Skip to main content

Ottawa transit rider information

Otranspo
FR
Back to rider information

Data methodology

Every rider-facing result identifies its source, observation time, confidence, and unavailable state.

Departures and predictions

Scheduled departures come from the currently imported OC Transpo and STO GTFS service calendars. A realtime departure is shown only when a trip update from the matching operator can be matched to the scheduled trip and stop.

Cancellation and skipped-stop relationships are displayed explicitly. Failed, malformed, unexpectedly empty, stale, or partially failed feeds are not called healthy. When live data is missing, the schedule remains labelled scheduled.

Trip planning and accessibility

Trip plans come from the configured OpenTripPlanner graph built from the imported transit feed and routing data. Otranspo does not invent a fallback itinerary.

Wheelchair and step-free preferences are sent to the routing engine. Each returned leg retains the source accessibility state; missing states remain unknown.

Public report patterns

Individual reports are private. Public patterns include only moderator-confirmed non-safety and non-operator reports grouped by category and route, with at least five reports in each cohort.

A public count is community-submitted evidence, not an operator-confirmed incident count.

Operational status

Status is computed from direct probes of the database, current migrations, imported schedule, routing engine, independent realtime feeds, durable worker, email, storage, geocoder, and required configuration. Optional components degrade independently.

Reliability method correction

On 4 August 2026, the on-time method changed from counting repeated archived predictions to one settled observation per arrival: the last prediction within ten minutes of scheduled time, evaluated with a two-sided five-minute window. On the same archive, the result changed from 86.7% to 77.4%. This is a correction to the method, not evidence that service changed on that date.