I like the downtown-only Red Line trains that Metro's been running on weekends. Although I'm obviously biased because I live in that area, the NY Ave to Van Ness trains that have been running for the past few weekends are nice, because they've actually been very frequent.

(On the other hand, the Next Train displays are less than useless on these days. Some of the Van Ness-bound trains weren't showing up at all. I did notice that Metro finally programmed the signs at NY Ave to reflect the single tracking through that area, but didn't account for the fact that they were, in fact using both platforms at that station, or predict for the next departures of the Van Ness trains.)

by andrew on Jan 19, 2011 10:40 am  

Great suggestion! Part of the reason the London Underground is well-regarded is that it runs on a timetable, making it far easier to plan journeys and work around track disruptions. WMATA would be wise to adopt timetables for the weekend, and distribution would be very simple (online, on the big LCD screens being installed systemwide, etc.). Once people get used to the timetable, the system would rise in many folks' estimation.

I also see how timetables on weekdays could also be beneficial: additional trains would shuffle in easily, but there could be "core" trains that would allow for predictability.

by Rudi on Jan 19, 2011 10:48 am  

The Dutch railways run trains, at least on major routes, on 15-minute headways or better, thinking that people won't need the complexity of a timetable because a train will always be leaving soon. Then again, these are mid-distance trains covering the equivalent of Metro out in the suburbs.

by David R. on Jan 19, 2011 10:54 am  

I thought that Metro already has timetables on their website for weekend service and non-peak weekday service. Though I do agree that the timetables would be easier to interpret if they listed every station, and didn't get rid of the center of the schedule by making you figure out the times for an "every 6 minutes" frequency, not to mention that it ignores any issues about maintenance that could mess it up. But I think this post would be more useful if it didn't ignore that there are already weekend timetables, and instead discussed specifics on how those timetables could be better.

To andrew (first commenter) I agree that more frequent downtown-only trains are helpful, but I'd opt for them terminating at Grosvenor and Silver Spring the way they do on weekdays. Mainly because my station is Tenleytown, so it really stinks when they terminate at Van Ness.

by Douglas Bell on Jan 19, 2011 10:59 am  

Seriously, what qualifications do the people writing this blog have when the majority of the posts here about how to improve Metro are completely impractical ideas? Each single track zone requires a different timetable, you can't create a "standard" weekend schedule under the assumption that the schedule from week to week is going to vary, and it usually does. For the most part, you never get more than a week or two of the same weekend service pattern since each trackwork project moves on its own timeline, so the idea of paper timetables (which are already available online to depict "normal" service and have been available for all times of day except rush hour for over 10 years) should be dead on arrival.

What Metro could do is provide better information in advance of schedule changes that create headways longer than 20 minutes and develop a service advisory page similar to the one that the NYC MTA has that is better organized and easier to follow than the information buried in wordy press releases. I've been caught on a Metro platform for FORTY MINUTES due to trackwork (the operator I spoke to said the trains were coming every 30, the form letter response I got back from Metro when I complained said their website said "20 minute delays" so I guess that meant 20 minutes extra on top of the regular 20 minute headway at that time but who really knows). And the trip planner has started to incorporate service changes into its result, so while it was a great suggestion to make, the GGW bloggers were a bit late on that one. Oops.

Keep your day jobs, bloggers...

by OTP on Jan 19, 2011 11:00 am  

The Van Ness-New York Avenue trains are the regular Grosvenor-Silver Spring trains, truncated to continue providing a high level of service betwen the single track zones. They would go to their regular terminals if they could get there without creating a bottleneck up and down the line waiting to get through the single track zone.

I'm glad I'm not the only one here who knows that Metro has been issuing timetables for years. However, you wouldn't want a timetable that listed each train when it runs every 6 minutes from 10 AM to 8 PM for the entire system, that is a total of 100 trains on each of the three main trunk lines. You want the timetable to be easy to use or do you want something resembling the even 30s bus schedules?

by OTP on Jan 19, 2011 11:04 am  

I really have to think perhaps the answer is just shutting down lines altogether on weekends.

The weekend trains that I do see seem to be pretty full, so I think they are getting a decent demand. However, between overtime, usual labor costs and electricity they might save money by shutting down the orange/blue line altogether rather than single tracking.

by charlie on Jan 19, 2011 11:13 am  

@OTP:
You probably don't care about this since it was created and is maintained by a blogger, but there is an easy way to find out ahead of time about scheduled trackwork.

http://greatergreaterwashington.org/trackwork.cgi

It's a nice calendar format, which you can access from any computer or smartphone.

As for the service disruptions, as far as I can tell, Metro's Trip Planner is now only incorporating schedule disruptions for major closures, not all work. Regardless, a "service alert", like the one I proposed last week might still give someone enough information that they leave earlier or take an alternate route.

http://greatergreaterwashington.org/post/8851/metros-trip-planner-should-offer-service-alerts/

But you clearly think this is a bad idea, so feel free to oppose any innovation from outside the agency. Clearly no one who does not work for Metro has any good ideas.

I do agree with you, however, that the idea of a weekend timetable specific to individual trackwork is not particularly feasible. The exigencies of trackwork would make it unreliable anyway.

But it would be great if through scheduling or use of OCC Metro could implement timed transfers during trackwork. It's very irritating to get off the Green Line and get upstairs to discover the Red Line left 30 seconds ago, and it's 20 minutes until the next train.

by Matt Johnson on Jan 19, 2011 11:28 am  

That would require Sarles to actually do his job. It will never happen.

by Redline SOS on Jan 19, 2011 11:35 am  

charlie- fyi- people depend on Metro as their primary means of transportation, and this included weekends. So, " just shutting down lines altogether on weekends" is not the answer, as you put it.

by Matt on Jan 19, 2011 11:50 am  

On the timed transfer topic:

I've often wondered if Metro could do a better job of easing transfers where lines converge and it's popular to switch from inbound on one line to outbound on the other (esp. Rosslyn, L'Enfant Plaza green and yellow). Since all the transfers would be doing the same thing (e.g. inbound Orange to outbound Blue or vice versa), it would be helpful if such train pairs were scheduled to provide minimal wait times. It wouldn't have to be exact, just close. Is there a reason this wouldn't work?

It might be difficult to time transfers at the major downtown stations, though, since there are many different combinations of trains riders might be taking. But it would at least make sense to hold trains in the station for a minute if they arrive close together and the headways are long.

by RichardatCourthouse on Jan 19, 2011 11:52 am  

Your calendar app doesn't do anything beyond what the Metro site does, aside from a format change. Hopefully someone finds benefit from it, but the issue isn't only the presentation of the information, but also what information is (or isn't) being made available, so it doesn't solve the issue I brought up. But nice work nonetheless.

Timed transfers would be nice, but even on normal schedules with no trackwork, the Red Line comes every 15 minutes and all the other lines are every 20 to create a 10 minute headway in the core. Either the Red Line would have to be cut to every 20 minutes, increased to every 10 minutes, or the other lines would need to be increased to every 15 minutes to have timed transfers. In this financial situation, I don't see that happening unfortunately. Throw in trackwork projects, and unless they are taking place at the far ends of the lines, I'm not sure there is a way to reliably provide timed transfers. Something has to give, the issue with what I've read here and on other Metro advice sites is that no one is willing to give up anything like timed transfers AND they want the system maintained and low fares. It might work in the blogosphere, but it doesn't work that way in reality, unfortunately.

I never said including more information on lines affected by trackwork related changes was a bad idea, only that Metro is already a step ahead of you by updating the trip planner to reflect trackwork related changes. I don't play with the Metro trip planner each week to see if each and every project is incorporated, though I do use it to plan my own trips. And it has been including NY Avenue turnbacks for awhile now, and also had the closures around Stadium-Armory from this past fall, too. Metro also used to issue timetables special for major trackwork with the departure times for each station, though back then the trip planner wouldn't include the updated times.

I don't oppose innovation from outside of Metro and I'm not going to try to convince anyone that the past few years are the brightest spots in Metro's history. I'm all in favor of reasonable innovation, but most of the ideas proposed here are not reasonable, don't acknowledge the reasons the prior decisions were made and/or just pass those past ideas off as stupid, shortsighted, or negligent, and the mentality of what works in X must work in DC is prevalent in many posts and usually wrong, which is why the historical decisions were usually made in the first place (how long have your writers lived in the DC area on average? It seems the institutional memory on this blog doesn't go back more than a couple years, which is unfortunate in my opinion). But hey, you all are allowed to blog and offer your opinions, and as a long time DC area resident who has been riding the Metro for my entire life (and with an urban planning degree and a job in the transportation field, but not with WMATA, and who knows quite a bit about WMATA's history and operational practices despite not working there), I'm entitled to challenge your opinions and offer mine instead. Seems fair to me at least.

by OTP on Jan 19, 2011 12:09 pm  

I was amazed during my first subway trip in NYC when I learned that the express was waiting at the platform until the local arrived. "What?!" I cried, in disbelief. "You mean, the driver actually waits for the local riders, so they can transfer quickly? This can't be real!" My years of WMATA experience wouldn't let me grasp the concept. Imagine, courteous train drivers in DC helping riders get to where they need to be, on time and professionally.

If only, in my wildest dreams, the Red Line upstairs at Gallery Place would pause, for just a minute, until the majority of the downstairs riders made it upstairs to transfer. Sigh.

by OX4 on Jan 19, 2011 12:39 pm  

MetroBUSes have a schedule; they rarely keep to it. Why would the MetroRAIL be any different?

by Vicente Fox on Jan 19, 2011 12:48 pm  

@OTP:
You are certainly welcome to express your views here. In fact, I encourage you to do so. Nothing in my earlier comment should be taken to mean that you should not challenge the ideas presented here.

I think you'll find that not all contributors agree with each other. Sometimes one of us thinks something is a good idea, while others don't. As I pointed out in my comment above, I don't think Vincent's idea is feasible. But he approached GGW with it, and we agreed to publish it.

If you have ideas you would like to have a forum for, feel free to contact us by email. We're always looking for new contributors and new viewpoints.

I, for one, understand why Metro does some of the things it does. Other things, I don't understand. So if someone suggests what they think is a good idea, everyone appreciates a rebuttal that might include some history about things. But that doesn't necessarily mean the idea is a bad one.

You're right that my institutional memory doesn't go back very far - at least not with Metro. It's true that I've only been a regular rider since mid-2007. But I was a regular rider on WMATA's sister system, MARTA, for 4 and a half years before that. And an intermittent rider before that.

And, like other contributors here, I'm far from a novice when it comes to transit. I've ridden every heavy rail system in the United States, save 1. I've ridden almost three-quarters of light rail systems and all but 3 commuter rail systems. I've ridden rail and bus systems in Canada and across Europe.

Other contributors have relied on Metro for years. Or on BART. Or New York's MTA. Or Cleveland's Rapid.

So sometimes, we see things that make sense for other transit agencies, and we suggest them here. Sometimes we do the same trip on Metro countless times on end and notice something that would make it better.

We each hope that by sharing something here, that we will effect positive change. That we will make our Metro better. Our region, too.

And yes, sometimes our ideas just won't work. Sometimes they're just not feasible. Something that works in Charlotte just might not work in Washington, and that's okay. Even if it did, Metro might never choose to adopt it.

But that doesn't make the mere act of suggesting it a waste of time. The point is, at least partially, to stimulate thinking about the subject and a healthy discussion of it.

So it hurts - and I mean that - when someone implies that we rarely have a good idea. That we should "keep our day jobs" - that we shouldn't bother. That you wouldn't miss us if we disappeared.

Yes, it hurts. And it's disappointing.

I do, in fact have a day job, which I have no intention of quitting. And that means the work that I do writing for this blog and editing its content comes out of my "free" time and my sleep cycle. It is a passion of mine that I do not, even in the least bit, regret.

In fact, like you, I work as a planner. I have a background in transportation planning. And while I don't work for a transportation agency at the moment, I have worked in transportation planning in the past. And I am not alone.

Several of our writers have degrees in planning or in related fields. Most of us have day jobs, sometimes related to what we write about, sometimes not.

But having a degree or a job is not a prequalification to write for Greater Greater Washington. It certainly helps, yes, to speak the language of planners. But what is more important is to speak the language of the lay-person.

I sincerely hope that through our work, we have helped to enlighten people who otherwise wouldn't have known how something worked or why it existed like it did. I hope that our efforts have led to positive change in the community, in the region, in the nation.

And I hope that sometimes our ideas will lead to real solutions. And if sometimes one doesn't, that's okay. At least we'll have tried.

I'd like to invite you, and anyone else who is interested, to submit something. We love counterpoints and new ideas, alike.

Thanks for reading. And thanks for offering your opinion.

by Matt Johnson on Jan 19, 2011 12:51 pm  

First of all let me thank you all for your comments. Whilst I do agree with some, I unfortunately have to disagree with others. Such is life in the Blogosphere.

Just to pick up on some points:

Timetabling: Reshuffling timetables to reflect service disruptions can be done through several means: Slowing down trains, longer turnaround times, adding an extra 30 seconds at each stop… Single track operations should not for example involve 20 minutes of idling just before a single track section as a consequence of a signal not being clear. But that involves planning and therefore a timetable.

Also one should not forget that weekend service usually follow the same pattern of service during the day. Therefore what will work between 9 and 10 AM in the morning on a Saturday will work just as well between 6 and 7 PM in the evening on the same Saturday. That makes the job of timetabling ahead a lot easier and way more feasible.

Transfers: If a timetable was put in place as I suggested, I would not expect every train to be able to provide connections to all other trains running on different lines. Some careful balancing would need to be done between operational constraints and observed weekend travel patterns.

Finally let me repeat my point: Weekend service should not be a quest for speed and/or frequency, it should be a quest for reliability and predictability.

by Vincent Flament on Jan 19, 2011 2:24 pm  

Metro repairs should be done at night when the system is closed. How many workers have been killed or injured in recent years while working on the tracks during operating hours? A very simple concept.

by jps on Jan 19, 2011 3:20 pm  

Personally, I'd be just fine if the PIDS worked. This past weekend, the sign at Rosslyn was completely messed up; every train was listed as "No Passenger" and arrival times would go way up, way down, or spontaneously disappear. There's not as much need for predictability if you can get online before you leave and check arrival times.

Timed transfers would be tricky to pull off. Let's say that all passengers can get to another platform in three minutes. So if a Red Line train gets to Chinatown at 10:00, Yellow and Green Line trains shouldn't get there until 10:03. Meanwhile, the Red Line train gets to Metro Center at 10:01, so the Blue and Orange Line trains shouldn't leave there until 10:04. But if the Yellow and Green Line trains get to L'Enfant at around 10:06, the Blue and Orange Line trains will be late by a few minutes, causing a 10+ minute delay. If the Yellow Line train waits, then it will take longer getting to the Pentagon to meet another Blue Line train, etc. And this is just looking at trains going in one direction. I'm sure there's software that can set up this kind of thing, but track work and boarding delays (i.e., during sporting events) could throw the whole thing off very easily.

Stating headways for certain periods of time may work better. For example, Metro could explicitly create a policy of less than five minutes' wait during rush hour, less than ten at other times during weekdays, and less than fifteen on weekends. Shorter headways during events and longer ones during major track work could be publicized.

by jakeod on Jan 19, 2011 3:21 pm  

@OTP: "what I've read here and on other Metro advice sites is that no one is willing to give up anything like timed transfers AND they want the system maintained and low fares."

A common sentiment on GGW last year was that fare increses were greatly preferable to service cuts; I even participated in an online petition to that effect either directly on the blog or via a link from it. In any event, enough people expressed this view emphatically enough that Metro did indeed chiefly raise fares instead of cutting service.

by davidj on Jan 19, 2011 4:13 pm  

@OX4
It is nice when the subways stop to allow transfers, but two things to note:
* the trains are usually adjacent to each other (for example, the 1 and 2/3)
* waiting is hardly universal. It is nice when it happens but they don't always wait.

by movement on Jan 19, 2011 5:02 pm  

Terrible idea.

This is supposed to be a rapid transit system serving a dense city, not some 9-5 commuter railroad.

Metro should be constantly criticized for running headways over 15 minutes at any hour of any day. This is DC, not Miami. We shouldn't accommodate their ineptitude.

Maintenance should be done when the system is closed....just like Chicago, Boston and Philly do. Not forcing people trying to get to dinner at 6pm to deal with 20 minute headways.

Metro is already competing against minimal traffic and cheap parking on weekends. Adding enormous waits only makes people even more likely to drive.

If absolutely necessary, shut down the system at 10pm on Sundays to get some extra work in.

by JJJJJ on Jan 19, 2011 6:13 pm  

@jakeod I agree, just make the PIDS system work when they are single tracking...how exactly does that system work does anyone know? Do they have sensors at every station or is it based on when a train leaves the end of the line or some other designated point?

IMHO a timetable will never be accurate especially when you have a track repair project going on...sometimes Control needs to hold trains and stop or slow them going through the work area.

by Chris R on Jan 19, 2011 7:43 pm  

Many of us would love to have more funding so that Metro could run at rapid-transit frequencies all weekend, and they could afford the extra costs inherent in doing all that trackwork overnight. Given that that has not occurred I am really glad to see this post.

I don't think it would be that hard for Metro to be more user friendly for weekends with trackwork. It would, however, require Metro to be more flexible and customer-oriented than it has been so far. This change in thinking, as GGW has documented in many areas, is a real challenge for WMATA.

It is a relief that the Trip Planner supposedly is taking into account track work. Apparently it is also completely new. Better late than never, but it says something that this step took so long. However, the larger point remains - Metro should make it easier to know when a train is coming on weekends, and they should design their schedules on weekends so that trains can stick to the schedule (ie with some padding somewhere).

While I am at it, I think Metro also has to be more flexible about operations: Not just always run trains on the same routes come hell or high water. Even the Van Ness - NY Ave trains seem like evidence of flexible thinking to me! Previous posts on GGW have pointed out that Metro should run express trains after a substantial service interruption. This requires Metro to trust employees to make decisions, and communicate those decisions to the public. But it would be most worthwhile.

Metro ought to be flexible about routes during outages like this past weekend. Running both Blue and Orange lines separately from Vienna and Franconia to Rosslyn was completely non-useful. Why not run Blue line trains north to Rosslyn, then turn them around and run to Vienna? The trains had to reverse out of Rosslyn anyways. This would have given riders a much quicker ride into downtown, via Pentagon and the Yellow line, than the thoughtless set-up they used last weekend with shuttle buses. But, it would have required some flexibility from Metro.

by DavidDuck on Jan 19, 2011 7:55 pm  

I like the idea of timetables, but they're not absolutely necessary. It'd be nicer to see "-- Train" changed to something more definite.

@ Rudi:

The London Underground does not operate on a timetable. TfL's timetables are approximate frequency guides, and service is so high on most lines that a timetable would be useless. In reality, only NYC (usually not according to any schedule) and BART have timetables in terms of US rapid transport systems.

@ OX4:

That's because the local-express transfers occur cross-platform, that is, one simply goes to the other waiting train. In many cases, conductors do not wait, especially at rush hour, so as to avoid overcrowding. There is no reason that the Metro would need to adopt such a strategy, in part because there are no cross-platform transfers (yes, I'm counting Rosslyn and Pentagon) in the system. It should, however, figure out a way to reduce getting knocked around by people running to catch a train on the upper levels of transfer stations.

by Phil on Jan 19, 2011 8:24 pm  

Actually, the NYC conductors don't decide on their own whether to hold for a connection. Dispatchers indicate to the trains to hold in the stations by illuminating a set of amber lights above the platform. Once those lights go out, the train closes up and moves on. Since the dispatchers in the towers who decide whether to hold the trains often only know the approximate location of where the train is as opposed to its exact location, while many times connections are made successfully, there are also many times where the connections are just missed because the dispatcher turns off the holding lights too soon. Also, it is standard procedure that trains not be held for connections during rush hour unless the schedule calls for a particular connection.

by OTP on Jan 20, 2011 7:17 am