- Reading
- Data methodology
- On-time percentage
The share of selected final provider predictions no more than five minutes early or late relative to schedule; arrivals are not independently measured.
- Definition
- The share of selected final provider predictions no more than five minutes early or late relative to schedule; arrivals are not independently measured.
- Formula
- 100 × COUNT(|settled prediction delay seconds| ≤ 300) ÷ COUNT(settled final provider predictions), rounded to one decimal place.
- Units
- percent of selected final feed predictions
- Population
- One last eligible provider prediction per scheduled trip, stop and Toronto calendar date; no independently measured arrival.
- Measurement window
- Route summaries: rolling 30 days. Route-hour evidence: recent 14-day scan.
- Coverage
- Matched published predictions only. Numerator and denominator describe selected final feed predictions, not measured arrivals or all scheduled or operated trips.
- Publication eligibility
- Regular comparable bus routes; a route-hour needs at least 100 observations across at least three service days. Rail and purpose-specific routes are excluded from ordinary rankings.
- Scope
- Route summaries use a rolling 30-day window. Recurring route-hour evidence uses the recent 14-day scan and is published only with at least 100 observations over at least three service days.
- What this number can support
- Schedule difference in final eligible feed predictions, not independently measured punctuality. It is not OC Transpo’s 99.5% service-delivery target, a cancellation rate, or proof that every scheduled trip ran.
- Sources and identifiers
- Method · Settled provider-prediction reliability method
last-prediction-within-ten-minutes-v1 - Current API · Current network reliability public API
reliability-evidence
- Method · Settled provider-prediction reliability method
Implementation status
- Source candidate
- Implemented in this source candidate.
- Public production
- The corrected metric version is implemented in source; public production deployment and verification remain unrecorded.
Read the status before citing this number.
The final provider-prediction calculation is implemented in this source candidate. Public production verification for this corrected version is still not recorded.
Exclusions
Predictions more than ten minutes before the scheduled arrival are excluded, and delays outside ±90 minutes are removed as timetable-wrap outliers.
OC Transpo rail is excluded because the operator does not publish realtime rail observations in the feed used by this method.
Uncertainty and interpretation
The result describes matched published predictions, not every operated trip. Feed gaps and trip-matching limits can reduce coverage.
Current network evidence is recomputed and is not stored here as an immutable historical snapshot.
Current evidence
Change history
| Revision | Date | Record status | Reason | Human approval | Production deployment |
|---|---|---|---|---|---|
| correction:reliability:2026-08-04:settled-v1:revision-1correction:reliability:2026-08-04:settled-v1 | Source complete; production pending | The prior method counted repeated predictions as independent arrivals and treated large early departures as on time. The corrected method keeps the last prediction within ten minutes of each arrival and uses a two-sided five-minute window. | Pending the named independent verifier; no approval evidence is recorded. | Pending; no production release identifier or verification time is recorded. |