Realtime data estimates seem off for trains
Hello,
I run the Danish site www.togtid.dk, which offers realtime train information about trains relevant to Denmark, including Swedish trains going to/from Denmark.
When I implement the SIRI realtime estimates for the Swedish trains I get what looks like quite a few data errors. Also when I compare to the official site on Swedish realtime train data. So I am wondering if I use the correct dataset. What dataset would you recommend for this?
Also I am interested in the Snälltåget planned schedule data and realtime data. What dataset do you recommend for this?
Thanks!
Best regards,
Jens
Togtid.dk
I run the Danish site www.togtid.dk, which offers realtime train information about trains relevant to Denmark, including Swedish trains going to/from Denmark.
When I implement the SIRI realtime estimates for the Swedish trains I get what looks like quite a few data errors. Also when I compare to the official site on Swedish realtime train data. So I am wondering if I use the correct dataset. What dataset would you recommend for this?
Also I am interested in the Snälltåget planned schedule data and realtime data. What dataset do you recommend for this?
Thanks!
Best regards,
Jens
Togtid.dk
Följ inlägget
2
följare
But for Swedish trains I still recommend going straight to Trafikverket APIs - their TrainAnnouncement API requires quite a lot of work but data quality wise when you get it right - it's at least my preference so far.
Best regards,
Jens
www.togtid.dk
When it comes to trains however, the GTFS data currently has some limitations.
- SJ and Snälltåget do not include track information
- SJ and Snälltåget do not have realtime information
We are working to make Trafikverket data available through GTFS Regional and GTFS Regional Realtime datasets so you can obtain all train data through one dataset, but this is not expected to be in place before 2027. In the meantime, you can use Trafikverkets API as Lars mentioned, where you can obtain all trains including realtime information and platform number. Their API allows you to filter events so you only get departures at those stations you are interested in. Note that Trafikverkets API is limited to Swedish stops and does not include Copenhagen etc.
In which (realtime) dataset are you seeing times which are off? Please let us know if there are any issues, so we can correct them. We do however need an example where a trip has an incorrect delay (including trip or serviceJourney id, actual delay + source for the actual delay, delay from our dataset, in case of realtime data with a snapshot of the realtime data file if possible).
Regards,
Bert