3 ms·

There's a reasonable argument for busways over light rail -- the non-disaster failure mode of buses is a lot more reasonable than the non-disaster failure mode

by jowiar 11y ago

There's a reasonable argument for busways over light rail -- the non-disaster failure mode of buses is a lot more reasonable than the non-disaster failure mode of rail. If a bus has mechanical issues, it doesn't stop the entire bus system. With rail -- a bit of engine trouble breaks the whole system. I've seen them work reasonably well in Pittsburgh, along with the Bus lane going opposite traffic on the one-way streets (as opposed to say, Philadelphia, where everyone drives in the bus lane).

sologoub 11y ago
I haven't seen them work out too well in San Fernando Valley. They seem to be less used than I would have guessed.
- mixmastamyk 11y ago
  The Orange line is packed at rush hour, not sure what you're expecting?
m_mueller 11y ago
The problem with buses IMO is the planning - most of the time planners take the cheap way out and just let them use the same road resources as cars. IMO a good local transport is completely separate from the roads used by cars, such that congestion on the car road network doesn't affect the second system - i.e. you always get a real alternative. Having one lane for buses, taxis and ambulances would be a pretty good solution, but that's usually not how it works out. Also, you have to look at the whole system in terms of how much surface space it needs - in any reasonably dense city you soon start to need to go either underground or stilted. Railways have a much narrower footprint due to their efficiency - so if you can make them reliable (which is clearly doable, see Japan and Switzerland as examples), it's just a more space saving solution, thus you need to tunnel / stilt less.
dredmorbius 11y ago
With rail -- a bit of engine trouble breaks the whole system. That depends on how you engineer the system. With reasonably frequent crossings, you can route around stalled vehicles. If you've got dedicated rights-of-way (rather than mixed-mode auto-and-LRV routes) then wrong-side operations isn't so critical. With sidings or triple-tracking, you can offer only modestly degraded service, though at markedly higher initial cost and land use. Still far better than highway / surface-street designs. Or you can "bounce" vehicles to either side of an outage, running service to the problem location and running truncated shuttle service on either side. It's largely a matter of political will, costs, and frequency of outages. Which, generally, even on less-reliable system aren't exceptionally prevalent. The other issue is that busses tend to fail (for various reasons) more frequently than rail.

by jowiar 11y ago

sologoub 11y ago

I haven't seen them work out too well in San Fernando Valley. They seem to be less used than I would have guessed.

mixmastamyk 11y ago
The Orange line is packed at rush hour, not sure what you're expecting?

m_mueller 11y ago

The problem with buses IMO is the planning - most of the time planners take the cheap way out and just let them use the same road resources as cars. IMO a good local transport is completely separate from the roads used by cars, such that congestion on the car road network doesn't affect the second system - i.e. you always get a real alternative. Having one lane for buses, taxis and ambulances would be a pretty good solution, but that's usually not how it works out. Also, you have to look at the whole system in terms of how much surface space it needs - in any reasonably dense city you soon start to need to go either underground or stilted. Railways have a much narrower footprint due to their efficiency - so if you can make them reliable (which is clearly doable, see Japan and Switzerland as examples), it's just a more space saving solution, thus you need to tunnel / stilt less.

dredmorbius 11y ago

With rail -- a bit of engine trouble breaks the whole system. That depends on how you engineer the system. With reasonably frequent crossings, you can route around stalled vehicles. If you've got dedicated rights-of-way (rather than mixed-mode auto-and-LRV routes) then wrong-side operations isn't so critical. With sidings or triple-tracking, you can offer only modestly degraded service, though at markedly higher initial cost and land use. Still far better than highway / surface-street designs. Or you can "bounce" vehicles to either side of an outage, running service to the problem location and running truncated shuttle service on either side. It's largely a matter of political will, costs, and frequency of outages. Which, generally, even on less-reliable system aren't exceptionally prevalent. The other issue is that busses tend to fail (for various reasons) more frequently than rail.