How the World’s Best Transit Systems Achieve Amazing Frequency
Vložit
- čas přidán 30. 10. 2023
- Watch this video ad-free on Nebula: nebula.tv/videos/rmtransit-de...
We've talked many a time about how important frequency is to a transit system and its riders. But how can a transit system achieve frequent transit? Let's discuss.
As always, leave a comment down below if you have ideas for our future videos. Like, subscribe, and hit the bell icon so you won't miss my next video!
=PATREON=
If you'd like to help me make more videos & get exclusive behind the scenes access and early video releases, consider supporting my Patreon! Every dollar goes towards helping my channel grow & reach more people.
Patreon: / rmtransit
=ATTRIBUTION=
Epidemic Sound (Affiliate Link): share.epidemicsound.com/nptgfg
Nexa from Fontfabric.com
Map Data © OpenStreetMap contributors: www.openstreetmap.org/copyright
=COMMUNITY DISCORD SERVER=
Discord Server: / discord
(Not officially affiliated with the channel)
=MY SOCIAL MEDIA=
Twitter: / rm_transit
Instagram: / rm_transit
Website: rmtransit.com
Substack: reecemartin.substack.com
=ABOUT ME=
Ever wondered why your city's transit just doesn't seem quite up to snuff? RMTransit is here to answer that, and help you open your eyes to all of the different public transportation systems around the world!
Reece (the RM in RMTransit) is an urbanist and public transport critic residing in Toronto, Canada, with the goal of helping the world become more connected through metros, trams, buses, high-speed trains, and all other transport modes.
Way too many North American rail transit agencies run longer trains at lower frequencies, when the better model is shorter trains at higher frequencies. BART and LA Metro are both infamous examples of this problem, which is a huge reason they've faced such difficulty recovering ridership to 2019 levels even after virus restrictions ended.
The issue is labor. They are terrible with route planning, transit system and vehicle designs
Well this also means that the rule of “quality over quantity” also applies to transit, even Singapore can run a MRT line with high demand efficiently with 3-car trains just by having higher frequencies
Jeremy Clarkson once said, "Americans don't really get luxury and quality, all they know is how to make things bigger". He was talking about cars, but it's surprising just how many things that statement applies well to.
LA’s trying to get their frequency back up, but they’re apparently facing a shortage of operators
I'd suggest this is mostly an American problem, as opposed to Canadian systems
The Victoria line in London has terminals with just a single island platform. The train arrives and the driver gets off, and at the other end another driver is waiting, and gets on to drive the train back the other way. As the train leaves the first driver walks down the platform to the other end, and while he walks, a train arrives and does the same on the other platform. Shortly after he reaches the other end, the next train arrives, its driver gets off, and this driver gets on to drive it back.
The Victoria line runs every 90 seconds at peak times, so it's a pretty efficient system and works well, and no need for reversing sidings.
I mean, it works, but it's definitely not an optimal solution. Even the peak 36tph (100 second headway) can only be achieved for a short period because of these 2 track termini. And 'stepping back' has obvious issues in terms of resilience to disruption, as it requires virtually all drivers to be in the right place at the right time, all the time.
So problematic are the 2 track termini on the Victoria Line that there have been repeated schemes to extend the line one stop south from Brixton on to a loop line to Herne Hill - *just to get rid of that bottleneck*
IIRC the Victoria line does 32-36tph so not quite 90s?
The stepped back operation is effective, but not quite optimal!
@@RMTransitThe Victoria currently has at least 34tph in each peak, scheduled services sometimes get to 85s between trains, with a total line capacity of 40tph, the reason why it doesn’t run at 40 at the moment is because of long dwell times at stations like King’s Cross St Pancras and Victoria, where already the amount of people often makes trains up to 20s late which causes the trains to sometimes run as close as 70s behind each other. This is one of the reasons why crossrail 2 was proposed as the amount of people using the Victoria keeps on increasing but it is becoming impossible to retain its high reliability and operate more trains, so crossrail 2 would alleviate pressure and allow the Victoria to finally reach 40tph, which would mean it becomes double the bakerloo line peak frequency.
I don't know the tph figure, but the Waterloo & City line is an interesting case study. They have 2 dead end platforms at bank, and reversing sidings at Waterloo. In the rush hour, the trains seem to dwell for only about 1 minute at most. In fact I think they have 2 drivers in them. One at each end, and they just ride in the back cab on the return journey rather than getting out and walking along the platform.
@@haltendehand1 But dosnt all sorts of ways to reverse train at such frequency require "all drivers to be at the in the right place at the right time?" (apart from automated reversing)?
At such frequency, even with reversing sidings beyond the terminals, the time between the train stops at the end of the track and it needs to go the other way usually is not enough for a single driver to {switch off the cab at one end, walk the full length of train, switch on the cab at the other end, do all the checks, start moving the train}, so the norm for metros with such frequency is for another driver (driver B) to step in the cab at the rear of the train when it comes into the terminal station, and when the train reaches the end of the reversing siding, driver A switch off the cab and handed control to driver B who is now ready to move the train in the other direction. I believe this is the norm and it has the same require "all drivers to be at the in the right place at the right time?"/ "resilience " issue that you mention.
And the alternative as I mentioned before is automated reversing, which may not be able to be implemented in older systems. ie when the train stops at the terminal platform, the driver hands control to the computer, while the train doing all the {driving to reversing siding, reverse, move back to the other platform}, the driver walks along the train and by the time the train reaches the boarding platform, the driver that drives the train to the station is now ready to go at the other end.
(PS. I believe at Victorian line terminals, they step back by two trains, which means at the peak there are always more than one driver waling back at the platform, hopefully resolving the "resilience " issue
“Transit should be frequent”
My local route: 2 buses per hour during weekdays, take it or leave it
To be fair, If the 2 buses per hour are on time, this can be useable.
I have been on a system running 2 per hour when I could not tell if the bus I was on was very late or just a bit early.
2 per hour is unlikely to encourage you away from a car if that was the alternative, depends on other factors regarding costs and time.
@@stephenlee5929> two buses per hour, on time
Isn't this how Swiss villages work?
for a small town or village, if it easily connects to rail, that can be quite good. For a large city it's horrible.
Indeed, when you combine reliable service with solid speeds lower frequencies can actually be ok!
one thing that wasnt mentioned in the dwell time section is level boarding. its not just important to people like me who use wheelchairs, but also speeds up boarding emensly with busses and trains. especialy if you look at trains like go trains or old trams with 3 stairs
Yeah, dwell times increase significantly when the guard has to go fetch a wheelchair ramp!
Fun fact: the S-Bahn Berlin was the first public transit system in Germany that adopted level boarding for the very reason to decrease their dwell time in the 30s. This is also the historic reason that the S-Bahn systems in Germany use platforms with a 96cm platform hight.
Level boarding also minimizes the risk of getting injured while step on the train (for example older and unstable people and those who are handling with luggage or a bike for example). At my work as a public transit planner, I therefore try to explain all my partners that everyone benefits from level boarding, not just "those in a wheelchair"
Yes for real. Also for people with canes, strollers, walkers, suitcases on wheels, little kids, carrying heavy things. Level boarding and small gaps.
@@mdhazeldine As the poster noted, if access to the train is through a narrow staircase, boarding is immediately slower even for passenger without luggage. And if you have a heavy bag, especially luggage with wheels, that you have to drag up the stair, the difference is huge.
In the UK, most buses sadly only have 1 door all the way at the front so at the busier stops, you often have to join a queue to get off if you're sitting at the back. In the rest of Europe, buses often have 3, 4 or even 5 doors.
Buses in the UK are rough :/
My local operator Brighton&Hove run core routes, i.e those with daytime frequencies of 12min or less, using twin door double deckers, those introduced in recent years being battery hybrids.
The less intensive evening and Sunday services see these vehicles utilized on outlying routes which normally rely on 'cascaded' vehicles.
B&H tried 'bendy buses' for a good few years, but they didn't mix well with a 200 year old city centre, cyclists, pedestrians and 'over-refreshed' stag and hen parties..
@@howardrisby9621The point is outside of London, more than 1 door set per bus is the exception not the norm. Unfortunately.
In the US buses usually have two doors but the rear door is often unusable due to the poorly trained drivers coming up to the curb so that the bus is at an angle or the artic bus is crooked, and passengers would be confronted with too high a step or too wide a gap to mind.
in greece they have 2-3 the short ones, the long ones have 3-4
in cities other than athens its also 2 entrances
I'm on an overseas holiday in Beijing at the moment. I've been using the subway a lot, over a dozen trips so far, some of those with multiple interchanges, so something like 20 trains. My longest wait has been three minutes. Sometimes you're at the top of the stairs going down to the platform and you can hear the train you've just missed as it's leaving the station. If you walk at a normal pace, by the time you walk down the stairs and to the middle of the platform, you can hear the next train coming.
Trains that are more frequent than traffic lights.👍
Been trying to adjust to other cities after being in Hong Kong. I miss that so much
Excellent! Might be worth looking at Moscow. Its Metro has by far the greatest frequency on the planet, 75 to 90 second headways! It is more a human conveyor belt than a transit system.
I've seen it multiple times now: WHY on earth are trams in the US/Canada driving in that slow manner? How are they supposed to compete with cars if you have to be slower than them on the SAME street, even with their own lanes?
This is why grade separation (metro lines) are key!
@@frafraplanner9277 Yes, but you can clearly see in so many footage clips, that the trains are driving slowly even on basically grade seperated lines.
I could outpace them by bike.
@@TheWeirdaholic not any different in many European cities: might be quicker while driving parallel, but the tram will be slower due to the frequent stops
Infrequency of a transit system leads to sort of chicken or egg problem. Is there low ridership due to low frequency? Or do transit systems have low frequency in response to low ridership?
Absolutely, frequency needs to lead!
Spanish solution is an amazingly practical layout if there's enough traffic to warrant it. Can't widen the existing island platform because there are tracks in the way? No problem, just put additional platforms on the side. In the case of two perpendicular lines, you are a single escalator ride from one line's exit platform to either of the other line's boarding platforms. You still get the benefit of being able to transfer cross platform between tracks on the same line. Tracks 2 and 7 at Jamaica station on the LIRR utilize Spanish solution and it makes being able to cross to another platform without needing the stairs much easier by allowing one to cut through the train. It is not idiot-proof, but there are things you can do to coax people to use it correctly.
Properly design and sign stations and passages so that people naturally go to the correct platform to board. There are several examples in the US where Spanish solution-style platforms are in place, but station design and access constraints make it unworkable. Park Street station on the Red Line of the MBTA is one such example. Open the car doors on the alighting side before those on the boarding side, with a delay of several seconds. You want enough time for a clear pulse and flow of all passengers towards the correct doors to start before you open the doors for passengers to start boarding. That way, passengers don't even need to think about it, they just go with the flow.
2:36 Munich S-Bahn commuter rail trunk line in downtown Munich runs at a potential capacity of close to 50k passengers per hour per direction during rush hour, as the commuter trains of the different lines come together on a 2 minute schedule / adding up to 30 trains per hour per direction
Well said about the doors! Whether it's the NYC Subway, SEPTA's Market-Frankford Line, or the PATH, the trains have three or four doors on each side! They notice how effective having three doors is on a rapid transit train, and yet they don't think "Let's do that with suburban rail so the operations are just as effective". Like the LIRR got fancy M9s back in 2019 and yet they fall into the same problem as the M7s and M1/3s, it's just two doors. The LIRR C3 bilevels having small steps to get to the seats is even worse for crowd flow! Having a third or fourth door is great for avoiding congestion at the doors and maintaining good crowd flow, or having two much bigger doors like on the Macau LRT as you pointed out can do the trick too!
And of course, Spanish solution works like a charm as well! Besides European examples, a North American example of Spanish solution is at the 33rd Street PATH station, opening the doors on one side to let all the 33rd Street passengers off and then opening the other side for those getting on the Journal Square-bound train (or Hoboken-bound when Hoboken has its own Hoboken-33rd Street trains on weekdays). Epcot, TTC, and Magic Kingdom stations on the WDW monorail uses Spanish solution too! Driverless metros also allow to run higher frequencies without a huge operational cost, then they don't need to use more people to operate more trains, and platform screen doors simplify station layout, thus increasing safety.
GOA-4 operation also massively improves frequency, allowing trains to instantly reverse without waiting for the driver to walk to the other cab. Lyon line B (which was automated last year) has 2 minutes headways despite its northern terminal being single tracked for historical reasons.
Frequency makes transit so much better!
I wish Seattle would increase the frequency in its existing downtown tunnel rather than building a second tunnel a block away and forcing people to walk between two tunnels for transfers between lines.
I still like 8sec headways a gondola can provide - no train will ever reach such - makes transfers so much more pleasant!
An interesting point. I never did try the gondola transit in Rio.
If you go just by frequency walking everywhere would be perfect. Gondolas are just far too slow and often low-capacity to be better than metros
@@jan-lukas To be fair, you need some degree of safety as well as frequency. So Walking does not work in some US and Canadian cities, see not just bikes Fake London.
Seattle's situation with Link is very weird!
@@jan-lukas gondolas won't compete with a metro on long lines, but make for great short feeder lines where frequency is far more important than travel speed
On some lines in London, they improve turnaround time by having another driver waiting at the other end of the platform to get in at the other end to drive the train out. The driver who took the train into the platform leaves the train, then walks to the other end of the platform to drive out the next train. It requires an extra driver per shift to do that because at any given time, one driver will be walking to the other end of the platform while the others are driving trains. But you do get more journeys for the same number of trains.
The problem you mentioned at 5:01 is the problem the Tung Chung Line and Airport Express on the MTR faces! When British Hong Kong was planning to build it, the Chinese government raised concerns about the effect of the project on the territory's fiscal reserves, which eventually forced the Hong Kong government to reduce the cost, and this led to many limitations. The power supply system restricts the number of trains running between Kowloon and Lai King stations. The system can accommodate a maximum of one Airport Express train and two Tung Chung trains travelling in both directions at one time. The minimum headway on this section of the line is 3 minutes 30 seconds.
The airport rail link was originally designed to accommodate four tracks, two each for the Airport Express and Tung Chung line. It was later reduced to two tracks where both services share the same trackage. As a result, signal failures can affect both services. The Lantau Link section of the line only allows one train to pass through each direction at the same time, raising the minimum headway between trains to 2 minutes 15 seconds. As a result, some Tung Chung line trains terminate at Tsing Yi during peak hours, instead of travelling the entire line. The signaling system doesn't give priority to Airport Express trains, thus Tung Chung Line trains at Sunny Bay frequently impede them.
Can You please make a video on poor condition of bus services in Indian cities other than Delhi (like Kolkata) a constant reduction of buses in suburbs.... Govt is investing less and less.
Also, love your videos, from from the city hosting India's first metro system ❤️🇮🇳
Thanks for your comment, its hard for me to cover something so local from so far away!
Nice to see my normal commute feature in your video (Wimbledon on SWR 455s). I hope you had a nice time here. It seems like you got lots of new footage!
I was in Paris just a couple of weeks ago and every time I am there, I am amazed by how frequent it can be. On some lines, you can actually see the train leaving and entering the station pretty much at the same time. Even though, some of these lines were build 100 years ago, it is amazing that you walk down the steps, a train is leaving and as soon as you are on the platform, another one is coming. I have not yet seen any transit system with a more frequent service than in Paris.
I'm always amazed how much information you manage to pack into each video.
Makes me always proud seeing a train from my hometown Berlin. A bit silly I know but…
Thank You Reece for the very interestng and informative video. Watching the Incheon Airport Maglev "Ecobee" i was always asking myself for what the X-Cross switches at the entry of the station are needed. Now i know it. For boarding and deboarding.
This was a great and concise look at this. I'd add that platform doors (in places that don't already have them) can likely help with being able to run higher speeds into and out of stations without compromising safety. It would probably help keep the tracks much cleaner too, especially in places like NYC. Those trash-driven track fires really slow down trains...
Recently opened Pune Metro in India uses the same technique of boarding and deboarding as you mentioned. Just that Pune has it Overground on Viaducts.
I’m a sucker for a design video so I loved this one! Thanks
Berlin S-Bahn acceleration always gets me. Especialy compared to trains on neighbouring tracks that ultimately reach higher speeds but those S-Bahns just speed up so fast its crazy. Outruns trains departure the same moment every time. Not even close.
Stammstrecke is great in Munich. It does have problems with late trains and delays which create ripples through the network occasionally.
But they are planning on adding an express Stammstrecke route.
Before electrification, Brisbane used to have full-width compartment carriages; each carriage had nine compartments (seven for southside trains), with two full-width cross benches (seating 5 each). So there would be nine doors on each side of every carriage. Worked well at peak, and these old wooden carriages were prefered to the more modern SX sets in peak.
Bondi Junction (in Sydney) had a relay driver. Train would arrive and the driver would walk through the train to the rear end. Relay driver would get in and drive the train forward into a short sutb in the tunnel. When there, the original driver would drive forward onto the departure platform.
Olympic Park (also in Sydney) has an excellert variation to the Spanish solution. It's on a loop. The arrival track divides into two, and the platforms are side-track-island-track-side. Arriving passengers disembark onto the island platform, then out to the world. After the train is clear, the island doors are closed and the side-platform doors are opened. Departing passengers for westbound services are on one side platform, and city and east-bound services on the other side platform. Good solution for departing passengers as they know which platform to go to. Good for arriving passengers, as they don't have to fight for waiting crowds. This staion is busiest at popular events, such as the annual show.
Sydney Trains has something called a "City Circle" where three suburban lines running at suburban frequencies converge onto a single pair of tracks to run through the busy CBD.
It's pretty neat on paper, as it operates with metro-like frequencies within the City Circle, but if one of those lines gets delayed due to some emergency the knock-on delays would affect THREE lines instead of one. It's a pretty amusing achilles heel until it makes it onto the news with yet another "train delay paralyses entire network angering commuters" headline.
With the result that Australian governments are spending billions on new cross-city tracks instead of millions on improving reliability i.e. signalling, trains and communicaton.
@@jack2453 I don't think they could realistically have done any better. Better signalling wouldn't have solved the achilles heel of the system in squeezing 3 entire lines worth of trains into one, unless accleration and deceleration is improved.
With that being said, I reckon the best course of action is to build facilities to turn trains around after Central, to 1) alleviate congestion around the City Circle and 2) in the event of delays, reduce knock-on.
The problem is that I guess Sydney Metro's the accepted solution so we'll just have to roll with it. I guess robot trains can't go on strike...
@@kaihang4685 Multiple lines running through the central core branching further out is exactly how the Paris RER and the Berlin and Munich S-bahns work. It is not a fatal flaw. It just needs good management.
I've talked about the circle in a number of videos, while interlining like this can lead to knock on delays its done around the world without issue, just need good planning and operations!
For buses there are classic doubledecker buses, eg in London and in Berlin. many decades ago, they had a single stairs to the upper deck at one end. then they added second set of stairs to allow for entering on one end and leaving on the other, and used a central walkway with 2seater benches instead of having the same single walkway for entering and leaving at one side with 4seater benches. more recently they started replacing old doubledecker buses with articulated buses which are twice as long, have more and wider doors, and have the same capacity (around 100 people) but no longer need stairs. I still love the tradition and hate when they get rid of most doubledeckers, but i completely forget about this when i can very quickly enter and leave on short stops. and on most innercity routes, i have to wait no longer than 5 minutes for the next bus on the same line, or even much less if (for short travel distances) there are 2 or 3 lines which partially overlap. (similar frequencies on multiple lines apply to trams)
similar applies to U- and S-Bahn (capacity up to 1000 on long trains) which (for me, in the center of the city) come every 2-3 minutes during rush hour, 5 minutes normal, and even 10 or less at other times. who needs to look at and learn timetables with good frequencies ?
the S-Bahn on lines to and beyond city limits often only has a "low" frequency of 20 minutes (still not bad when you then sit in the train for 30+ minutes for those distances), but through a big part of the town seven lines share the tracks and thus people mostly can use any train every 2-3 minutes. the next problem are probably some old stations (built 100+ years ago, right in the center of the city) that almost can't be rebuilt easily or at all (at least not while continueing normal operations, and not when some base are protected historical buildings), creating bottlenecks on stairs and elevators instead of on the train doors.
We really need to take pages out of Japans book!
We should be using locomotive pulled trains only on intercity services, and they actually work great here. All suburban routes should use high capacity EMU's like you said, but I feel we need to design a better trainset that uses a mix of both single deck cars for door capacity and double deck cars for extra capacity in general. Picture something like the Paris RER NG sets but with a few more intermediate single deck cars. Or, ideally, picture something like a JR E231 / E233 (my favorite trains in the world) but with a few more double deck green cars (with wider doors and stairways then current cars of cource.) I really do think America really needs proper Japanese trains!
It's also not impossible to run both high frequency service with both express service also local service. I've calculated this myself; It's possible to run 32tph local and express service if you put crossing tracks at every other station. You can still have top speeds of 140 km/h which can mean the fastest service can average ~95 - 100 km/h! Also, you can coordinate schedules that can result in 8 express, 16 rapid (with varying stopping patterns), and 8 local trains. Meaning, you can have up to 24 express-type trains per hour which can be massive capacity for regional trips that suburban rail serves! (Nerdbrain moment: If you use Tokaido line style 15-car JR style trains, which would actually be shorter than the current GO trains, you can have up to 100,800 pphpd per line! 20+ lanes of traffic! With 84,000 pphpd for just the 24 express-type trains per hour! Now THAT'S capacity!) It's hard to explain here how it all works but it's much easier than it looks; plus I'm sure you've already figured things out like this anyway!
Overall, we should learn not just from Europe, but from Japan! I'd love to see a JR east style suburban rail network in North America!
I saw on Madrid's line 3 on a rainy day I decided to 100% the line (having only gone south of Sol on 2 occasions, one of those times was to the train museum). At Moncloa, it does switch sides, but there was another train sitting on the other side at Villaverde Alto. Line 2 and Line 4 do that too sometimes at their western terminals of Cuatro Caminos and Arguelles-but there it makes sense to have most trains have access to both the side and Island platform (which at Cuatro Caminos is a quick transfer to continue northbound on Line 1-which was useful to me being one stop north), so it simple cuts out some inconvenience where that case at Villaverde Alto proved well as I was just going to take that train back to Legazpi...and then I think I walked over to the Planetarium (I had no plan and had the day off from class)
I am jealous, I need to visit Madrid!
@@RMTransit well Pinar del Ray was closed and Cuatro Caminos a construction zone most of my time. Line 3 extension should open in the spring and line 11 in phases from late 2026
Accessibility is also vitally important. I was recently in Chicago with the family. We have always been a fan of the L. Recently the CTA has been making strides in making trains and stations accessible, adding elevators etc. However, wheeling my grandmother on and off the trains in a wheelchair we found that the nearly level boarding was not level enough. We had to yank the chair backwards up to an 8 inch jump on or drop off the trains. There is also up to a 8 inch gap between the train and the platform that the small wheels could get trapped in. As you can imagine, this made getting on and off take longer. We were told on the last day that there are ramps, and that the people at the station can help with them, and call ahead to the disembark station so they will be ready too. But that also sounds like it would take some time and slow things down.
Now, most of the points only truly matter for a rapid transit operation which operate in a very high frequency, though you still considered that for mainline operations.
The only things I'm missing from this video is the lack of references to some of your older videos like the science of flying junctions, cross platform interchanges and the like as they too affect the frequency of operation as well as some of the basics like number of tracks (though the operation for multi-tracked operation is a science of its own) and services types of operations.
On that aside, the idea of tail tracks is quite interesting as from a passenger's point, the in- and outbound platforms are still clearly defined and also gives of an illusion of a continuous streams of trains. For the other, I commonly them at service termini with continuation for other trains on the line (pocket tracks) which makes sense because they're the easiest way for a train to turn around without disturbing other trains, all happing at grade but without any crossings (which is why I consider them interesting).
They also act as small yards for trains which always fascinate me for some reason when I see one in person.
I’m fixated on the door issue and dwell time, and can’t help but see the complete inefficiency of airplanes in this regard. Airplanes are great for long haul, but for regional trips, boarding and unboarding can take as long as the flight itself, with modern rail being an obvious alternative.
5:00 Actually the Frankfurt Stadtbahn (aka Subway) reduced their service in u6 and u7 routes. Some trains would stop in the tunnels as there wasn't enough electricity. The VGF instructed the drivers to not accelerate to fast hahaha
Thank you for this. I wanted to help my city improve service frequency as the trains have gotten overcrowded and this video gave me some insights on how to handle peak rush load.
Could you share more videos on the Indian metro which are in progress - Delhi metro and Bengaluru metro to begin with. The latter is facing overcrowding issues
Back in the 70’s the schedule for the Budapest tram and metro lines just mentioned the interval trams and trains would arrive during the day. Ranging from every 30 seconds on the Great Ring trams 4 and 6 (the busiest tram lines in the world) to a max of about 5 minutes for less busier lines. The frequencies have decreased now thanks to car use but are still high enough so that you never have to look at a schedule …
6:19 From what I know the Berlin S- and U-Bahn trains termini also have two platforms but trains arrive on one wait ther for most times 5-7 Minutes to allow the drivers to change cabs and then set off again
Please do a explainer of the transit system in La Paz! Its a reslly unique system and I'd love to see you make a video about it
Rmtransit, just watching your video, and was very interested in "loops ' as you called it I be very appreciative if you could do a video on this piece of rail infrastructure. Thank you
There have been a lot of simulations of airplanes with a wing architecture, but tube and wing persists for known flow in emergency situations. Livestock seem to have used ventilated box cars with sliding doors. An interesting concept to use retractable doors.
I wish I could remember which trains were going to be "short-fast" . The intermodal trend was long and they seem to be relatively fast. The heavy haul are long and relatively slow. It must have been regular freight. Railroads have needed to increase passing tracks.
Viewing this from Boston, where the MBTA seems to be in a death spiral of worsening service frequency and reliability on both bus and light rail (probably also on the rapid transit, but I haven't been on those lines much). And the traffic just keeps getting worse, including well after midnight, after the T has stopped providing service at all (and on the rare occasions that they have tried Owl service, it has always been blatantly designed to fail).
Don't forget station placements and layouts. The best type of line stop placement would be one where every station switches sides of the train. this creates good passenger flow onboard during crowded times. And yes platforms on both sides of the tracks along with double platform boarding is a great thing -- especially when you have one track serving multiple routes to keep people waiting for certain segments on different sides and keep confusion to a minimum
Regarding the doors: the Avenio trams used in Munich have two doors per carriage compared to the one door per carriage in the R2 and R3 cars. The last two have that many doors were the good, old P type cars 🥰
And the new 200 m S-Bahn train presented recently has varying doors per carriage: most carriages have three doors, but some have two, but wider doors 😲
SD Trolley's main topological error is that the trunk branches into the Blue and Green Lines at Old Town, despite both branches having extremely strong outer anchors. And now the transit agency is crazy enough to propose an airport rail link running only every 15 minutes that would branch off the same trunk, via a flat junction, even closer to Downtown. (Luckily that plan is dying and the latest plan is an automated rubber-tyred light metro connecting the airport with Downtown with 2 minute frequencies!)
Also, San Diego's new UTC terminus has a scissors crossover at the station approach instead of beyond the station...whoops!
Great video.
I always love these tech videos.
Please have a look at the metro system in Sofia (🇧🇬Bulgaria). The layout design is really good
You should do videos about small North American cities, but cover 3-5 cities so its still a normal length video. That being said, you could do a video explaining how to improve current public transportation in cities like Barrie, Orillia, and Collingwood, ON.
Excellent content, well presented and impeccably edited. Small request, please could you say FEWER doors/trains not less doors because doors/train are a countable item. 😀
Well... since the size of the doors matters too, less might not be entirely wrong!
A couple things you might not have thought about:
1) Some doors will have been wide enough for two people when the stock was first built, but now aren't. I won't be surprised when these new trains come up against the same issue, as people continue to get wider.
2) Doesn't matter how wide the door, if people block others getting off. If people could use their brains for two seconds, they might realize making room for people to get off would make it quicker for them to get on.
In my country I think some boarding passengers panic when the door that they've chosen to board their train at still have people alighting whereas people have already finished alighting from other doors, so other boarding passengers that've chosen to board from those other doors then get to board earlier, & may end up taking up all the empty seats left on the train. Conversely when alighting from a train, passengers may avoid alighting at the same door as passengers on motorized wheelchairs, as they are slower & thus passengers at other doors will get to alight 1st & get further up in front of the escalator queue (to take you out of the station you alight at)
Any time a driver is involved in the frequency equation, there will be delays due to said driver. For best frequency, Driverless (GOA3) or Unattended (GOA4) operation is of benefit. Humans get tired, computers do not. ;)
As for the criticality of turnbacks , many systems used a "degraded" form of driverless by automating just the turnback operation. This allows the driver to walk from one end of the train to other while the turnback is in progress.
Please consider making a video on planning a TOD neighborhood
German "intercity trains do sometimes get delayed"... understatement of the century
6:42 it's NOT true that side platforms are more efficient for a terminal!
There are tons of high frequency metro systems around the world with Island platforms at the terminus. The track layout can be exactly the same, just with more space between tracks to fit the platform between them.
Actually it's better to have island platforms in many cases: you can have an additional crossover before the station that gives you the option of turning the trains around at the platform.
this is very common on the Vienna U-Bahn
I don't think he meant that side platforms are optimal, he just said that they were "typical". But yeah a lot of European systems just have them turn back at either side of an island platform which seems more efficient
Plus island platforms allow for a fully automatic reversing procedure which further increases capacity as it prevents the driver from walking the entire length of the train at every turnaround.
Two questions: 1. Which specific substack article were you referring to? 2. Is it possible to access OpenTrack without a university/industry connection?
Now there's something that will throw a wrench in a lot of what is considered optimal to increase frequency. Accessibility support. Fast loading and fast unloading often is not very friendly design for wheelchairs and other handicap modes
Many rail transit lines have built in restrictions . There is a diesel rail line in Denmark called Gribskovbanen which runs 4 trains an hour each direction on SINGLE TRACK for most of the day . Very old established environmental reasons demands that a they cannot build a 2nd track nor string catenary .
1:14 to could also have 2 island platforms per direction and longer dwelling trains with higher capacity and achieve similar frequency
Two lessons from London:
Many Elizabeth Line stations in the central core have entrances/exits at both ends of the platform leading to completely different exits at street level. So with a 200+m platform you effectively have two stations, and platforms can clear much more quickly.
Operating procedures can also make a massive difference to potential frequency. South Western Railway operates with guards enabling door opening. When the train stops, the guard opens their door, steps onto the platform to make sure the train is correctly stopped, then steps back onto the train to release the doors. This process can add 30 seconds to station dwell times compared with operation of the doors by the driver, a significant delay when suburban stations may be just 2 minutes apart.
Don't let the RMT hear you, they'll bang on about how it's sooo unsafe and ackshually passengers want guards! Even though the Tube, Thameslink and many other railways in the UK and across the world have used DOO for decades now with minimal issue.
The Spanish solution sounds fascinating.
It is!
I'm wondering why tail tracks beyond the terminal allow higher frequency of the train? Doesnt it require the train to stop twice and require drivers to walk from one end to another while the train is stopped at the tail tracks, not picking up passengers?
I'd guess that it's just the extra space you get from it, you can have 4 instead of trains at the terminal at the same time
Short answer: for a train to switch back when it has reached a terminal, it will limit the capacity because it needs to cross the path of incoming trains, which in turn then have to wait to enter the station. By having a train go to a track past the terminal, you eliminate this hurdle.
The short answer: Grade separation, the removal of level crossings i.e. points of conflicts.
Long answer: The purpose of tail tracks to prevent trains from blocking the tracks of other trains just because one is about to start or terminate. This is two-fold: On a pure terminus, this typically means a train will only start driving if the other train is about to enter the platform area and thus flexibility in operation but tail tracks don't need to care about blocked switched. On top of that, tail tracks leave the tracks clear for trains which proceed further without requiring them to slow down if not stop for both in- and outbound trains, further addign flexibility to the operation (without that, a terminating outbound train would have to proceed to the next turnout, wait there until the inbound train enters the station if one is about to enter all while blocking the tracks of the other outbound train and not much time to spare). On a minor scale, it also results in consistent platform assignments i.e. you always know that platform A has the starting and platform B the terminating trains (which IMO is more natural than having to figure out which train to take) and they also double as a storage space as seen in e.g. the S-Bahn tunnel of Berlin.
Ironically, tail tracks are a form of grade separation which happens at grade but it's also the only one I can think of.
I like this explanation, I endorse!@@@nigiiobi1154
The train has to stop twice, but you effectively get room to fit in a whole extra train. Regarding the driver walking down the train, the Elizabeth line in London uses completely automated operation for the turnback portion of the line, which means the driver can walk down the train while the train drives itself to the next platform!
What is OpenTrack?
3:48 You are showing subway line U5 in Berlin here, and it’s a bit ironic, because this is the “wrong” train, which has caused much consternation on the line. These trains are much too narrow for U5 (2.40 m vs 2.65 m; you can see the extension boards near the exits). Reason being: A local manufacturer had issues supplying new trains, so the public transport company had to borrow trains from the narrow line U2. Luckily, this won’t be forever, otherwise it would indeed make that line less attractive.
I covered this in my Berlin explainer video
does anyone know how to download the opentrack software? because i tried digging around their website but cant figure it out but id like to test some train frequency models
Low frequency can really ruin an otherwise great transit route. My parents live at the edge a big city and you can take the bus that takes 10 minutes in each direction to a local shopping street and a S-Bahn stopp in 2 different city districts, where you can either shop or take the Train to reach almost every relevant city nearby, but due to missing staff it only stops there every 30 minutes at the moment, which makes taking the train a pain because you have to wait almost 15 minutes for the train to arrive (almost the same time you need to get to the city center by car via Autobahn) and have the same issue on the way back. While I lived there as a kid the bus stopped every 20 minute, which was passable, but ever since I can remember they wanted to bump that down to every 15 minutes...
My biggest problem with getting in and out of a train are those idiots with the bicycles. Especially when there is no space for those bicycles and they are blocking the doors.
The most used lines of Milan's underground run every 1 and a half minute on rush hour but are still pretty full
We were in London (the UK one) for the weekend and on Friday afternoon the Jubilee Line had an issue. Instead of a train every 2.2 minutes it was 12-13 minutes. The overcrowding was crazy with people having to wait for 30+ minutes or so as they couldn't get on. It's always crowded at the best of times, but it really showed how important frequent trains are when people are used to using them!
That must've been really bad. I was on it the other day and had to wait 7 minutes, and then it took me 3 more trains to be able to get on!
@@mdhazeldine It was not great - we had cases with us that didn't help! We squished in and were popped out at North Greenwich (O2) where the dangleway was very quite to Excel :)
I have no problem with being squashed after 60+ years on the tube, but my niece wasn't best pleased lol
Can you do a video covering the Grand Trunk Railway?
Where is that loop at 6:05? Thank you 😊
It occurs to me that driver pay is also an obstacle to frequent service. Halve the number of cars in a train and you can run twice as frequent, shorter trains for the same number of cars total. But you’re paying twice as much for drivers. This suggests automated systems are a useful tool to increase frequency. Run as many trains as you need, and as your system can handle, and don’t pay any more for drivers because they aren’t needed. Really high frequency transit should probably work as much as possible like a horizontal elevator. Doesn’t need a driver, easy to get on and off, you might have to wait but not for long, and most importantly it’s just *there* and people just take it for granted that it’s there and don’t waste mental energy worrying about whether there will be a train or not.
Any idea what is the most frequent service in the world? Do any top the busiest tube lines at 36tph?
Good news. Did anyone hear about Edmonton finally opening its new Valley Line?
Thanks for beeing.
The Tube uses a technique at some terminal stations called "stepping back" to keep turnaround times very low. In this, the train pulls into the platform at the station, the driver completes their sign-off, then steps out of the cab and presses a plunger on the platform. This signals to the next driver (who is already waiting at the other end) that they can get into the cab at the outbound end and start their prep to drive the train back out. The first driver then walks the length of the platform, ready to drive out the next train that comes along. All this time, people are alighting and boarding the train, and it's all ready to go within a couple of minutes without ever having to turn the train around or actually waiting for the driver to walk the entire length of the train.
Speaking of frequency there is actually another compromise, which is capacity. That’s because when you’re designing with frequency in mind you’re more likely to choose smaller trains, which is not a good thing for a line that has the capacity of 2 million people a day.
It‘s great there is science for how to move people into the train. Our regional train service based on a bombardier talent 2 is certainly no good example. There is a wheelchair enabled bathroom right at the bicycle door which creates a smaller walkway already, yet the bottleneck can be used as bike storage on one side and flip down seats on the other. And _then_ there is the large bike/push-chair/ wheel-chair area. The washroom is needed, but don‘t invite people to sit there / or park there bike in a crucial entry point into the train
Oh - these trains are overcrowded and always to late.
Does Open Track have a free online version for testing this or was that a joke Reece?
You must come to know São Paulo metro system. Lines 3 is a non-stop carrousel.
When buses run every fifteen minutes, we call it off-peak.
California HSR changes plan from LA to Anaheim. 33 miles in 45 minutes doesn't sound fast at all. Thoughts?
Even 45 miles in 33 minutes wouldn’t be “high speed.”
I think you'd love to know that in Barcelona Sants, the suburbian railway trains enter and depart from the station at the same time but in the same track. I've also seen sometimes that a train enters and goes to the end of the platform, and then another train enter the same platform but stops at the beggining of it (because the trains are half the length of the platform) I don't think that's efficient or safe, but at least it's interesting!
It can be both efficient and safe, its known as double berthing!
I think one thing you're forgetting is that you have to staff all these trains. Each extra train you put on the route will cost three or four extra driver salaries per year. This is why driverless metro systems are such a game changer.
6:47 how do I open up Open track?
I'm really interested in simulating train movments
You did not mention the Victoria line with fantastic frequency sometimes almost a minute between trains, stand on the platform as a train enters the tunnel another train is entering the other end 40 seconds later. JH
6:36 - That's exactly how it's done at the southern (Waterloo) end of the Waterloo & City Line, while Bank doesn't have reversing sidings and - instead - the drivers change ends when the train is sitting in the platform.
That sort of layout could also be used as future-proofing in case the line is ever extended, but that won't happen to the W&C.
As for the Spanish Solution, I've seen it used at only one station here in New Zealand. The current Newmarket station (on Auckland's commuter network) has three tracks, two island platforms and four platform faces. When the 'AM class' electric trains pull in on the middle platform, the doors open on both sides, which means passengers don't need to use the escalators if they end up needing to quickly get from one side of the station to the other.
There are trains every 65s during peak hours in my city (Toulouse). VAL is really a great system. Funny enough, no modern CBTC system runs trains that frequently (at least not in sustained operation).
By the way,that Spanish Solution was actually invented by the Pennsylvania Railroad,when they designed Jamaica Station in the early 1900's! The use of platform trains,is still used to this day,and Jamaica is working on its second century! Thank you 😇 😊!!
Nope existed in many stations long before 1900
can confirm one case for @nilskoresaar8710 . South Brisbane Junction (now Corinda) in Queensland had this by 1888, and probably from 1875. Not used to speed up dwell times, but more of a convenience so that passengers didn't have to cross track to a branch service.
@@mrewan6221 The sheer amount of history,I was unaware of,but typical! The US histories tend to obscure the other users of similar operations! Thank you,for your courtesy,and information! Thank you! 😇! Thank you 😇 😊!!
Re: bus/tram doors. Here in North America, we rely on there being a single point of entry to ensure that riders pass by the fare box and pay. In much of Central Europe, you can board on any door and validate your ticket at many points inside the vehicle. Fare enforcement is by undercover inspectors.
Given the number of firearms in circulation here in the USA, I don't think you're going to end up with a lot of enthusiasm for people to demand fare payments from people carrying concealed weapons.
There are just never are going to be enough inspectors to drastically reduce fare evasion tho. London tried the enter at all doors approach with the new Routemasters but ended up going back to the the old system because it was costing them over £3.5 million in lost fares every year
we have that even in Russia
For resiliency, you usually want to add some storage tracks along the route, where trains can be parked e.g. if there is a defect. A major drawback of turning loops vs. the other designs is that it's much harder to include additional tracks, where delayed or defective trains can be overtaken.
in athens, there is a railway corridor, but at least until 2025, it will be single tracked in a section. In a european capital. The government gives money to the airplanes for safety from turkiye, but i don't think that there will be a war with turkiye anytime soon. Anyway, when, finally, the corridor is built there might be suburban trains frequently in this underused corridor (greek railway is among the worst in europe, and its basically abandoned)
RUN MORE TRAINS! (btw "fewer" is the opposite of "more" for count-nouns)
Regarding the reversing sidings at 6:30 -- the Canberra light rail does something like that, except funnily enough the platforms at both termini are *after* the double crossover, and there's no guaranteed "deboarding" and "boarding" side. Whenever you're approaching the terminus, it's 50/50 whether your tram will approach the left platform or right platform (and the crossover point is pretty tight so it's kind of uncomfortable for passengers when the tram switches sides). When you're boarding the service at the terminus, if there's a tram at both platforms, you have to refer to signage to figure out which tram is going to depart first. The overall effect on the drivers is the same as a regular reversing siding, but it's less comfortable and less convenient for the passengers. Still, both ends of the light rail are currently in very busy urban centres, so there probably wasn't room to extend the tracks at all -- hopefully if the tram line gets extended to less busy places with more room for the tracks, they could set up a reversing siding arrangement.
Need to persuade the transit authorities that designing a wheelchair ramp that doesn't take a full 60 seconds to deploy is a system wide timetabling issue
Interesting, but I think you overlooked one aspect: passengers and passenger discipline. I commute daily in Berlin, and people are just too stupid? Lazy? Obnoxious? For frequent service. Doors are being held; stepping into the middle of a car is rare; doors are blocked. And everybody grumbles about train service, of course 🙄. Anything that could be done for this problem?
Terminals are also built like this because when you need to extend the line, all you need to do is... extend the line. That's what they've done with line 14, considering they knew it would get extended eventually (I'm not sure how far ahead the planning was about line 14 tbh), all they need to do is extend the tunnel and turn the reverse tracks into a regular track in the system.
Service pattern is also incredibly important. If you want to run high frequencies, then running a single all stops service patten is probably the ideal.
If you absolutely, positively must run a mixed express/local pattern, you need to understand how this impacts frequency and capital cost. A New York system with fully segregated express tracks allows you to run high frequency express trains against high frequency local trains. But quad tracks are expensive to build and more expensive to maintain, money which has to come from somewhere. Alternatively, a Japanese style system with passing loops allows you to run high-frequency express trains against low frequency local trains without needing very much extra track. But this leaves the non-express stations with only infrequent (and probably inconsistent) service, which may or may not be desirable either.
What you mention at 9:00 is interesting (the widening of an approach track to 2 with an island platform at a through running mainline railway station) because Brussels never implemented this at its Central station, which has now become a notorious bottleneck on the Belgian railway network. That said, the issue hasn't been replicated everywhere as Brussels-Luxembourg station has 6 tracks (with 2 island platforms and 2 side platforms) even though there are only 4 approach tracks.
I have an example: Im living near Berlin and 2006 the New S-Bahn came every 20min. Almost nobody used it... Then they decided to create an 10min frequency and the the rate of passangers exploded.