehicles will be able to coordinate with the transit systems, and both will achieve better levels of service without manually driven vehicles interfering in their non-coordinated way in the corridor. This is already somewhat in place on the surface level of Market St. and in the downtown section of the Geary St. corridor where lanes are dedicated to transit and (human driven) taxis. I could even imagine the Sunset tunnel between the Lower Haight and Cole Valley getting paved and allowing self-driving vehicles to share it with the current Metro system, which could reduce a lot of trip times for the self-driving vehicles.

This is once again a place where I see potential synergy between transit and robocars, whereas you more often seem to see opposition between them.

And I hope what you're "writing up in more detail elsewhere" doesn't end up being "Here are MORE reasons robocars are going to mean the end of mass transit," as I think that kind of thing only serves to antagonize, where I see many places where bridges can be built between proponents of both mass transit and robocars.

I agree, it's a fault of mine that I jump around in time like that. I have a long-term vision painted out and I should be more careful about what stages in it I am talking about. However, the path is reasonably incremental -- at least potentially. At least presuming this is a tech which gets and adoption curve similar to the late-20th century curves of new computer related technologies. That is not a small claim, since this is computers taking over transportation and transportation has different rules from the other things computers took over.

I do expect decent robotaxi services to start showing up reasonably early on, and then to grow. No, they won't be with ultralight cars at first, and yes it takes volume before they can be with cheap cars. But they will be cheaper than regular taxi fairly soon -- it turns out labour is 57% of the cost of a Manhattan taxi, for example.

$1/mile taxi outcompetes transit trips under 3 miles at the subsidized price, and under 6 miles at the real cost. San Francisco is only 7 miles wide.

Fully packed trains at rush hour do indeed make good use of their ROW, but that is cream-skimming. Trains must keep headways of 5 minutes because they have online stations. The train ROWs are not going away, of course, I am speaking more to the question of what new things we would plan looking forward, particularly with BRT (which should become Bus+conformant robot lanes) or dedicated ROW LRT. (Already many are starting to realize that BRT is a clear win over LRT in terms of cost effectiveness, but they need to go further.)

Trains can actually do even better numbers than you say, but only to serve a limited population that lives and works along them. Indeed, their ROW is so valuable that people move jobs and houses to be along them.

I do think there will be synergies, actually. And the ability of robocars to replace the big mass transit systems is more distant. But there is a problem, which is mass transit is only really efficient -- in terms of energy and in making use of that expensive ROW -- at rush hour. The ideal approach in the medium term is the mass transit at rush hour, and the robots the rest of the day. But there will be resistance to buying transit equipment to use only a few hours/day, and hiring staff who only work rush hour. Commuter rail -- which is the most efficient form of transit in the US -- does this in some cities, so it may be more possible than I think.

And the robojitney -- computer coordinated automatic vanpools -- may also make a lot of sense. One of transit's big issues is the modern city is not nearly so hub and spoke as that of the past. A majority of trips in the city do not involve the CBD, but most transit systems tend to take you though it to change trains.

You may be right that I overemphasize the issues around transit. This is in part because the things I have learned about the economics of transit are surprising and so largely unknown.

"presuming [robocars are] a tech which gets [an] adoption curve similar to the late-20th century curves of new computer related technologies. That is not a small claim"

Indeed, I think you are being overoptimistic by several degrees. We in high tech are used to this pace of change, and as science fiction fans, as well, we have visions of things we feel we know are surely just over the horizon. Then we run into the reality of the sargasso of multiple layers of city, state, and federal politics, plus opposition from existing entrenched industries, public opinion that is influenced by that opposition, other overoptimistic projections by private contractors and suppliers, etc.

"I do expect decent robotaxi services to start showing up reasonably early on"

I do as well, though I fear the opposition from existing cab companies and drivers as much as I fear opposition to self-driving busses and trains from transit drivers' unions. The cab companies are just as perniciously entrenched in municipal politics, and they are already facing disruption from things like Uber, so they have more forewarning and have more time to prepare for further threats like robotaxis.

"$1/mile taxi outcompetes transit trips"

Chasm jumping again. This is not near-term, and it requires many things to fall together into place just-so. (What was it you were saying before about complicated vs. dumb systems?) It also presumes the roads being able to handle the increased usage that would come with robotaxis being cheaper than transit. And it presumes that pricing wouldn't be affected by both peak pricing, which is more easily applied to robotaxis than transit, and congestion charges, which are similarly much more easily applied to robotaxis than to private vehicles.

You say "Fully packed trains at rush hour" and yet in your next paragraph you say "Trains can actually do even better numbers than you say." Indeed, I even mentioned I was describing the LOW end of capacity, and yet your counter-argument implies "cream skimming."

"BRT (which should become Bus+conformant robot lanes)"

We totally agree, there. The synergy should be obvious to all.

"[trains only] serve a limited population that lives and works along them."

But again, robocars and trains have synergies, here!

  • Quick and easy first- and last-mile robocar trips to and from nearby transit stations
  • No need to park-and-ride
  • No need to walk to a bus stop, to wait for a bus, to get you to the train
  • Quick and easy train station-to-destination travel at the other end of your train trip
  • On-demand personal vehicle access during the day while at your destination, so you don't need to worry that you won't have your own car available if you take the train
  • Faster travel, over a longer distance, at a lower price than either personal cars or robotaxis, with almost the same amount of convenience
  • None of the hassles of needing to find (in a self-driving car) and/or pay for (for both self-driving cars AND personal robocars) parking at your destination
  • No need to own a car in a city to reverse-commute to some suburbs (e.g. Silicon Valley)
  • Weekend recreational trips can be done with a cheaper, faster train for part of your trip and then you hire a robocar to drive around at your destination
  • Easy mixed-mode trips with transit in one direction and a robocar in the other (or multi-point mixed-mode travel), bringing back cargo, visitors with their luggage from airports, etc.

Over longer distances the synergies are clear, but long distance trains don't only have stations at the ends, they also have stations in the middle. This means long distance trains are also short distance ones for some people. Transit corridors remain highly valuable, useful things.

Robocars are actually more likely to increase than decrease the desirability and demand on non-local mass transit in the near term, increasing its efficiency.

"the ability of robocars to replace the big mass transit systems is more distant"

Presumption of desirability and feasibility. "More distant" may be asymptotically so. In the meantime the impression I get is that you're picking unnecessary fights with transit people by implying that one will replace the other.

"mass transit is only really efficient ... at rush hour" the suburbs.

And you said yourself that peak transit time isn't even rush hour, it's at lunch hour. In urban areas "peak" transit time is all throughout the day, with a few early, mid, and late spikes (which is why I try to avoid using the term "rush hour" rather than the broader "peak time").

"there will be resistance to buying transit equipment to use only a few hours/day..."

10 to 12+ hours a day in cities. (7am to 6pm? Yeah.)

Also, the economics are different if you're starting from scratch and buying all your equipment than if you're talking about places where there are already existing transit services and infrastructure, ROW is already owned, some land such as park-and-ride lots and employee parking can be repurposed, etc.

"...and hiring staff who only work rush hour."

As I said before, the same robocar tech can and surely will be applied to transit vehicles, as well, reducing payroll, insurance, pensions, etc. Yes, there will be transit union opposition. However, as I said above, there will similarly be cab company and cab driver opposition to robotaxis. I'd say that's a wash.

"Commuter rail ... is the most efficient form of transit in the US"

Ironically operating on a ROW that is either dedicated, or shared only with non-passenger carrying freight trains. :)

I have been talking about these synergies with planes more than with trains though, because of how short most train trips are, notably urban trains. It gets better if you can do just-in-time coordination.

The problem comes, with urban train trips of under 10 miles, of comparing a trip where you go 1 mile to the train, then N miles on the train, then 1 mile from the train to your destination with just staying in the car. Instead of 12 miles with 2 mode changes (even if done very well, which I think they eventually can be) you have a direct trip in the car. It costs a bit more, but it's still door to door with no parking hassle.

And we know what the public's answer to this question is. They overwhelmingly own and take cars on these trips. Transit has a tiny fraction of the travel in the USA, and while its fraction of commute miles is larger it's still overall quite small. And that's when cars have the hassles of parking and driving. The only advantage the train gets is a possibly slightly lower cost -- a cost that clearly does not mean that much to most commuters -- and the possibility of a shorter trip due to the dedicated ROW. Gone is the train's advantage that you can read on the train.

That's challenging, because in spite of congestion, the travel times of trains are usually not much shorter (especially on the door to door time) and if they are, it's only at rush hour. I sometimes would ride the N-Judah from our house in the Sunset (which was only 1 block from Judah) to downtown. 30 minutes that takes. It's decent and cheap at rush hour, especially with parking downtown being what it is. But come 7pm, it was an 11 minute drive and no way I would take the train. I expect the robocars to make that worse for the train.

Alas, mass transit is not efficient even in the dense urban areas outside of rush hour. It is not energy efficient -- green cars do much better -- and when traffic is off-peak it is not especially road efficient and certainly not time efficient. That's transit's problem -- great at rush hour, moves a lot of people for reasonable energy and reasonable road space. But then it has to run all day when it's horribly inefficient, resulting in an inefficient overall performance. BART, New York MTA, Boston T -- none of them beat two people in a good electric car. In fact, they are pretty close to a solo driver. Forget SF Muni -- it's worse than the average gasoline car.

No, the peak driving time is lunch hour. I believe the peak transit times are the rush hours. I will need to look at the load numbers but my understanding is that rush hour is "standing room only" and non-peaks are much more relaxed. It has to be for the averages to be so poor -- 9 passenger per bus on average, 23 per train car, in the USA.

Yes, commuter rail's ROW is a big attraction for it. The reason commuter rail is so efficient is that many of them are very rush hour oriented and run mostly full trains. The most efficient commuter rail routes are the ones that are rush hour only and so long that trains do only one run, and wait in the city to come back out, and wait overnight in the burbs. No empty trains at all. These are attractive in places with high road congestion and they offer the commuter the chance to relax and read. Outside of NYC, where transit is most popular, they are the transit that rich people ride. But you have to be a 9 to 5er to ride these sorts of lines.

On another note, one thing I would like to model is "what's the optimal size of a transit vehicle?" -- in particular if, like in today's systems you can have only one. (Not quite for trains which can change length.)

The first obvious thing is that there is an optimum. Make the vehicle too big, even big enough for the ridership at 5pm, and all those empty seats the rest of the day cost you too much.

Factors to put in the eventual full model would include:

  • How much does extra seating capacity increase the weight, and thus vehicle cost and energy to move it?
  • Less capacity means more energy per passenger in full vehicles
  • Smaller vehicles can (and need to) run with greater frequency, which increases demand due to shorter wait times.
  • Alternately to greater frequency, smaller vehicles can mean more routes and a larger service area for more demand.
  • More vehicles means more drivers -- until robocars
  • More people means more starts and stops to pick up and drop off and longer waits at stops. (Regen braking only does so much.)
  • Demand-response (ie. taxi-like) can involve not even going to places with no demand, but also some empty vehicle movement
  • Direct A to B trips reduce overall mileage compared to trips involving transfers
  • A smaller number of larger vehicles creates less road congestion

This is a complex model, and I haven't built it in full. It might be a thesis project to do it well. However, my initial math is suggesting the optimal size is small, or rather a mix of mostly small sizes, and thus the debate here about transit. A simple demonstration of that is to take the average city bus in the USA which gets under 4mpg and carries 9 people on average. Here it's clear that a Prius wins. (The 8mpg hybrid bus beats out the solo Prius driver but not the average 1.5 pax Prius.)

Given the amount of land around the current malls their owners could (depending on location) make a reasonably large amount of money redeveloping the land for housing and office space. Which in turn could make them strong proponents for ROD.

I hate to dump a lot of cold water on this very cool technology, but when the first passenger gets killed in a robot car, the lawsuits will bankrupt the companies who designed it, built it, and/or sold it.

Shakespeare was right: First we kill all the lawyers.

Most of the discussion as taking the technical development point of view. This discussion also needs to look from the consumer point-of-view. I'm a lay person, but have been fascinated with the idea of self-driving personal transportation. Personal Transportation will become a service that the consumer will subscribe to. The fees will be based on quality of service, distance and time of travel. What the subscriber won't have to pay for is significant: auto loan, registration, insurance, maintenance or fuel. Also the user won't need to have a driver's license.

Driverless transportation will also have a big impact on law enforcement and the judicial system, as many traffic-related crimes will not happen. For example, driving under the influence, not having a valid driver's license or insurance will be moot. Cities can gain cost-savings by reducing police forces, and the judicial system, which would include judges, prosecutors, jails, etc.

When discussing potential markets to be served by robocars, other groups of people to include, along with children, are: people who lost their driver's license, and the elderly who want to be independent.

One last thought. The future robocar doesn't have to be modeled on today's typical automobile. They only have to be large enough for a single passenger. A robocar can be small and lightweight. The transportation system doesn't have to be ground-based like today's road system. The future transportation system can be segregated so that passenger, cargo, utility, and emergency transport can be run on separate systems. These systems can be elevated in tiers. The utility system (garbage collection, deliveries) can be underground. Passenger systems can be elevated, leaving ground level for pedestrians.

Add new comment