When they were first rolled out around 2000, the Los Angeles Metro Rapid lines were the hottest thing, so hot that a famous system of branding (Rapid buses red, local buses orange) was developed around them. Rapid buses run long distances along major boulevards, stopping every half mile, while local buses run alongside them stopping every two blocks. I too was a booster of the idea at the time, and soon “rapid bus” products were appearing in many cities.
But of course, the branding distinction was about speed, which all motorists understand, as opposed to frequency, which they often don’t.
The first two Rapid lines (Wilshire and Ventura Blvds) had all kinds of great features. There were architecturally designed shelters, and the City of Los Angeles helped with signal priority. Then, however, the forces of envy set in. Rapids made sense only where:
- the agency could afford very high frequency (generally no worse than 10 minutes) on both Rapid and local buses, so that it was worth waiting for the Rapid even if the local came first, AND
- corridors were extremely long with long average trip distances, because you have to be going some distance for the speed advantage of a Rapid to be worth any added walking or waiting that a Rapid would require.
But once the first two Rapids succeeded, there came the cries of “why does their street get this cool thing and mine doesn’t?”. And while the two points above were good answers to that question on many streets, LA Metro was pressured to roll out Rapid lines all over the region, in places where they made sense and places where they didn’t. Some, like Soto St, were just too short for the speed difference to be valuable to many people. Others didn’t have the frequency needed for their speed to be useful, with some coming as infrequently as every 30 minutes all day. Most of them had nothing like the signal priority of the initial two, nor the distinctive shelters. The buses were red, though, so it looked like some cool thing had been spread across the region. (For more on this political dynamic, which I call the Fishing Pier Problem, see here.)
So the result was outcomes like this:
If you’re on Venice Blvd but between Rapid stops, as in this example, you could walk less and use a local bus or walk further and use a Rapid. As this shows, the difference in travel time isn’t enough for that to make sense. The Rapid is only three minutes faster for that distance, but you’ll spend six more minutes walking.
The upper blue bar shows that by combining the Rapid and local buses into a single line that runs twice as often (with fewer stops than the local but far more than the Rapid) the result is a shorter total trip, because of the shorter wait. In this case, the customer walks six minutes to a single line instead of four (because the local stops are a little further apart) but then waits half as often (because the two lines are combined) and rides a trip that’s a little bit faster than the current local (again, because local stops are a little further apart). It turns out that lots of people along these long boulevards are in this situation.
Combining Rapids and local into a single more feequent line is one of the key recommendations of the newly proposed Los Angeles metro bus network redesign, the work of our respected competitor Transportation Management & Design (TMD) working with Cambridge Systematics. Russ Chisholm of TMD, whom I used to collaborate, actually led the planning that created the Rapids in the late 1990s, so it’s fitting that he’s also gotten to plan for their obsolescence.
Here are the outcomes. (“Reconnect with our customers” is the no-growth redesign, the plan that reallocates existing service instead of adding new service. “Transit First” adds bus lanes and other infrastructure, for even more improvement without adding operating cost.)
The vast increase in the number of people with access to frequent service, from 900,000 to 2.15 million, is the key to why this plan is likely to succeed. A huge share of this outcome results from combining the Rapid and local services into single lines, since many streets that formerly had both a local and a Rapid every 15 minutes will now have a bus every 7.5 minutes or so.
As always, a redesign that doesn’t add more service involves cutting some unproductive service, but here only 0.3% of riders losing walk access to transit, which is also impressive. These are the least transit-oriented places in the region. Still, we can expect ferocious complaints. It may seem like 0.3% of the ridership isn’t much, but they and everyone they know, with some public relations skill, can make it sound like the plan is a disaster. Even if nobody were losing their service, some people will be angry when you change anything. So if you live in Los Angeles, it’s important that you engage with the plan!
That brings me to my main critique. In exploring the website, I found the plan difficult to learn about. There’s no shortage of materials selling the plan to me, and there’s no shortage of route-by-route details, but I wish there had been a report that makes the argument for the plan and explains the thought process that led to its design. (No, PowerPoint slide decks are not reports, because they don’t show the logical relationships between ideas; they are useful only with narrating voice attached.) The plan’s data viewer is pretty good, especially the tool that helps people see how the plan changes where they go. We do similar things on our projects and they should be standard procedure now.
But I can’t find much on the website that seems to be speaking to non-riders, including anyone who cares about outcomes that the plan improves (congestion, climate, urban redevelopment, access to opportunity, social justice etc etc). If you might support the plan for any of those reasons, the comment survey (a tab within the data viewer) will frustrate you. It assumes that you’re evaluating the plan only selfishly, in terms of whether it will improve your travel. (This also discourages feedback from non-riders who could see other selfish benefits, such as a business that gets better access from potential customers and employers, or a benefit for a friend or relative.) Getting these plans across the line requires selling a big picture to the biggest possible audience, especially given that some angry riders will be yelling. I hope that, in some forum that I can’t find on the website, that pitch is being made.
I wish LA Metro the best with this redesign. It looks great. It presents huge opportunities for better access to opportunity, more sustainable urban form, climate benefits, reduced local emissions, and safety. It deserves to be allowed to succeed.
I’m very much looking forward to the redesign and am OK with losing most of the Rapid Bus network if it means improving the network overall. That said, I also think dedicated bus lanes for any route currently used by Rapid Buses (and local buses that use the same routes) would be a vast improvement — as would celebrating their routes in the same way many writers/riders do Metro’s trains… which is why I wrote this piece hoping to encourage both.
It’s going to be disastrous. Look at the bus lines currently without a Rapid Line. The 14 is never on time & half the buses are packed like sardines while the other half have 3 riders. Metro has gotten gradually worse. Maybe instead of a redesigned system they need a change in management.
Dammit, the Metro Local buses are not “Orange” they are “California Poppy”!
Which California Poppy? Do you mean Eschscholzia californica or Papaver californicum? If you’re going to be precise, let’s be precise! 😉
So why are brand new busses now arving in Los Angeles painted Red. Nobody passed the word. They
Will have to be used on non Rapid lines confusing passengers even more. ex CEO Art Leahy wanted
To have all the buses painted Orange. He was overruled.
Thats because Art Leahy is a transit professional as opposed to the amateurs running the MTA.
To play devil’s advocate, let’s take a closer look at the Venice Boulevard example. Don’t compare the first and second lines. Compare the first and third lines.
The “NextGen” alternative that cuts out stops saves one minute of in-ride time (basically unnoticeable on a ride of more than half an hour), but adds two minutes of walking time for people whose starting and/or ending stops are eliminated. The “NextGen” features actually make the trip longer.
The real benefit between the first and third lines is from the increased frequency and reduced waiting time (which is exactly what Jarret has been advocating for years).
This might have to do with certain stops not being used too often. In the current local’s schedule, it might be that Metro assumes that the bus gets to skip a certain percentage of stops. With a high-frequency all-local service as planned, the increase in ridership would mean that the bus has to stop at more stops that it does right now, increasing the scheduled travel time. The time savings of the hybrid could be more in the range of 3-4 minutes when compared to a local service with higher ridership.
This is interesting. In Alon Levy’s blog post about the Queens bus redesign (that he mostly liked) the author had some relevant comments. While discussing the idea of having both an express and a local on the same route, he said this:
The off-peak frequencies are so bad that saving a few minutes by running express doesn’t actually save time counting wait time.
Later on in the comment thread, he repeats his idea, tying it back to how Americans like to “innovate”, but not in a good way (https://pedestrianobservations.com/2019/12/31/queens-bus-redesign/#comment-70537).
I agree. As I commented in that blog, a good rule of thumb is whether the bus is so crowded that service is oriented towards reducing crowding, instead of adding frequency. It is unlikely that an agency would increase frequency from 3 minutes to 2, just to reduce waiting. But it is fairly common — at least at rush hour — for agencies to add extra buses to deal with crowding. That is the only time in which it makes sense to have both an express, and a local. The express deals with the crowding more efficiently, while all of the riders maintain a high level of service frequency. Otherwise, just run one bus as often as possible, making stops somewhere around 400 to 600 meters apart.
It looks like L. A. is doing the right thing.
Don’t know whether this is the case in this particular situation, but if the road is chronically congested and the plan is to have bus lanes, an express bus cannot pass a local without merging into the car lane. If drivers don’t let the bus in, the express bus just gets stuck behind the local, so skipping the stops doesn’t actually save much time.
Of course, off-peak hours, this tends to not be the case. But, off-peak hours, the local bus will spend less time stopping, so there’s less to gained by having an express route that skips the stops.
Yes, and during the middle of the day, the express and local have poor frequency. This means that the rider of the express loses more to waiting than they gain by having the bus go faster (and that doesn’t count the extra time walking to the bus stop).
One technique that does work reasonably well is to have the express (which only runs during rush hour) go on a completely different path (such as a freeway). That is pretty common in the greater Seattle area.
That’s exactly what’s needed