vbb-change-positions

Crowd sourced database of station-specific platform positions for changes between VBB (Verkehrsverbund Berlin-Brandenburg) metro, suburban and regional lines.

You're invited to help, see the data structure and contributing sections!

Installation and Usage

If you're using JavaScript, you can use the package by installing:

npm install vbb-change-positions

const readChangePositions = require('vbb-change-positions')

const changePositions = readChangePositions()
changePositions.on('data', console.log)

changePositions will be a readable stream in object mode.

{
    fromStation: {
        id: "900000045102",
        name: "Heidelberger Platz"
    },
    fromLines: ["S42","S46"],
    previousStation: {
        id: "900000044101",
        name: "Hohenzollerndamm"
    },
    fromPosition: 1,
    toStation: {
        id: "900000045102",
        name: "Heidelberger Platz"
    },
    toLines: ["U3"],
    nextStation: {
        id: "900000045101",
        name: "Rüdesheimer Platz"
    },
    toPosition: 1,
    samePlatform: false
}

Browser support

Using bundlers like WebPack or browserify, this package will work in the browser. Keep in mind that it contains quite a lot of data (~200k uncompressed, ~10k gzipped).

Data structure

The dataset is located in data.ndjson, a ndjson file which you can edit using a text editor or vbb-change-positions-cli.

Let's take the following example: I want to take the S46 in the direction of Königs Wusterhausen to Heidelberger Platz, where I want to change to the U3 towards Krumme Lanke.

The dataset row would then contain the following information:

key name	description	required	example
fromStation	Station at which you arrive. Object containing the keys below.	yes	{id: "900000045102", name: "Heidelberger Platz"}
fromStation.id	Arrival station ID*	yes	900000045102
fromStation.name	Arrival station name (only for readability of the dataset)	no	Heidelberger Platz
fromLines	"Arriving" line names. Note that the lines must run on the same platform and share the same fromStation. For the lines that fit those criteria, there should exist only one dataset row grouping them, though. This key exists mostly because information on fromTrack is much harder to get/remember than the rather simple line names and sometimes the track changes but the platform doesn't.	yes	["S42", "S46"]
previousStation	Previous station on the line before changing. Object containing the keys below.	yes	{id: "900000044101", name: "Hohenzollerndamm"}
previousStation.id	Previous station ID*	yes	900000044101
previousStation.name	Previous station name (only for readability of the dataset)	no	Hohenzollerndamm
fromTrack	Arrival platform (track)**	no	empty
fromPosition	Number where to leave the arrival platform. Between 0 (at the rear end of the station) and 1 (at the front "driver's" end of the station) ***	yes	1
toStation	Station at which you depart. Mostly the same as fromStation, but different for interchanges like Kaiserdamm <-> Messe Nord/ICC that are marked as one node on the public transport map. Object containing the keys below.	yes	{id: "900000045102", name: "Heidelberger Platz"}
toStation.id	Arrival station ID*	yes	900000045102
toStation.name	Arrival station name (only for readability of the dataset)	no	Heidelberger Platz
toLines	"Departing" line names, see fromLines	yes	["U3"]
nextStation	Next station on the line after changing. Object containing the keys below.	yes	{id: "900000045101", name: "Rüdesheimer Platz"}
nextStation.id	Next station ID*	yes	900000045101
nextStation.name	Next station name (only for readability of the dataset)	no	Rüdesheimer Platz
toTrack	Departure platform (track)**	no	empty
toPosition	Number where to enter the departure platform.*** See also fromPosition.	yes	1
samePlatform	Set to true if both trains stop at the same platform (entire platform, not "only" track). fromPosition and toPosition will be ignored and should be set to 0.5	no	false

* See this document if you don't know how to find out some station's VBB station ID

** If unknown, just leave empty/unset like I did here.

*** see additional guidelines

Finally, our example would give us the following data row for the NDJSON file:

{"fromStation":{"id":"900000045102","name":"Heidelberger Platz"},"fromLines":["S42","S46"],"previousStation":{"id":"900000044101","name":"Hohenzollerndamm"},"fromPosition":1,"toStation":{"id":"900000045102","name":"Heidelberger Platz"},"toLines":["U3"],"nextStation":{"id":"900000045101","name":"Rüdesheimer Platz"},"toPosition":1,"samePlatform":false}

Additional guidelines

Parallel lines

There's a common situation where multiple lines run in parallel for a few stations in a row, like S41/S42 and S46 between Westend and Neukölln or S1 and S2 between Yorckstraße and Bornholmer Straße. A general rule of thumb for this case is that only if it's your last chance to change between two lines at a station, this specific interchange scenario should be mapped. However, if it's a common things for commuters to change at another station than the "last possible", then and only then that station should be added as well. To give you a few examples:

• Neukölln should be mapped for S42@Hermannstraße -> S45/S46/S47@Köllnische Heide, S45/S46@Hermannstraße -> S42@Sonnenallee and S41/S42@Sonnenallee <-> S45/S46/S47@Köllnische Heide, making 4 interchange scenarios. (For simplicity, I ignored the U7 in this example, that should be mapped as well, of course).
• Westend should be mapped for S46@Messe Nord/ICC -> S41@Jungernheide, because the S46 terminates here, making it the last possible point of interchange. No other interchanges should be mapped for this station, though.
• Plänterwald should not be mapped at all, because it's not your last chance to change between any of the lines running there.
• Priesterweg should be mapped for 4 scenarios: S2@Südkreuz -> S25/S26@Südende, S25/S26@Südkreuz -> S2@Attilastraße and S2@Attilastraße <-> S25/S26@Südende
• Heidelberger Platz should be mapped for interchanges S41/S42/S46 <-> U3, but not for interchanges like S41 -> S46 or even S41 -> S42
• Flughafen Berlin-Schönefeld Even though technically the last station to "change" between S9 and S45, no interchanges should be mapped since both lines terminate here. For places where only one line terminates while the others continue, see the Westend example.
• Jannowitzbrücke Even though technically not the last chance to change between S-Bahn and U8, the interchange S3/S5/S7/S9@Alexanderplatz <-> U8@Alexanderplatz should be mapped, since a lot of people prefer to change at Jannowitzbrücke instead of Alexanderplatz because the ways are shorter and it's less crowded.
• Wannsee Same with this station. Even though one might assume people would change between S1 and S7 at Nikolassee, it's actually much more common to change at Wannsee, because trains stop at the same platform there. So S7@Griebnitzsee -> S1@Nikolassee should be added here.

Other

• If you're not too sure about the exact position on the platform, just take one of 0, 0.5 or 1 that fits best, in order to prevent us from having data that seems really accurate but actually isn't.
• fromStation and toStation will mostly be the same, except for connections like Messe Nord/ICC and Kaiserdamm.
• If there are multiple ways connecting to platforms, either add separate rows for all of them or just add the shortest connection.
• fromLines and toLines must be different, previousStation and nextStation can be identical, however.

Contributing

Please note that this dataset only contains data for metro, suburban and regional lines; buses and trams are not included.

Have a look at this list for an overview of stations that are not yet covered.

If you want to add information to the dataset, fork this repository, add information and finally submit a pull request. If you don't know how any of this works, you can also just open an issue with the information you want to add in text form and I'll add it to the dataset for you. The same applies if you have found an error or want to change anything about the data structure.

Please note that by contributing to this project, you waive any copyright claims on the information you add.

License

This dataset is licensed under the ODbL license (v1.0).