There are simulators for this, and of course there's data from places that have actually done it.
In Dublin we have a bit of a mixture of newish bus routes which largely have a sensible number of stops, and ancient routes (the oldest evolved out of tram routes laid out in the 1870s), which tend to have a stupidly high number of stops, because once you put one in it's very contentious to remove it. The super-regular stop routes are _so slow_.
The opposite actually tends to increase ridership; complex fare systems put people off and tend to make getting on the bus more time consuming (may require each passenger to interact with the driver).
I suspect the evidence here would fall mostly on the side of "it increases ridership", though it's probably hard to study, as it's rarely done in isolation, but more commonly as part of route redesign.
So this used to happen on Dublin Bus, but a while back they solved it with an astonishing innovation... a second door! You get on at the front and off at the back. Given that this has been common elsewhere forever, it's unclear why it took them so long, but...
(Bafflingly, they went through a transition period where ~all of the buses had two doors, but the driver rarely opened the back door. It wasn't really until covid that using the back door became standard. Improved things greatly.)
> and jockey for position and for a seat before the bus can start moving again
Do urban buses where you are require people to be seated? Didn't realise that was a thing anywhere. Any (urban, non-intercity) bus I've ever been on takes off as soon as the last person gets in.
A second door is good. Make it even better by people getting on and off from either - those getting off should be first. Many systems work this way and it works great. Trains oten have even more doors, but for a but that often isn't possible.
It's usually buttons in Europe. The cord things always make me think of train emergency stop cords (though these days those are usually "break glass" buttons).
But also, I don’t think even Google would claim that their LLM stuff can solve problems like this.
reply