GTFS-History #616
Description
Describe the problem
There are a number of useful datasets that aren’t covered by GTFS-Schedule or GTFS-Realtime that could improve the user experience, simplify the existing specs, reduce complexity and data size, or create new experiences and support new use cases.
Use cases
Some use cases that aren’t supported:
- historical data and stats
- financial information
- stop, route and vehicle history and usage
- maintenance information
- historical alerts and a subset of other old realtime info
- usage, outages
Proposed solution
I propose a new zip file GTFS-History[-YYYY].zip (where -YYYY is optional for larger systems to chunk data by year) which would contain some of the following information:
| File | Description |
|---|---|
| Ridership.txt | Ridership counts, breakdowns and summaries optionally by - time (year, quarter, week, month, day), location (system, mode, line, stop, segment, vehicle), type (tap, cash, transfer, etc.). |
| Maintenance.txt | Maintenance activity log - time, vehicle/stop/line/segment, type (scheduled/unscheduled, activity (long list - brakes replacement, rail grinding, etc.) |
| Finance.txt | Financial breakdown and summaries by - time, location, income/expense, division/groups, etc. this could include things like fare income, advertising income, fuel costs, etc. |
| Alerts.txt | Historical alerts from GTFS-Realtime |
| Delays.txt | Time, Location ( including multiple stops), Length, Alternative Service provided, reason (ie. trespass, obstruction, etc.), links to investigation reports |
| Historical Routes and Stops | Duplicate design as gtfs schedules - used for retiring lines, but keeping information around if useful for historical reasons |
Additional information
If this idea is worth exploring, obviously it would need to be fleshed out and more work put into developing a spec that was compatible with the existing feeds and systems.
At this point I haven’t put much work or thought into the spec details, but am willing to put more effort in if the community sees value in something along these lines.