A Tour of Muni’s Control Center
Photo by phrenologist
Awesome person and Muni rider @cripsahoy, who writes A Streetcar Called Taraval, today describes her recent tour of Muni’s LMC (Line Management Center). Among other interesting tidbits, we liked this:
The trains are managed by one employee with the lines separated across four LCD monitors. He has a printed schedule and some more reference materials on the desk in front of him. As he watches the vehicles cross the screen, he calls operators and tells them to turn around trains and answers calls from stations about other train timing issues.
Read the whole post at A Streetcar Called Taraval and learn a little more what it’s like to manage the mess we call Muni.
Oh PLEASE. What a bunch of MTA generated tripe.
Myth: “Running a three-car train is not a realistic solution to easing transit woes.”
Truth: MUNI ran 3-4 car Boeing trains in the subway. It can and does help. MUNI spent extra money on fancy pants couplers that would allow them to make/break trains very quickly. These fancy couplers were part of the reason you never saw mixed Boeing/Breda trains a decade ago. Boston stuck with a common design and could thus mix their Bredas with their other rolling stock. In fact, 3-4 car Breda trains do not happen because a.) 3+ car Breda trains have caused overhead wiring damage before and b.) there typically aren’t enough drivers at West Portal or Duboce to make this a significant speedup. IOW it is realistic to run fewer, longer trains, but labor issues prevent that.
Myth: “You could probably get out on the first door at 30th stand there and have the last door of the third car open in front of you when it reaches 32nd!”
Fact: MUNI already operates two car trains at stops where the stops are quite a bit shorter than the train. Take a look at Sunset or 22/23/24th and Taraval. They leave passengers hanging out in the middle of the intersection /all the time/.
Myth “The MUNI Metro runs on automatic mode in the subway on what is called a Moving Block System. The vehicle must maintain a minimum distance from the one in front of it. That’s why you get to know the inside of the tunnel at Van Ness so well”
Truth: All train control systems will have some sort of minimum following distance. A fixed block system uses only one size block, and this requires that the block (which is for all intents and purposes the minimum following distance) be as long as it would take the train to stop. A moving block system will allow /much/ shorter distances (say a couple feet instead of a quarter mile). The reason you get to know the inside of the tunnel so well at Van Ness and West Portal is because the train control system that was chosen by Willie Brown’s consultants at Booz Allen Hamilton can only “park” one train in the station at a time. At the time we bought into this system (SelTrac), it had never been used on anything other than a small, driverless system. Sometimes you’ll get two trains entirely in the station, but the train control system cannot handle this situation and only opens the doors on the one train. The folks at the LMC /could/ open the doors on the second train, but they /don’t/ for reasons unknown.
Same thing at West Portal. Standard operating procedure would be to go out of auto mode so that trains could double berth and speed up the outbound process. Officially you needed authorization from central control. Because CC would rarely respond in a timely manner, SOP became authorization via a wink and a nod. After that last wreck (not due to the dangers inherent in manually operating a train, but rather due to a morbidly obese driver with uncontrolled diabetes driving a train… who had no business being behind the controls), this no longer happens. Thus you’ve got an extra 5-15 minute delay getting through West Portal outbound.
Myth: “The only places they can really turn them around is out in the Avenues.”
Truth: There are crossovers at Embarcadero, Castro, (both before and after) West Portal, and SF State. There was also another turnback built near St. Francis Circle, dunno if it was temporary or not (but it was used DAILY while SFC was under reconstruction). The only reason most of these are not usable is because the MTA simply does not maintain its equipment. That’s the whole reason the T was interlined with the K (thus killing any prospect of two car K service). Turning that many trains back at Castro was just too problematic. Despite spending millions of additional dollars on these turnbacks in the subway (keep in mind the Market Street subway is only forty years old), when we procured these new Breda trains, no steps were taken to ensure that they’d work with the existing infrastructure. The Boeings were built for the esoteric requirements of Boston and San Francisco, and they worked fine (well as fine as an upside down helicopter ever could). The Bredas were just not built to spec, and nobody at Booz Allen Hamilton or MUNI saw fit to keep the Italians in line.
You can dig through the RescueMuni archives to find out why the SFSU turnback had been mothballed for seven years. Cliffs: the NIMBYs dictated a complicated series of crossovers with signaling equipment neither supervisors nor drivers understood resulting in a series of numerous derailments.
Myth: “If there were many empty trains carrying 3 people to the beach”
Truth: Oh COME ON. They dump out packed trains all the time with no outbound service in sight. This is a fallacy promoted by the MTA to make themselves look good. What good is it to have service in the tunnel if that service won’t get you to where you’re going? Not only that, but you’ve just deprived all of your inbound customers (nee riders) of any sort of service. Because we all know that MUNI will go to the affected stops and let people know that they’re turning back trains early and they should walk to the next usable stop.
Myth: “The LRVs are especially susceptible to delay for a few reasons. ”
Truth: They just lowered the speed limit west of Castro both inbound and outbound to a walking pace rather than repair the tracks. If you think that surface traffic is a problem, think about the problems that forcing the trains to crawl at 5mph vs 35+ will introduce.
The problems that exist today at the MTA are entirely self-inflicted. There are better solutions, even now, but the MTA refuses to do anything but stick its head in the sand.
First, Thanks Alex for the fact check. I would add That specific to the Judah there is also a crossover near Arguello. The major problem in the tunnel is exacerbated by the Embarcadero shuffle. Because Muni is unwilling to dynamically resequence trains to maintain service on all lines, the platforms artificially fill up with riders waiting for the delayed route(s). This increase station dwell thus backing up following trains. Compounding these delays, automobile interference @ 4th & King slows both Ks and Ns. If Muni pushed the trains along faster they could carry more riders as the trips would be shorter.
BTW, the crossover @ Castro was used for a shuttle previous to the opening of the T. As to the time it takes to effect a turnback, here again, Muni sabotages itself by deliberately slow porocedures. Functional transit systems figured out better methods de4cades ago. Having the outbound (new) operator board the rear cab of an inbound train @ Emb so as to be immediately ready to reverse direction is way faster than walking the train length. The previous driver can then stepo off as the now outbound train loads up/. Simple, no?
I’ve spent a great deal of time waiting at the inbound platform at Embarcadero and have come to the simple observation that much of the downtown logjam comes from the incredibly slow turnback process. Without even getting into David’s excellent proposal above, the operators involved could simply move at a normal walking pace, or gasp, maybe with a bit of urgency, during rush hour when dozens of trains are stalled in the tunnel. Typically during the peak-of-the-peak rush hour, a train arrives, all passengers exit, the operator slowly gathers his/her belongings and walks through the train to the rear end, a new operator finishes up his conversation with buddies and saunters to the rear end of the train, exchanges greetings with the departing operator, slowly walks through the train, stows his bag, and finally shuts the doors. Then there’s another slow switch to the other end of the train at the turnback point at the end of the tunnel.
Simply ensuring that an operator is standing in place ready to go when trains arrive would be a huge improvement and doesn’t seem like too much to ask. The operators can change places while passengers exit, reducing the absurd dwell times at Embarcadero inbound.
Having operators available would also mean that there wouldn’t have to be shuttle buses down Taraval(!), and it would also mean that the MTA could make/break trains at West Portal to speed things up.
Of course, that’s a labor issue… and when the drivers are fighting tooth and nail to prevent the MTA from receiving any federal funding, well, there are obviously bigger fish to fry (or fire).
You’re welcome. It’s frustrating to see the MTA using someone like Katie as a pawn to promote their own dysfunction. As a member of the Citizens’ Advisory Council, I’d hoped she’d be less of a pawn and more of a check or balance.
And, quite frankly, that’s why I tried to avoid posting any specific /solutions/. The information is out there, it’s readily accessible. The problems as well as potential solutions are all very well known. It’s the inmates running the asylum at the MTA, and if the CAC can’t function even as an impotent balance… well I’m just glad I still own a car.
That said, I will point out one additional problem that I see too often: running broken trains either in revenue service or in the tunnel (or both). If the MTA would like you to believe that routine switchbacks in the subway would introduce horrible delays (and they did with the T), what do they think spending /ten minutes/ at the Embarcadero outbound platform trying to get a train to communicate with the train control system?
Yes, that’s right folks. May 10, 2011, the MTA held up outbound service at 5pm for /ten minutes/. The very same people who are so afraid of introducing any delays into the system. FFS.
BART gets this right, and their headways are nowhere near as short as MUNI’s (MUNI runs trains much closer together which just invites a larger ripple effect). If there’s a problem that takes more than a few minutes to solve, BART empties the train and it’s deadheaded (taken out of service, not stopping at any station) to the yard or a pocket track.
WIth MUNI, in contrast, I got on an inbound L @ 46th with faulty brakes where the service tech followed it in his truck, and reset everything at every stop instead of breaking up the two (incompatible, apparently) car train, or putting both cars out of the way near the Riptide or on the elsewhere along the way.
Likewise I’ve been on a train that was (presumably accidentally) turned back (inbound to out) at Church station(!). Even with only one driver it didn’t take any longer than a normal stop. And, if memory serves, the Castro crossover was designed /for the Breda trains in the first place/.
There have been too many missed opportunities to do the constructive thing (ex: West Portal intersection… which was ripped up fairly recently). Far too many…
The Castro crossover is used routinely by S trains and that process seems to generally work fine.
Out of curiosity, does anyone have any data on metro performance in the subway? I’m wondering about average travel times during peak and non-peak hours, average headways, etc… Seems like fairly basic information, but I’ve never seen it.
I just don’t see how the system can carry the volumes we need without running longer trains at reasonable headways. As Alex says, it works for BART. In fact, this morning things were so bad that I rode from Castro to Civic Center, climbed up to the concourse, went back down to BART, boarded a train, rode to Embarcadero, climbed back up, and crossed back down to MUNI, and continued on to 2nd & King (on a LRV with a broken front door, natch), and I came out well ahead compared to spending the next half hour stuck in the subway. It cost me an extra fare and is a completely absurd way to travel, but that’s how bad the system routinely operates. I realize that four car trains aren’t going to work through the Sunset, but when the subway is jammed up yet again, why not go into manual mode and start coupling up?
I just don’t see any effort at innovation whatsoever. The one and only novel idea we’re trying is to add a bus line to cover the same route the N Judah will already be traveling. In other words, SFMTA has admitted that it is too incompetent to provide remotely passable rail service, so it’s going to spend yet more money to duplicate the existing service. Surely we could be trying any one of the numerous ideas to improve metro service, many of which wouldn’t cost a dime.
And if the metro is such a disaster now, imagine what the additional complexity of the Central Subway will bring.
I fail to see how being an asshole gets us anywhere. It sounds like you have some grievances you’d like to air. The MTA, the CTA, the Board of Supes all have public meetings you can go to. Come on down to a CAC meeting. They’re at 1 South Van Ness, 3rd floor, first Thursday of the month. We take public comment. It’ll go on the record and you can follow up with it.
If you want to affect change, sending me a flurry of angry tweets and bitching online solves any of the problems Muni has. I’m trying my best. I’m reading all the library has, emailing my Supervisor and transit-minded people and I attend hearings when able. I’m learning and trying new things. Heaven forbid someone write a blog post about a neat tour they got to go on.
The SFMTA is the 800 pound gorilla in the room. It’s fun and easy to keep Muni as a punchline. It’s excrutiatingly frustrating to read about the union vote amid driver texting, reduced service, and lower quality rides. Individuals in the SFMTA are good people. Most of them are trying to bring better service to San Francisco. We’re all frustrated that the people’s railway is being run into the ground. It’s more effective if you don’t insult the other party and shut down all communication. There’s still another human being behind the other LCD screen.
As long as you continue to simply repeat the MTA’s gospel /verbatim/, you’re being counterproductive. There are /no/ rider advocacy groups around. The MTA has its 6+ figure PR budget (as well as a few billion a year to defend themselves), the TWU created its own “riders” union. As for the actual MUNI riders? We, the riders, get nothing. So, yes, when I see yet another MTA press release put forth by someone who should be advocating for the riders it is /extremely/ offensive to me.
When /you/ say stuff like “Running a three-car train is not a realistic solution to easing transit woes.” yes, it does call the tenor of your blog post into doubt… and it makes me wonder how much experience you’ve got using MUNI. One of /the biggest/ problems (and this has been brought up by the CAC and the news media before, btw) with the current trains and train control system is that they run one car trains through the subway. So, yes, Thales (nee Alcatel) can tout that (slightly) more trains are going through the tunnel. But what’s being omitted is that they’ve got 1/3 the capacity of the Boeing/cab signaling setup. Oops.
A quick look through the RescueMuni archives (on Yahoo groups), a quick look through the old CAC minutes, a look through some of the Examiner’s coverage, or even just /riding the train/ would have given you more than enough information to come up with a more accurate post.
Your post was not about a “neat tour” the minute you started retelling the miserable excuses that the MTA uses to avoid implementing any improvements. It’s not a moral judgement of good or bad, I’m sure that most of the people at the MTA are good people… but they’re /not/ good at their jobs and it shows.
When that person behind the LCD is so slow to respond (if they respond at all) that LRV drivers manually enter West Portal outbound… and forty-odd people get sent to the hospital as a result, how polite are you supposed to be? When the street-level supervisors completely ignore headway adherence leading to congestion in the tunnel, how polite, how constructive are you supposed to be? When the TWU fights riders tooth and nail, just how much civility do you want? When the LRV driver has responded to my report of a safety problem by ringing the horn, the bell, flashing the headlights for /2+ minutes/ to gain the attention of a maintenance tech because they’re sleeping in their truck, how polite am I supposed to be? When I’m on an LRV full of passengers that a maintenance tech is field testing (malfunctioning air system and thus friction brakes) in the Twin Peaks tunnel in cutout mode, well in excess of the /auto mode/ speed limit, should I congratulate them on a job well done? When the MTA uses lies and deception to avoid equipping the F-line vehicles with NextBus transponders, who should I praise?
When the woman in charge of the TEP wanted to gut service on the 108 as a first resort, what’s the proper response? You can’t walk on or off Treasure Island it’s bus or car ONLY. But she knew that, it’s /her job/ to know that. Sure, I plastered the the island bus stops and the Transbay Terminal with flyers encouraging public comment. And, yes, enough people came to the 1 Van Ness and complained that she backed off… a little.
I took a friend on her very first trip through the subway and out to the boonies last week. On her FIRST trip she was absolutely puzzled by the same operational issues you want people to bring up at the public meetings. When it’s a constant fight to get anyone at the MTA to use a modicum of common sense that even a transit newb can muster, it’s time to call a spade a spade and point out that enough of the MTA staff does not do a good job that it’s having dire consequences for the residents of San Francisco.
As a rider, no, it’s not fun to watch the MTA being used a perpetual political football. Every time there’s a new administration, the exact same questions, the exact same answers. Round and round you go. If you go through the CAC minutes you’ll find PLENTY of stuff on record. Nothing I’ve said is new, and none of the improvements I could suggest are new. If you want to do something constructive, try raising some of these same issues again and be PERSISTENT with them. Hey look here’s something we’ve brought up five, ten, or even fifteen years ago with no solutions. But putting out an MTA press release and couching it as a neat tour? That’s a slap in the face to riders.
You want something specific? Fine. Look at Zach’s request for service information. That type of information was something that could have been gleaned from the daily service reports. When Ford came to power, the reports actually stopped. They returned, but have become increasingly lacking in content. Why is the CAC not /continuing/ to press this issue? If you want an example look at a DSR from last week, and then dig one up that was hosted on the RescueMuni site (google has indexed them).
And, yes, I’ve engaged the D4 sup on this issue. You know what Chu has done for MUNI riders? Shiny yellow stickers politely asking cars not to run us over, and at one point the removal of the track irons. Whoop de do.
By virtue of being on the CAC, I expect more from you than a list of excuses couched as a “neat tour”. If you want more information then either look harder or, you know, ASK. Streetsblog, Munidiaries, RM, transbayblog, the MTA’s web site. The information /is/ out there. If you’re curious enough, I bet David has more patience than I do to explain what needs to be done with the DSRs (or just compare and contrast).
Well, I think there’s a middle ground here. Personally, I’m grateful to Katie for taking her time as a volunteer to tour the LMC and report back what she was told. As a citizen volunteer committee member, Katie is not responsible for the MTA’s woes, and I think it’s sympathetic of how bad the communication problem is that people are lashing out at her over this issue. She took the tour and wrote a blog post relaying the party line as it was given to her.
From that effort, there are questions and oddities, such as those Alex and David and I posed above, that might make for good next steps. I’d rather that Katie be involved in gathering this kind of feedback and attempting to advocate for changes, not that anyone has ever been particularly successful at changing MUNI, than to see her blasted for her efforts. I came to this thread because I wanted to learn more about how the system functions in order to better understand why it’s not working.
As for myself, my sole grievance here is simply that the light rail system does not function according to any kind of remotely acceptable standard for performance, reliability, or safety. This is obvious to literally any visitor from out of town and to most San Francisco public transit users unaffiliated with the MTA. I don’t care who’s to blame or even really how the problem gets solved, I would just like safe and functional trains to get from Castro to Embarcadero in 10-11 minutes with reasonable headways and for the subway to be more reliable than the airlines.
Would coming to a CAC meeting and saying “I’d like the subway to work please” help cause any sort of improvement? That doesn’t seem like a useful exercise to me, but I haven’t tried so what do I know? What I do know is that going “oh so that’s how the system works” after a tour of the LMC isn’t going to unclog the tunnel, so what do we do now?
Well that’s my point. The MTA has billions of dollars a year to put out their party line. And they do just that pretty darn well. Take a look at their ads about how great the new bus ‘shelters’ are, or how many hundreds of thousands of dollars they’ve spent defending themselves against the TWU. They don’t need members of their citizens’ advisory board for that kind of stuff. It’s the Citizens /Advisory/ Council /not/ the Citizens Yesmen Council. Simply retelling the MTA’s half-truths isn’t much to be commended for, IMO. When I look at the list of people on the CAC, I see at least three names that jump out as people who’ve been around long enough to see through the half-truths and outright lies that were sold to Katie. Perhaps there should have been a group tour? Or instead of rushing to the presses, the other CAC members could have been consulted first and the council could have worked /as a group/?
There are enough resources, enough information, and enough knowledgable people around to do better than simply towing the MTA’s party line. You don’t need random people to comment “on the record” for that. You’ve already got CAC /members/ on the record bringing up the issues. Everything that David’s mentioned so far has been proposed to the MTA for years on end. At CAC meetings, at TEP planning meetings, within RescueMuni, on Streetsblog, at various other pre-MTA hearings. These /are not/ new issues or new fixes.
Here’s an example:
“The SFMTA CAC recommends that a direct telephone connection between the Line Management Center (LMC) and the Metro Rail Operation (MRO) supervisor on the platform of the Embarcadero station be established, using the existing landline telephone to the platform and new wireless headset for the MROs. This would enable the LMC to re-direct LRVs from one line to another for their outbound run. As a consequence, this would reduce bunching and service would be better distributed.”
“Daniel Weaver would like to know the dwell time delay and ridership totals for 1997.”
As far as three car trains:
“Steve Ferrario stated that the passenger throughput of 1997 was a record high. He questioned why Muni no longer coupled trains at the portals. Ms. DeVlieg stated that the ATCS can accommodate coupling, but Operations chose not to do so due to problems with signals on trains that are longer than two-car trains. Also, Operations is concerned about schedule complications and coupler maintenance of repeated coupling and uncoupling of cars.”
“Mr. Wieser asked if ATCS was the cause of the slow train crossover at Castro. Ms. DeVlieg stated that she did not have knowledge of the crossover, but the ATCS “go-slow” zone for that location was possibly directed by Operations.”
Here’s another example:
“Mr. Haley stated […] there isn’t a capacity issue in the subway.”
“Jim Kelly, senior operations manager, Central Control, [stated] the current LRV fleet is adequate to include service for the Central Subway”
“Greg Riessen asked about purchasing trains with functioning windows. Mr. Kelly stated that functional windows posed a safety hazard in tunnels.”
So, yeah, when you’ve got your head burried so far in the… sand, what sort of rational discourse do you expect? These exact same discussions have been going on for years and years. Action from the MTA from the CAC is needed, not public comment.
Ugh. The links should have been as follow:
All CAC minutes, all accessible on the MTA’s web site.
Thank you Alex. I don’t think it’s fair to say Katie has her head buried far into anything. She clearly joined the CAC recently and is trying to learn more about the situation. The only thing that’s important is what can happen now.
The quotes you pasted above bring a lot of things to light in fact. Following up on them, I came to see February 2011’s CAC minutes (http://bit.ly/ifYqml), which also happens to appear to be when Katie joined the committee. At that meeting, the CAC apparently approved Motion 110126.05, which addresses a couple of these brain-dead simple common-sense ideas for improving LRV service:
– “The SFMTA CAC recommends that a direct telephone connection between the Line Management Center (LMC) and the Metro Rail Operator (MRO) supervisor on the platform of the Embarcadero station be established”
– “The SFMTA CAC recommends that at least two MRO supervisors, equipped with wireless headsets, be stationed at the Embarcadero station during peak periods, one for the inbound and one for the outbound platform and one for the outbound platform.”
– “The SFMTA CAC recommends that the MRO supervisor at Embarcadero station ensure that regular and relief operators exit and enter the LRVs as quickly as possible, in order to reduce dwell times for inbound and outbound vehicles.”
I cannot fathom why these suggestions weren’t obvious to SFMTA on their own, but now that these ideas have been officially recommended, what is their status? As far as I can tell, operators are as sluggish changing places at Embarcadero as they were back in February. I’ve certainly never seen a supervisor make any effort to reduce dwell times even when the subway is heavily congested during rush hour, and it’s pretty rare to see a supervisor in general at Embarcadero.
There are plenty of changes that are difficult or costly, such as fixing ATCS and getting the wrecked LRVs back in service, but we’re not talking about those kinds of changes here. We’re talking about a telephone line, a cordless headset, and a supervisor to encourage both operators and passengers to enter and exit trains quickly. To anyone but the SFMTA, these would be easy and basic requirements and would have been done years ago. It should be obvious to everyone involved that if there are three L trains in a row, several should be dynamically reassigned at the turnback to maintain headways, but how often does that happen?
If nothing else, the psychological effect of these kinds of changes is important too. It’s like the open kitchen at In-N-Out Burger: you might have to wait a while for your food, but you see a whole army of staff rushing about and working hard, so you don’t mind the wait as much. MUNI riders might be slightly more satisfied if there’s an attentive supervisor standing on the platform who is actively hustling to keep the system moving.
We know this stuff. These basic, simple, and inexpensive proposals have been recommended for years, and yet nothing changes. I don’t doubt that there are good people in the SFMTA who want to improve service, but where are the results?
That’s a reference to the MTA staff having their heads covered, not Katie. Look at the general clueless tenor of the MTA responses, and look at who is and isn’t showing up to these meetings. It’s absurd.