I wrote the longer post below in response to your assertion that redundant vehicles need to be cruising around. I have improved some of the calculations just now. Did you read it?
>>Household vehicles were driven an average of 64.6 minutes on a typical day in 2022 (including all trips made that day) and parked for the remainder of the time (95%).<<
It seems clear to me that Waymo should prepare for peak demand. Driverless cars are fundamentally different from taxis with drivers, because there is no driver to pay. Owners of private automobiles are not losing sleep because their cars are parked 95% of the day. And they're not worried about the interest on the money they paid for the car which continues even when the car is not in use. The decision to purchase a car and bear that interest has already been made because of the utility of having a car available when they want it. After that decision has been made, the only cost that owners of private vehicles are concerned about is maintenance. And vehicles that are not moving are not degrading.
So if private owners are quite happy with a situation where their vehicles are not being used 23 hours of the day, than Waymo should be happy with a situation where their vehicles are used only half of the day.
>>What do you do with redundant vehicles during the off-peak? Why, you have them cruising the streets, so that customers who would otherwise have to wait 8 minutes for a pickup instead wait 4. <<
I see no reason why they need to be moving around when unoccupied. They just need to be spread around the city in a smart way. Yes they need a parking place, but parking unoccupied driverless vehicles must be a much smaller problem than parking all the vehicles for the commuters into the city.
Let's imagine that the goal is five-minute response time to a call. As a potential rider, 5 minutes seems quite reasonable to me. Now Waymo (or whoever) needs to have enough unoccupied vehicles that even at peak demand there is still at least one waiting vehicle within 5 minutes of every point in the city.
I can imagine Waymo buying a city lot that is vacant or covered with less valuable real estate. They excavate the lot down about 12 ft and build a basement with a ramp to the surface. Then pour a concrete roof over the basement and build a commercial establishment on top (or a park etc.).
>>The number of cars that can park in a city lot depends on the lot's size, layout, and whether it's designed for efficient parking or includes landscaping and walkways. A standard acre can hold roughly 144 traditional parking spaces, but practical layouts typically accommodate 100-115 spaces. <<
>>As of 2024, Waymo's fifth-generation robotaxis were based on Jaguar I-Pace electric vehicles augmented with automatic driving equipment that according to Dolgov costs up to $100,000.<<
Rounding down, lets say 100 spaces for driverless cars (DVs). Each car must be connected to a charger. Since they are parked by robots, the margins can be very tight. So 100 vehicles would actually require much less than an acre.
Now Waymo needs one of these basements within a 5 minute drive of every point of the city.
>>During rush hour, city traffic speeds typically drop significantly, with average speeds ranging from around 10 to 20 miles per hour, depending on the city and time of day.<<
We definitely need to be conservative in this case, so assume 12 miles per hour. One mile every 5 minutes.
Now we need a grid of basement parking lots every 1.5 miles in two dimensions. (Visualize a square 1.5 miles on a side with a parking lot at each corner: most of the points within the square are no more than one mile from a corner. A point in the center is 1.5 miles from a corner when following a rectilinear path. The unlucky client in the exact center of the square must wait 8 minutes for pickup.)
That works out to one parking lot per 2.25 square miles. Underground parking means zero usurpation of commercial/living/recreation space.
Imagine a city of 100 square miles. That's 44 parking lots (@ $10 million per) and 4400 DVs (@ $100,000 per with cost reductions) to provide 5-minute response time. (Less than $1 billion total -- cheap!)
Quite a bit of slack there: you can go from 100 parked vehicles to 1 parked vehicle and still provide 5-minute response time. Of course at rush hour, riders are constantly exiting DVs and those randomly-spaced vehicles can immediately respond to calls. So the average pickup would be less than 5 minutes at busy times (and at slack times, average speed would be greater than 12 mph). Also note that when a vacated DV is not needed, it goes to the nearest parking basement. No need to return to where it began.
While your underground parking proposal is clever, and meets the need, I'm stuck on the price tag. A billion dollars per city is a huge investment for any company, even one with deep pockets... especially as just having empty cars drive around costs almost nothing. For a company trying to launch in dozens of cities, that's a pretty appealing difference.
I do think you're right that smart positioning of vehicles is crucial. I'm curious to see how Waymo and Zoox attempt to solve this (Tesla won't bother). Thanks again.
Note that the billion dollar price tag is the total cost including 4400 cars. Also, it is quite possible that with a little practice, a basement parking lot (mostly poured concrete) could be built for less than $10 million dollars.
Finally, what would be the cost of a subway system with cars arriving every 5 minutes in every part of a hundred square mile city?
Certainly we will not see Waymo making any billion dollar investments anytime soon. I'm basically trying to get a sense of the potential for driverless vehicles. I'm quite encouraged.
P.S. Substack tells me that I must update my profile before I can comment. Facebook would never have been a success if it was this hard to make a comment.
I had to find my substack app and refind your reply in order to post my reply.
Note that parking lots are only so large because you need arbitrary access for all cars. For waymo you can send any car of the right specs, so you can just line them all up one after the other in a LIFO style queue. Disadvantage is if a single car breaks all other cars are stuck, so you don't want it too deep, but 5 cars per queue seems fine.
Thank you for joining the discussion! I hope you say more.
I had exactly the same idea. Oh, you can take any car! They can line up bumper to bumper and move forward automatically. I was thinking 10 queues of 10 cars each.
But then I remembered that they all need to be attached to chargers. So I thought about how each car could automatically disconnect from one charger, move ahead in the line, and then reconnect. But after playing with that idea for a while, I decided: Nah, too much trouble, just use a little more space.
Right now I'm thinking diagonal parking (like the Japanese). Something like this:
* Line of 10 cars angled left and connected to chargers on the left wall
* Access lane
* Line of 10 cars angled right
* Line of 20 chargers
* Line of 10 cars angled left
* Access lane
(3 more of above)
* Line of 10 cars angled right and connected to chargers on the right wall
How all this relates to the ramps to street level is a little complicated, but I'm sure can be figured out.
So that's five access lanes which can be hardly wider than the width of a car. Not bad.
Here is another thing to think about. I live in the Dominican Republic. We have these tiny vans ("rutas") that run around with no door on the passenger side and bench seats inside. They pick up passengers until they are jammed full. (I find them uncomfortable.) A ride costs about 85 cents (50 pesos). Waymo doesn't allow ride sharing, but I see no reason why that might not be an option in the future. An option chosen on the app when you call for a car.
Not only would it reduce the cost per passenger, but it would reduce the time waiting for a pickup (because of the numbers of DV cars passing by that could pick you up). No bench seats please.
You only need to move the cars along if it's a FIFO queue. If you use a LIFO stack there's no need, they can stay in till they're ready to leave. Disadvantage is you can get some traffic between cars coming in and those going out, but that might be worth the simplicity of not needing to disconnect and reattach chargers.
Alternatively you could use inductive charging, that's less efficient but might be worth the labour cost savings.
Should you become interested in designing a parking deck for WayMore, the pillars in the deck I toured today are 18" x 18" and the spans are mostly 15' with some longer ones.
Could I talk you into designing an underground parking lot for 100 Waymo vehicles?
Jaguar I-pace
Width 215 cm -- 85 in
Length 470 cm -- 185 in
85×185 = 15,725 = 110 sq ft
You need an entrance/exit ramp, walls, and pillars. I'll get you some specs on pillars and spacing tomorrow if you wish. There is a parking lot underneath a hotel where a friend of mine works. I'm sure I can walk through the deck and make a few notes.
If you want to have some fun designing a driverless vehicle company, I think we should call it WayMore.
I was alarmed when I saw the land prices in LA ($5 mil + for a 10,000 sq ft lot), but then I remembered that we get all that back. When the parking deck is finished there's a smooth concrete surface at street level ready for anything anybody wants to do with it.
One final idea: I know next to nothing about induction charging, but surely the closer the coils are together, the more efficient the transmission of energy will be. So imagine the coil in the car covering part of the bottom surface of the car. Seems like the coil in the parking lot could raise up a few inches so that they are very close to the car.
Yes, I was thinking exactly that: FIFO. I think the two-way traffic with LIFO could be handled easily, because a computer is controlling everything. If you (the computer) have a car coming in on one stack, then you just pick a different stack when you need a car to exit. Sold!
Induction charging could work fine, because the cars will spend a lot of time parked. I like the simplicity very much. Error proof!
Note: I would use induction charging even with LIFO because of the simplicity and automated handling.
I wrote the longer post below in response to your assertion that redundant vehicles need to be cruising around. I have improved some of the calculations just now. Did you read it?
>>Household vehicles were driven an average of 64.6 minutes on a typical day in 2022 (including all trips made that day) and parked for the remainder of the time (95%).<<
https://www.energy.gov/eere/vehicles/articles/fotw-1356-august-19-2024-household-vehicles-were-parked-95-typical-day-2022
It seems clear to me that Waymo should prepare for peak demand. Driverless cars are fundamentally different from taxis with drivers, because there is no driver to pay. Owners of private automobiles are not losing sleep because their cars are parked 95% of the day. And they're not worried about the interest on the money they paid for the car which continues even when the car is not in use. The decision to purchase a car and bear that interest has already been made because of the utility of having a car available when they want it. After that decision has been made, the only cost that owners of private vehicles are concerned about is maintenance. And vehicles that are not moving are not degrading.
So if private owners are quite happy with a situation where their vehicles are not being used 23 hours of the day, than Waymo should be happy with a situation where their vehicles are used only half of the day.
>>What do you do with redundant vehicles during the off-peak? Why, you have them cruising the streets, so that customers who would otherwise have to wait 8 minutes for a pickup instead wait 4. <<
I see no reason why they need to be moving around when unoccupied. They just need to be spread around the city in a smart way. Yes they need a parking place, but parking unoccupied driverless vehicles must be a much smaller problem than parking all the vehicles for the commuters into the city.
Let's imagine that the goal is five-minute response time to a call. As a potential rider, 5 minutes seems quite reasonable to me. Now Waymo (or whoever) needs to have enough unoccupied vehicles that even at peak demand there is still at least one waiting vehicle within 5 minutes of every point in the city.
I can imagine Waymo buying a city lot that is vacant or covered with less valuable real estate. They excavate the lot down about 12 ft and build a basement with a ramp to the surface. Then pour a concrete roof over the basement and build a commercial establishment on top (or a park etc.).
>>The number of cars that can park in a city lot depends on the lot's size, layout, and whether it's designed for efficient parking or includes landscaping and walkways. A standard acre can hold roughly 144 traditional parking spaces, but practical layouts typically accommodate 100-115 spaces. <<
>>As of 2024, Waymo's fifth-generation robotaxis were based on Jaguar I-Pace electric vehicles augmented with automatic driving equipment that according to Dolgov costs up to $100,000.<<
Rounding down, lets say 100 spaces for driverless cars (DVs). Each car must be connected to a charger. Since they are parked by robots, the margins can be very tight. So 100 vehicles would actually require much less than an acre.
Now Waymo needs one of these basements within a 5 minute drive of every point of the city.
>>During rush hour, city traffic speeds typically drop significantly, with average speeds ranging from around 10 to 20 miles per hour, depending on the city and time of day.<<
We definitely need to be conservative in this case, so assume 12 miles per hour. One mile every 5 minutes.
Now we need a grid of basement parking lots every 1.5 miles in two dimensions. (Visualize a square 1.5 miles on a side with a parking lot at each corner: most of the points within the square are no more than one mile from a corner. A point in the center is 1.5 miles from a corner when following a rectilinear path. The unlucky client in the exact center of the square must wait 8 minutes for pickup.)
That works out to one parking lot per 2.25 square miles. Underground parking means zero usurpation of commercial/living/recreation space.
Imagine a city of 100 square miles. That's 44 parking lots (@ $10 million per) and 4400 DVs (@ $100,000 per with cost reductions) to provide 5-minute response time. (Less than $1 billion total -- cheap!)
Quite a bit of slack there: you can go from 100 parked vehicles to 1 parked vehicle and still provide 5-minute response time. Of course at rush hour, riders are constantly exiting DVs and those randomly-spaced vehicles can immediately respond to calls. So the average pickup would be less than 5 minutes at busy times (and at slack times, average speed would be greater than 12 mph). Also note that when a vacated DV is not needed, it goes to the nearest parking basement. No need to return to where it began.
Sounds good to me.
Thanks for working through this so carefully!
While your underground parking proposal is clever, and meets the need, I'm stuck on the price tag. A billion dollars per city is a huge investment for any company, even one with deep pockets... especially as just having empty cars drive around costs almost nothing. For a company trying to launch in dozens of cities, that's a pretty appealing difference.
I do think you're right that smart positioning of vehicles is crucial. I'm curious to see how Waymo and Zoox attempt to solve this (Tesla won't bother). Thanks again.
Thank you very much for responding!
Note that the billion dollar price tag is the total cost including 4400 cars. Also, it is quite possible that with a little practice, a basement parking lot (mostly poured concrete) could be built for less than $10 million dollars.
Finally, what would be the cost of a subway system with cars arriving every 5 minutes in every part of a hundred square mile city?
Certainly we will not see Waymo making any billion dollar investments anytime soon. I'm basically trying to get a sense of the potential for driverless vehicles. I'm quite encouraged.
P.S. Substack tells me that I must update my profile before I can comment. Facebook would never have been a success if it was this hard to make a comment.
I had to find my substack app and refind your reply in order to post my reply.
Note that parking lots are only so large because you need arbitrary access for all cars. For waymo you can send any car of the right specs, so you can just line them all up one after the other in a LIFO style queue. Disadvantage is if a single car breaks all other cars are stuck, so you don't want it too deep, but 5 cars per queue seems fine.
Thank you for joining the discussion! I hope you say more.
I had exactly the same idea. Oh, you can take any car! They can line up bumper to bumper and move forward automatically. I was thinking 10 queues of 10 cars each.
But then I remembered that they all need to be attached to chargers. So I thought about how each car could automatically disconnect from one charger, move ahead in the line, and then reconnect. But after playing with that idea for a while, I decided: Nah, too much trouble, just use a little more space.
https://www.reddit.com/r/askmath/comments/1522pm6/is_this_car_park_in_japan_more_space_efficient/?rdt=48323
Right now I'm thinking diagonal parking (like the Japanese). Something like this:
* Line of 10 cars angled left and connected to chargers on the left wall
* Access lane
* Line of 10 cars angled right
* Line of 20 chargers
* Line of 10 cars angled left
* Access lane
(3 more of above)
* Line of 10 cars angled right and connected to chargers on the right wall
How all this relates to the ramps to street level is a little complicated, but I'm sure can be figured out.
So that's five access lanes which can be hardly wider than the width of a car. Not bad.
Here is another thing to think about. I live in the Dominican Republic. We have these tiny vans ("rutas") that run around with no door on the passenger side and bench seats inside. They pick up passengers until they are jammed full. (I find them uncomfortable.) A ride costs about 85 cents (50 pesos). Waymo doesn't allow ride sharing, but I see no reason why that might not be an option in the future. An option chosen on the app when you call for a car.
Not only would it reduce the cost per passenger, but it would reduce the time waiting for a pickup (because of the numbers of DV cars passing by that could pick you up). No bench seats please.
You only need to move the cars along if it's a FIFO queue. If you use a LIFO stack there's no need, they can stay in till they're ready to leave. Disadvantage is you can get some traffic between cars coming in and those going out, but that might be worth the simplicity of not needing to disconnect and reattach chargers.
Alternatively you could use inductive charging, that's less efficient but might be worth the labour cost savings.
Should you become interested in designing a parking deck for WayMore, the pillars in the deck I toured today are 18" x 18" and the spans are mostly 15' with some longer ones.
Could I talk you into designing an underground parking lot for 100 Waymo vehicles?
Jaguar I-pace
Width 215 cm -- 85 in
Length 470 cm -- 185 in
85×185 = 15,725 = 110 sq ft
You need an entrance/exit ramp, walls, and pillars. I'll get you some specs on pillars and spacing tomorrow if you wish. There is a parking lot underneath a hotel where a friend of mine works. I'm sure I can walk through the deck and make a few notes.
If you want to have some fun designing a driverless vehicle company, I think we should call it WayMore.
I was alarmed when I saw the land prices in LA ($5 mil + for a 10,000 sq ft lot), but then I remembered that we get all that back. When the parking deck is finished there's a smooth concrete surface at street level ready for anything anybody wants to do with it.
One final idea: I know next to nothing about induction charging, but surely the closer the coils are together, the more efficient the transmission of energy will be. So imagine the coil in the car covering part of the bottom surface of the car. Seems like the coil in the parking lot could raise up a few inches so that they are very close to the car.
Excellent! Two great ideas.
Yes, I was thinking exactly that: FIFO. I think the two-way traffic with LIFO could be handled easily, because a computer is controlling everything. If you (the computer) have a car coming in on one stack, then you just pick a different stack when you need a car to exit. Sold!
Induction charging could work fine, because the cars will spend a lot of time parked. I like the simplicity very much. Error proof!
Note: I would use induction charging even with LIFO because of the simplicity and automated handling.