It's a cliché that you wait ages for a bus then two come at once. That's because it's routinely true - left to their own devices, busses aggregate. I think the most I've seen in a row is three, or maybe four. People arrive at bus stops effectively at random intervals. Individuals in the know may arrive at a particular bus stop 'at the right time', but the total number of people waiting varies. And the longer it's been since a bus arrived, the more people are likely to be present. If a bus is delayed for any reason (or the previous bus was early), it picks up more people, which slows it down - both because they need to get on, and later the bus needs to stop and let them off, where maybe no-one wanted to get on. Conversely, if a bus is ahead of schedule, or it is behind a delayed bus, there are fewer people for it to collect and it speeds up until it reaches the other one. Busses often have difficulty overtaking each other, but even if the unladen bus does, it's going to start picking up passengers and slow down again.
I propose the following remediation: Suitable bus stops along the route are designated as transfer stops. These need to have appropriate space for two busses to park up without obstructing traffic, and space on the pavement for passengers to reorganise. If two busses become aware that they are close to each other temporally, they both stop at the next transfer stop. An announcement is made prior to this, to get people ready. Passengers are transferred, both directions. Passengers destined for the following, nearby stops[1] transfer to the later, rear bus. (Exceptions are made for disabled and vulnerable passengers who would find it difficult to transfer.) Passengers for the stops beyond that optionally transfer to the earlier front bus. The first bus sets off as soon as possible, displaying a sign "no pickups until <specific location>". In the first instance, that location would be the one after the transferred passengers[2]. For some bus routes, a shortcut may be available. Beyond that point, it reverts to normal behaviour. The second bus sets off when it has sorted itself out, and behaves normally.
[1] Number to be determined initially though experienced guesswork, later with empirical data for each route. [2] Experimentation would be useful to evaluate whether this is optimal.-- Loris, Dec 19 2024 https://en.wikipedia.org/wiki/Slip_coach [pocmloc, Dec 21 2024] Clever and useful. You should send this suggestion to people who manage bus schedules.-- Voice, Dec 19 2024 To quote Olga Diaz, " Where's the money going to come from? "-- normzone, Dec 20 2024 There's a slightly different practice where I live; if the following bus is within sight of the leading bus, the leading bus will refuse to stop to pick up passengers (it will only stop to let them off). This means that passengers waiting at the stop wait maybe 60 seconds longer, but the passengers already on the leading bus get to make up some lost time, and the passenger loads are slightly better balanced.-- pertinax, Dec 20 2024 The bus stops do not necessarily need to accommodate both busses at the same time, especially if the second bus is stopping at all stops.
That should cut the cost of implementation down to near zero.
I'm pretty sure the busses in Oxford used to use pertinax's pattern when I used them to commute (10+ years ago now).-- yamahito, Dec 20 2024 Yes the transfer is the one thing that lest this idea down. Apart from that it seems almost baked in certain places.-- pocmloc, Dec 20 2024 The issue I see with the "leading bus doesn't pick up passengers" strategy is that it doesn't make enough of a difference. The bus needs to drop off the passengers it has at many stops - and the implication is it's more heavily laden than the rear bus. And a bus refusing to pick up passengers once it's stopped is... well, that's just not going to happen. Apart from anything else, bus drivers already get too much hassle from the passengers, they couldn't possibly enforce that rule for long. So the lead bus gets slightly ahead, and the number of passengers onboard goes down slightly. But it can't keep that up for far enough to actually get back on schedule, and it has to follow its route exactly (in case anyone wants to disembark on any arbitrary stop). So at best this strategy means you get two busses travelling about one minute apart for the remainder of the route, or perhaps even the shift (in line with what pertinax describes). Very obviously sub-optimal. The "unloaded rear bus overtakes, then doesn't pick up passengers" strategy is somewhat superior, where possible. But it does run the risk of saturating the new rear bus (which is a disaster), and again the new front bus simply can't refuse passengers for far enough to make enough distance to replace the first bus in the schedule.
Stopping once to transfer passengers back is a big efficiency win overall - the lead bus can then miss many stops entirely, and possibly even take short-cuts. There's a hospital near me with a 'stem-loop' (shared between busses, taxis and cars dropping off/picking up patients) which it can take the busses at least five minutes to get round at busy times; avoiding that would be massive. I think many routes will have the odd section which is awkward and could reasonably be avoided.
The reason for transferring passengers forward is that it reduces congestion. The system starts to fall apart if the rear bus becomes full, and everyone starts to resent it. But the passengers who go forward are getting to their destination earlier, so it's a true win for them, and it costs very little extra, since both busses are already stopping. Obviously the passengers who go back are losing a little time. But probably not all that much over the transfer time, since they were only a few stops from their destination anyway.
Most bus routes I've travelled on have larger stops, or places where multiple busses can align every few stops. If there isn't one, but the drivers can communicate, then dropping off passengers to catch the following bus only would be a reasonable compromise.-- Loris, Dec 20 2024 OK so here's the modification that makes this idea world-beating. Have the fronts and backs of buses fitted with articulated connections like you get on railway carriages. If or when the rear bus catches up with the front bus, the rear bus docks with the front bus and they combine into a 2-unit bendy bus. The front bus driver controls the unit and the rear driver uses an electric cattle prod to herd the passengers into the transfer while the combined unit is travelling between two stops. At the next stop they decouple and the front unit zooms ahead on the non-stopping shortcut while the rear unit picks up and drops of passengers.
If the system was slick enough then they could decouple at speed to save the front bus slowing down at all.
If the roads in the route were sufficiently dualled then the two buses could couple side-by-side for much faster transfer of passengers.-- pocmloc, Dec 20 2024 I think it goes without saying that any innovation which allowed vehicles to reliably connect and separate while in motion would be phenomenal. The problem is, it would have a scope far beyond improving the efficiency of existing public transport methods, it would revolutionise everything.-- Loris, Dec 21 2024 Tagline?
" it would revolutionise everything "-- normzone, Dec 21 2024 I thought it would have something to do with specifically green busses but this is better.-- 2 fries shy of a happy meal, Dec 21 2024 random, halfbakery