Transit system planning
Before measuring on-time performance, define “on time”
Agencies need a clear window, timepoints, data rules, early-departure policy, and reporting method before comparing routes or operators.
An on-time performance percentage looks precise, but the result depends on choices made before the calculation. A route can appear reliable under one time window and unreliable under another. A useful study defines the standard, applies it consistently, and explains what the number does and does not show.
Set the acceptable time window
Define how early or late a vehicle may be relative to the published schedule. Consider whether any early departure is prohibited at designated timepoints and whether the same late tolerance applies to all services.
FTA technical assistance notes that transit providers must first define what counts as on time. The chosen window should be documented, approved, and visible in every performance report.
Further readingFederal Transit Administration: Title VI Fixed Route Transit Requirements transcript (opens in a new tab)Choose where performance is measured
Identify the route origins, destinations, and intermediate timepoints used in the calculation. Measuring only the first and last stop can hide early departures or delay that develops in the middle of the route.
Use timepoints that matter operationally and to riders, including important transfers. Keep the set stable enough to compare periods, while documenting schedule or route changes that require a new baseline.
Write the data-quality rules
State which data source is used, how missing or implausible records are handled, how detours and canceled trips are treated, and whether layover or deadhead movements are excluded. Preserve the raw record and the reason for every exclusion.
Automated vehicle-location data can create a large sample but still needs validation. A report should show the share of scheduled trips with usable observations, not only the final on-time percentage.
Break the result into actionable views
Review performance by route, direction, time period, day type, timepoint, and recurring operating condition. A single systemwide average may hide one direction that repeatedly misses a major connection.
Compare reliability with ridership, running time, operator feedback, road conditions, and customer contacts. On-time performance is one measure of service quality, not a complete explanation of why service varies.
Connect the measure to a decision
Define the threshold that prompts schedule review, supervision, traffic coordination, operator support, or another action. Assign an owner and a date for checking whether the intervention improved results.
Publish the method with the results so decision-makers and riders can interpret the trend. Contact Cathay to discuss performance measurement within a transit system evaluation and service planning study.
Your next step
Define a transit performance study
Connect with our team to discuss your needs and the available arrangements.