Interlining: Yes or No?
 

I've always felt that when it comes to rapid transit (i.e. grade-separated, electrified), interlining is a great way to double frequencies along shared route segments (two tracks) and is an easier, quicker, and more cost-effective approach to expanding rapid transit systems (as opposed to building single lines from scratch) because of the existing ridership base.

The downside of interlining includes reduction in the quality of service, with trains not being able to run as frequently, schedule delays more likely, and disruptions caused by construction.

In LA, the best case for interlining involves a potential heavy rail extension of the B (Red) Line down Vermont from Wilshire to 120th Street (and eventually farther south). Vermont has the second-highest bus ridership after Wilshire, so I see value in trains arriving every 2.5 minutes along the shared spine and 5 minutes along the branches during peak periods.

https://live.staticflickr.com/2825/3...cb2356f0_o.pnghttps://live.staticflickr.com/3886/3...282a87a1_o.png

I'm a fan of interlining. It gets a bad rap by a handful of transportation academics and thinkers, especially one in particular, but I think the benefits far outweigh the operational flubs that can, and admittedly do occur. Of course I'm really just thinking about NY here.

Which one in particular is so critical of it?

Alon Levy has written extensively on all aspects of interling, de-interlining, reverse branching with a focus on the NY subway.. it's pretty in-the-weeds with lots of numbers, but is highly researched and has a strong position even if it's pretty theoretical. I obviously don't agree with all of it but a case is definitely made and is worth reading for anyone highly interested in rapid transit operations.

Can automation and platform screen doors attenuate operational delays?

The Vancouver Skytrain's Millennium and Expo Lines once interlined from Waterfront to Columbia. Today, the "interlining" isn't in the form of two lines, but rather a singular line (two actually — Expo and Canada) with two branches.

Interlining makes the most sense when the spine is the major source of the ridership and the branches have relatively equal demand.

Thanks I'll check it out, sounds interesting.

Interlining is combining two routes into one, even though they appear as two completely separate and independent routes on the map, they actually use the same vehicles (e.g. the bus of route 1 arrives at the terminal, and then it becomes a bus of route 2 when it departs the terminal). What this thread is actually referring to is branching routes (e.g. route 1/1A/1B).

Problem with interlining is it is basically creating one very long route, and longer routes are unreliable unless there are rapid transit measures (ROW, signal priority or grade separation, longer vehicles, trains, all-door boarding). There is no problem with branching routes if they have the same rapid transit measures as the main route. A branching route might actually improve reliability if it is shorter than the main route, as length is one of the main reasons for unreliable service.

In the example posted above, the blue route would not be adversely affected by the red route, but the red route could be adversely affected by the blue route since the blue route is longer. However, if these are both heavy rail, meaning all rapid transit measures (including long trains and grade separation), there is little risk of delay. Bus and railsroutes typically have scheduled layover time of a few minutes in each direction which further help them to stay on schedule. If the blue line has enough layover time, then if it is a true rapid transit route, it should not interfere with the operation of the red line if the red line is also a true rapid transit route.

just bring that dam red line up to los feliz will ya? thats where i usually am.

There are already stations Vermont/Sunset and Hollywood/Western.

Flying crossovers make a big difference in reducing interlining problems.

At-grade junctions unintentionally cause delays for inbound trains, which can then cause bunching in the absence of strict signaling.

For example, in Boston, you will sometimes see no trains inbound from the Kenmore junction for 10+ minutes, then a parade of 3-4 trains in a row, sometimes with two from the same route back-to-back. This is caused by delayed outbound trains blocking the at-grade junction. Inbound trains that don't have to cross the active outbound track are free to continue moving inbound, which inevitably favors the right-most route.

Interlining really depends on the principal corridor frequency and the mixture of services on branches; most commuter networks around the world operate on this principal where the lower frequencies limit the number of conflicts, but with more heavily used rail systems, and particularly metro systems, this becomes more problematic.

London has quite a few sections of quite complicated interlining, but there has been a gradual rollout of digital signalling to mitigate against this problem, whilst improving journey times and frequencies. Ideally to achieve very high frequencies (>30tph) you really need to segregate or simplify the level of interaction. That happened with the Circle Line and will happen in the future to the Northern Line (it will be split into two lines).

I think that interlining can work, but it really depends on the route/frequency.