Earlier quoted context omitted.
I idly wonder if what would actually make sense here is a hybrid model that combines a gigantic fulfillment center with tens of thousands of products located "far" from people, with a large physical footprint and near to road/rail arteries, but with a mid-bandwidth, high-granularity, low-latency physical link to "near" places. For example, imagine you had an upscaled pneumatic tube system (don't get hung up on the ex…
Buried anything is just horrendously expensive. Partly because of other things that are already buried.
Even if it was a good idea (which I doubt, it's just a idle thought), I don't think there's a practical way to retrofit such a system in existing cities due to the costs, planning and presumably private funding for a non-public network, because the public road system exists, needs to continue to exist for large items and can be used for virtually free in comparison. So if/when the depot-to-neighbourhood leg is automated, it's much more likely we'll see drone vehicles on the road or occasionally in the air instead of dedicated pipeline-like delivery systems.
Even without the pipeline, you can conceive of self-diving heavy vehicles pulling up next to local delivery hubs and disgorging thousands of shipping pods into a robotic receiver. From there they either get picked up, droned, Starship'ed, cycled, whatever to the eventual front door. It's still possible just having then self-drive right to the door would turn out cheaper.
Sounds very sterile as an experience, but really it's only an optimisation of the small, but highly distributed, remaining segment of inefficiency in the existing global machine that already converts raw materials to a widget or food and gets it to within 100 miles of your house.