When It’s Okay for Transit to Not Be on Time

Transit systems are super complex, and sometimes by trying to make onething better we make another *much* worse. Transit systems need to learn to optimize for the public good and not arbitrary metrics! Find out more in today's video.

Check out @therailcinema 's amazing channel!

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.

=ATTRIBUTION=

=COMMUNITY DISCORD SERVER=

(Not officially affiliated with the channel)

=MY SOCIAL MEDIA=

=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.

[RMTransit]

Some great comments on this video!

One important consideration is that the rail traffic controllers don't always have information as fine grained as what you can see on the ground (well they never do) so what appears "obvious" might not be so! But, in this case based on the location they would have fine grained info - so its not a "train was in a 10 mile block and could have been anywhere!" situation.

That being said, better signalling and policy can certainly help reduce situations like this!

[SniperToHeadshot]

As a train traffic controller for SBB (Switzerland), maybe I could offer some insight how this works in Switzerland. Timetable padding is extremely route and line dependent. Some trains have next to no padding at all (example: IC3 between Zurich and Sargans already does next to the maximum line speed to achieve a one hour "Takt" between those cities and can only catch up 2 minutes) but other lines can have as much as 6-8 minutes of padding.
This is also dependent on the infrastructure used; trains running on a single-track line often have to wait for a train running in the opposite direction in order to reach a passing loop or station.
In Switzerland, trains wait a maximum of 2 minutes for connections because waiting longer would mean a lot of delays for the next trains, as our infrastructure is next to full capacity. On the other hand, because of our clock-face scheduling, even if you miss your connection, the next train leaves no later than 30 minutes after, so you never have to wait very long.
This is possible in Switzerland mainly because our trains run comparatively slowly when you look at Italy or France with few exceptions.
There's also a rule that a train running late is usually "sacrificed" in order to keep the other trains on time, as this usually reduces delays over-all.

[Ruhrpottpatriot]

What many people often miss out (because they are not specialists in the area), is that 4 tracks don't have double the capacity of 2 tracks. They have more than that. So by splitting your tracks into express and local, you're actually missing out. Best case scenario would be to build the flyover (or tunnel) and still connect the local tracks to the express tracks, so that you can have faster local trains take over slower ones.

[ghamerons]

Bruh just got jumpscared by seeing my local trains in the thumbnail

[therailcinema]

Loved your perspective and honored to have been a part of this!

[AverytheCubanAmerican]

I'm glad you did a video on this. While yes, we may not like it when a subway train is a little late, we have to remember that we shouldn't take things for granted. The MTR in Hong Kong or Japanese trains apologizing when it's two minutes late? That's a privilege! 5-to-10-minute headways (especially on a key line at rush hour) in NYC may lead to capacity constraints (not to mention being an oven during the summer), but when compared to service in other American cities (like Boston, SF, or Chicago)...every five minutes sounds amazing. This same thing goes for the LIRR. The LIRR is by no means perfect, but it's a masterpiece when compared to Minneapolis's Northstar or Nashville's WeGo Star, especially when it comes to frequency.

Those who live in cities don't know how much they have it with walkability, bike-friendly, and transit-oriented infrastructure. Moving from Jersey City to Long Island, it made me realize just how much I had it. Living in a car-oriented suburb without a car open my eyes to why advocating for urbanism throughout North America is so important. Drivers may think having a car is freedom, but having good sidewalks, having walkability and convenience, being able to do all your errands thanks to a five-minute walk or transit ride, is true freedom.

[krakowtransit]

Polish railway system has this thing where an on-time train can depart up to 10/15 minutes after planned departure to wait for a delayed train to provide transfers. It doesn't always work, but it saved me several times

[NickBurman]

The Yamanote Line in Tokyo is an example of "timetable relativity". All services on the line are theoretically timetabled; when services start early in the morning trains manage to maintain the schedule, however as rush hour kicks in and delays start piling up (because of a stuck briefcase on a closing door, a passenger which has felt unwell, delays in door closure due to number of onboard passengers, etc...), Control (and the Yamanote has its own set of controllers) chucks the timetable out of the window and works to keep the trains flowing into stations as frequently as possible (minimum 2 minutes). Also, to avoid the risk of injury due to potential tumults, control occasionally asks stationmasters to reduce the number of or even stop the turnstiles for a few minutes, to allow trains to clear the platform of passengers. By the time evening rush is over, service might be out of synch with the timetable by as much as 10 minutes, yet Control keeps frequency-feeding trains into stations until service ends at 1AM as it would be useless to try and set things back to rights. The recent installation of digital ATC and the moves by JR East to automate the Yamanote line might change this situation in the future.

[IIVQ]

Increasing timetable padding increases the required time for a return trip, which might mean requiring more trains for a given frequency. So removing the padding and occasionalky paying out a sum to late passengers might save the company money.

[MarioFanGamer]

Watching trains, knowing their schedule and which stations have passing loops, it's always interesting to guess when a train will overtake a different one. Here in Germany (where delays are infamous), there is a general hierarchy of intercity to regional (itself separated into express and local, albeit not so much as a strict rule) to freight trains in terms of (daytime) priority and it's easy to see when non-stop trains (including freight trains) have to drive slowly because they first need to pass a local train first which may take a while until they reach a passing loop. And YouTube commentors also mention that generally the slower trains have to wait which to me sounds like to not mess up the schedule of the other trains any further.

Also +1 for stations with platforms on passing loops only. It's typical for high speed stations (in part because of security reasons due to the high winds caused by trains driving at high speeds) but even non-high-speed lines benefit from these _because_ they enforce timed overtakes.

[aaabatteries]

In France we had local shuttles for TGV arriving passengers depart empty before the late running TGV arrives just to have good regularity number

[ricequackers]

The trick is to set different delay refund levels for different levels of delay. For Thameslink for example (which I'll speak about because it's my commute service), they pay out 25% for 15 mins, 50% for 30 mins, 100% for 60 mins and 200% for a 2 hour delay. That gives them enough flexibility to delay a train by a bit to ensure others can keep to their timetable without having to pay out.

As for the topic of running on quad track, line running is used on the Midland Main Line (shared with express East Midland services and the occasional freight train). The Thameslink trains usually do a carefully timed crossover from the fast to the slow lines at my station, just in time to get out of the way of the express. However, if the commuter train is late in arriving, the express gets priority so it can stay at max speed, and only then does the former get to cross over. Often the signaller will cross the train over to the slow line at an earlier junction so that neither train has to completely stop.

Oddly enough the East Coast Main Line that Thameslink also uses (among several others) uses directional running, so it doesn't usually have this issue at all.

[HaapsaluYT]

As much as I agree that we should allow a few delays, I physically cringe every time a Metro-North train shows up at my stop 30 minutes late

[Taitset]

Great video Reece! In Melbourne we have a thing where the private operator gets fined if the service runs late (5 minutes I think), which has definitely resulted in some questionable priorities over the years. For a while there was a policy of altering late running trains to run express for part of the line to make up time - which would save maybe a couple of minutes for those already onboard, but effectively cancel the train for everyone at the intermediate stations.

[ProfessorPancakes]

Spot on. You see it all the time with Amtrak running along the Northeast Corridor fighting with Metro-North and NJ Transit trains.

[nomadMik]

Then there's the transfer issue. The number of times I was on a BART train that was a few minutes late, and thus tried to sprint for the connecting FailRail train down the Peninsula, only to have the doors slam in my face and hear FailRail tell three dozen angry passengers that somebody down the line might be annoyed at leaving a minute or two late and it was better for us to wait FORTY MINUTES for the next train down the Peninsula… yeah, let's just say there's a reason I avoid CalTrain as much as possible these days.

[rtyschannel]

5:15 I've seen similar situations in the UK. Where I was on a train that was running direct from Reading into London Paddington. That train was held at Reading to let a train that was running late. Which I might understand except that train that also runs on the main line and stops at Slough. So now the train that doesn't stop at Slough needs to crawl behind that late train that does stop there.

[UchennaKema]

Its wild watching this from NJ, because we have to have all of these since we have freight and passenger sharing main lines along with express (Thank you Conrail very cool). Timed overtakes are also required as the NEC in NJ has a mix of 150mph very fast trains, 80mph slow old EMUS and 100mph express locomotive lines

[MrVlad]

When you said overtake tracks I remembered a very funny but sad situation that we have in Greece. A few decades back the Larissa - Thessaloniki section of the Athens - Thessaloniki got modernised, with double tracks, new high-speed-like alignment and new stations with overtake tracks. Then it came time to operate the line, which is where things went downhill... Remote control wasn't installed on the switches of the overtake tracks, which means that the switches have to be operated locally by the station master.
Unfotrunately, due to austerity and mismanagment there aren't enough station masters to go around for all the stations that need them. Which in turn means that for the local trains to be able to stop at a station with overtake tracks but no station master the switches have to be permanently pointed at the overtake tracks. Thus forcing the fast trains to slow down for the stations...

[domtweed]

Thanks for using a British train as the thumbnail. We appreciate the accuracy.