I think for instance we can't possibly build enough railway and bus network to serve every inch of the territory and it's simply to allow fast speed individual transport to distribute the load. It can be even automated with no pilot control, but it's impossible to schedule-base every location in a city without displacing people, rebuilding cities entirely around specific common transport etc. In Europe and Asia, where we build a lot of cities organically, it seems silly. In the US, where cars are even MORE prevalent to the point cities are built for them rather than humans, it will never disappear.
People love to respond that that's too expensive or unworkable or unrealistic or whatever, ignoring that it's actually the status quo that is too expensive, unworkable and unrealisitc, and that there are now a bunch of previously car centric cities that have been deliberately and fundamentally changed very successfully.
Cars are not the problem: it's their aggregation in dense city + their pollution. Remove the pollution, people stop dying, we can then discuss your dream of perfect public transport for optimization's sake.
I'd want to see some numbers to support that. What does it cost to provide round-the-clock availability of secure, climate-controlled public transport to every possible location, urban or rural? Bullet trains on standby at 0100hrs to take 1 guy from Osaka to Kanagawa would cost a fortune. In comparison, while PARKING and roads definitely have costs associated with them, the hardware/infrastructure is largely concrete, and inexpensive to maintain when operating below capacity. So 1-2 people can hop in a car and take a road trip at a far lower net/civilization-wide resource expenditure.
Japan has great public transport. Japan also has great cars and a very healthy car community. What part of Japan's transportation approach classifies it as "unworkable and unrealistic" to such an extent that completely banning cars is the optimal solution?