Rail transportation data, from network geometry to live departures.

Plus 2 more in this slice — each with its field dictionary and sample rows on its own page.

• Data Scientists & ML Engineers Rail transportation data for data scientists spans 13 datasets Datadory qualifies at relevance 2 or better, averaging 8.
• Developers & Data-Product Builders
• Market Researchers & Consultants Rail transportation data for market researchers comes from 15 qualifying sources led by Eurostat's commercial delivery terms 4.

Start with the unit of record, because this shelf refuses to flatten to one grain.

Questions about infrastructure resolve to polylines - 302,771 of them in the American network, millions of mapped elements globally.

Questions about journeys resolve to scheduled stops and their realised counterparts, stop-level in the Swiss and French collections, per-train station calls in Britain.

Questions about volume resolve to statistic rows: country-year and country-quarter tonnage, or a single national total per year at the macro level.

And questions about material condition resolve to photographs, point clouds and text rows.

Then weigh geography honestly.

Only the geometry layer spans borders cleanly - United States, Canada and Mexico in one file - and only the metadata catalogue reaches every member state at once.

The national shelves go deeper inside their fences: Britain for performance and finance, France for ridership, Germany for facilities, Switzerland for timetable archaeology, Brussels for freight comparisons.

Finally, match the clock to the question.

Live movement data exists for Great Britain today and tomorrow; timetable horizons roll months ahead; freight statistics land quarterly; accident history compounds across five decades; and the training corpora are fixed snapshots that will not correct themselves.

Frequently asked questions