Technology equivalent to contemporary tech.

A city with a population of millions built from scratch.

There is a need for an inexpensive underground system that would transport cargo (at best everything from packages, food, garbage, construction materials, etc.) within the city and free the streets from cargo transport.

The system should be reliable and pay for itself in the long term. (At best it should be able to effectively handle different sizes of packages)


1) Which technology should be realistically used? Tiny rail cars? Conveyor belts? Pneumatic post? Something else?

2) What is the biggest size of package that would be realistic?

3) Should such a cargo system in any way use the city metro system?


6 Answers 6


Realistically, underground is much more expensive than above ground using delivery drones. Delivery drones (multi-copters) require much less overhead to achieve much better logistics. Consider several miles per square mile of real estate for rails, tubes or conveyors vs perhaps a few dozen, or even a few hundred drones to get the same coverage and perhaps a few dozen charging stations. Even if the infrastructure is already in place, i.e. an existing underground network of tunnels, the cost of buying and installing any form of conduit far outweighs the cost of buying a handful of delivery devices which require no major infrastructure.

The next best option as far as my limited understanding and basic assumptions go (see the list below the image) is a pneumatic tube system. You would likely want something more complex and versatile than what is used in bank drive-throughs, but I am not up on the engineering aspects of such systems.

The basics are here (click the image).

enter image description here


I neglected part of the question which states that the city would be built "from scratch". In that case, implementing a system such as pneumatic tubes would simply increase the overall cost of design, materials, planning and engineering for the layout, but would not dramatically limit what could be done assuming that the expectation is that this option would involve a certain up front investment. Furthermore, California is seriously considering a tube transit system for people, but it could be duel purpose of done cleverly or simply for transporting packages. For example, given a tube car design with separate compartments for people and packages (above and below for example), such a system could conceivably trade packages efficently at stops while people are exiting and boarding the tube car.

The California Hyperloop concept is shown above ground, but the principle is the same whether above or below ground.

the tube design

The tube concept.

a tube car

A tube car cut away. Note that the turbine is for the purpose of evacuating the tube ahead of the car to reduce drag, not for the purpose of propulsion.

Below are considerations if the city must be retrofitted for such a system.


  • existing infrastructure may potentially be repurposed, i.e. old water pipes which may be leaky, could be emptied, sealed with any form of chemical sealer to make them air tight as this would no longer be a health hazard
  • underground network and infrastructure already exists making upgrades and installations less costly


  • packages would have a maximum dimension by diameter and length and all packages are shipped inside a canister designed to work with the tube system
  • the simplest way to design the system is to provide a single tube diameter which best fits most size requirements
  • larger sizes must be transported by another method
  • smaller sizes still render the cost of a single transit as they would simply be secured inside a canister
  • if using existing infrastructure, every building would have to be refitted to accept or send packages
  • if using existing infrastructure the cost of refitting every building must be subsidized because as any network operates (like the internet, or social media) its value is determined by how well connected it is - i.e. if there are thousands of miles of tubes and only a handful of end points using the system, then value of the system as a whole is well below the overall cost of implementation
  • $\begingroup$ Since the city is being build from scratch and you are already building a sewage system and laying down a few thousand kilometers of power and internet line, isnt the addition of one more system of tubes or whatever pretty cost-effective to add since you have the ground open anyway? And the advantages of not needing to worry about weather and navigation in a cityscape (especially with a high skyline) are very good in the long run. $\endgroup$
    – Demigan
    Dec 12, 2019 at 7:05
  • $\begingroup$ Yeah, for some reason I spaced on the "built from scratch" part, so there's a lot of extraneous information, but I decided to leave it up anyway in case anyone else happened upon the question. It is true that a new construction would considerably reduce the overhead for implementing such a system. $\endgroup$
    – Nolo
    Dec 12, 2019 at 12:47
  • $\begingroup$ I think the biggest economic problem is that trains can only go where rails have already been laid down. Cities grow and evolve, which is why cars, bikes, golf carts, and maybe drones are better: they're more adaptable. Maybe if there's something about the layout of this new city, for example, all blocks and streets are permanently fixed, and there's one rail "station" at the center of each block, so a rail network can be laid down once and doesn't need to change. $\endgroup$
    – workerjoe
    Dec 12, 2019 at 17:19
  • $\begingroup$ @workerjoe There's always the possibility of going vertical... consider a rail system built on columns - which themselves could be stacked, presumably up to a given number of levels based on the design of the columns. If a second layer of the rail system wanted to go where there was no other rail under it, then plain columns with stabilizing cross-members could be used to maintain the new level at the same height. This could be built over streets, or even interchangeably with automobile lanes at any given level. Only limited by technical and logistical ability of the society. $\endgroup$
    – Nolo
    Dec 13, 2019 at 13:00

Look into Walt Disney's plans for the City of EPCOT, which is not the park by the same name we have today.

The central part of the city would route all automobile traffic to a series of roads underneath the public portion of the city, to reduce pedestrian traffic access, and allow for trucks and supplies to navigate the streets to basement cargo bays and garages for their buildings, while foot traffic would be at the ground floor* and the only vehicles would be mass transit (People Movers as scene in Magic Kingdom in Walt Disney World for local commuting, with Monorails for rapid transit). One could modify this so that above ground roads* are for private vehicles (personal cars) while lower roads were for bulk cars, and lay the underside roads inline with the aboveside roads for ease.

While not city management it's common knowledge that Magic Kingdom's public area is served by a series of underground hallways* that run the length of the park and are able to remove garbage, place costumed characters and cast members dressed for specific theme lands, and restock gift shops without a single guest seeing an errant cast member.

*Because all of these were intended to go in the Orlando area by Walt, the use of Underground is not the best term but the easiest. Florida has a high water table and structures in Florida do not have basements as they would be flooded almost instantly. EPCOT City and Magic Kingdom would have/were built with the "underground structures" being on the ground floor and then the public portion was the "second floor". THere's some discreet landscaping to hide this Around Magic Kingdom.


Because of the assumption of totally new city construction there may be an alternate, feasible solution.

Each property/building requires a unique address with connection to communications, electrical power, potable water supply and storm water pipes. Optionally there may be a recycled water supply for gardens, toilets, etc. Plus organic waste, recycling waste and hard waste services. All of these require trenching for underground pipes and conduit. if this trenching work is done prior to building construction or as part of foundation work then the cost of increasing the diameter of services for package delivery is trivial. A previous answer suggested laying pipes, conduits, etc. on ground level and raising the buildings by one level. A variant would be to lay the pipes, etc. on ground level and back fill over the pipes and conduits to establish a new ground level. With the use of suitable back fill material you could aim for well drained soil.

With a suitable design with recycling pumps the potable water supply can have continuous flow. Increase the diameter of the water pipes and the system could be used to delivery water proof cargo capsules. As each location has a unique address then RFID or equivalent technology could be used for package routing and delivery. It should be possible to have a hierarchy of pipe sizes for packages. For additional cost a property/building owner could specify a larger diameter pipe. To reduce pipe pressure each property would require a pump to raise water pressure. Pump input could automatically swap between local storm water tanks and the low pressure, mains water supply. Return of used carriers could be using a continuous flow storm water system. In the event of a storm event the return carriers would be flushed from the system to increase storm water capacity. Without capsules the storm water system would be able to handle the increased water flows. Only sterilized capsules could have access to the delivery system (potable water supply). Every property would be restricted to the storm water system. At the storm water system pumping stations the there would be sorting of the full from the empty, return capsules. Sterilized capsules would be inserted into the delivery system. If required, empty capsules could be then be delivered to properties for reuse. There could be different capsules for different contents. With RFID technology high priority capsules could be given processing priority. Each property would be required to extract capsules from the system. so there can not be a backlog causing a jam. If there was a jam deliveries to the affected address would be held back. Aim would be automate the system. Using fluid logic as an alternative to RFID it should be possible to use the water flow to route packages.


How about a network of tiny underground railways, pretty much down every downtown street in a town like Chicago. Every street crossing would also have an underground rail crossing, with connector tracks in every direction.

A single track railway would suffice, since in opeations you would simply make all the tracks "one way streets".

It wouldn't use the metro subway system at all. The reason is the typical city metro system is too busy to accommodate the extra traffic.

Every building of importance would either have a siding and an elevator, or simply set the building sub-basement level even with the tracks.

Think tubes maybe 5 feet wide and 7 feet tall, just tall enough a human could walk down the tube without hitting head on the trolley wire. Max package would be about 4' x 4' x 20' or so.

The railway could be run by either humans or automation, depending on the tech level.


if you can build the city from scratch, you could build a transportation system on flat land and then build a concrete ceiling on top which will be the floor of the city (or the next underground level). That underground could be used by robots like our today automated storage systems. The maximum size may be about the same that todays subways have. Of course, you could also use these robots to transport humans. They could even be used for individual traffic. You could enter your destination and have a kind of 2D-Elevator that rides you through the city.


Whatever system you use will boil down to a few basic principles

  1. When delivering a load of cargo going to one place, a single larger transport vehicle is cheaper and equally fast per unit as many small transports.

  2. When delivering a load of cargo to many places, many small transports is both faster and cheaper per unit than 1 large transport making many sequential stops.

  3. The less you need to pack and unpack between transports, the faster and cheaper per unit your transport is.

So, the first part of answering your question is figuring out the optimal layout. If your city has a single massive port for ingoing and outgoing logistics managed by a single entity, then you can minimize the overhead of getting supplies in and out of the city to begin with. Instead of each corporation having their own distribution center, in your city you minimize waste on underloaded bulk freight by making them all share. This central hub would basically function like an Amazon warehouse just being a catch-all for all goods and services. From their you do not necessarily want to unpack all of your freight containers yet though. If you have an industrial district in one place, and residential district in another, then you can send individual freight containers closer to their final destination by just loading them right back on smaller subway trains to district hubs. From those hubs, cargo can be further divided up and sent on even smaller subway carts bringing goods directly to people's homes and buisnessess.

enter image description here

Your Regional Subways need to accommodate train car containers from 8x8.5x20 ft to 8x9.5x40 ft to make sure you never need to unpack one to get it to its right district and remain compatible with all international standards of shipping container sizes.

Your District Subways however are not your normal train system. They are more like tracks for self propelled mining carts that range in size from a shoe box for small parcels up to 4x6x12 ft carts designed for construction materials and large furniture. This way, instead of loading up a large truck and carrying your things to a lot of other stops before arriving at your home, you send the right sized vehicle for the job to pickup or deliver goods in just 1 stop.

Then every home or business has a basement that functions their own private receiving station. You get home from work, and your groceries are just sitting down there in a cart waiting for you to unpack it.

Why subway carts?

Pneumatic tubes: They require a seal that causes more friction than wheels and more precise construction to make; so, the cost to make and operate them is more than traditional subways.

Flying Drones: They require less infrastructure, but are much more expensive to operate. Carts passively resist gravity by sitting on the ground, but flying drones need to expend a constant energy just not to fall. They are good for premium expedited shipping of small parcels, but if you are moving around heavy bulk things like furniture, garbage, plywood, etc. then you are wasting a lot of money on them

Conveyor belts They will either need to be way overpowered for most of what they transport or they will need to have complex transmission systems distributed every few feet which will break often due to constant gearing up and down. Either way, they are not as efficient.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .