Turn your agency's NPMRDS export into the corridor numbers the Congestion Relief Program's Criterion 1 asks for: the travel time index, the planning time index and federal reliability. Your files are read by your browser and never uploaded.
Your files
In RITIS, use the NPMRDS Massive Data Downloader to export your corridor's segments for a full year, with 15-minute averaging and travel time in seconds. You'll get Readings.csv and TMC_Identification.csv. A second export from the trucks data set adds truck reliability.
Peak hour excessive delay (optional)
Delay needs each segment's posted speed limit, which your state's HPMS submission records by section. Load a file with tmc and speed_limit columns (mph), give one speed limit for every segment, or both: the file's limits come first. Traffic volumes come from the AADT in your TMC file.
Volumes by hour follow FHWA's continuous count stations in the corridor's state (2025), on urban Interstates and freeways where the state has enough of them, as the TMC file's state column says. Puerto Rico has no stations in FHWA's data, so San Juan uses the national pattern. Your agency's own counts would be better still.
For your application (optional)
Criterion 1A is also met when the corridor misses the PM3 targets your State DOT or MPO reported. Enter them to get a verdict; FHWA's State Performance Dashboard lists each state's. An expected cut in delay prices Criterion 2A's savings, and your organization's name gives the downloads the file names DOT recommends.
Six made-up segments over 2025, labelled as synthetic everywhere, so you can see what you'll get. Or open the sample evidence pack (PDF).
No files yet.
Results
Load your two files, or the synthetic example, to see the corridor's travel time index, planning time index and reliability.
The evidence pack follows the NOFO's format for application files: single-spaced 12-point type with 1-inch margins (Section D.1). The location file is KML, which the NOFO accepts for project locations.
Bottlenecks, worst first
Every corridor in your files, analysed the same way and ranked: by the cost of delay when you give speed limits, otherwise by the planning time index. In April 2026 DOT asked governors to name their two to five worst bottlenecks under its Freedom to Drive initiative; this ranks them from the same NPMRDS data, with text for each and a verifiable PDF.
Load your files, or the synthetic example, to rank the corridors.
Half-year performance report
After an award, the recipient files semi-annual progress reports with the performance measures the grant agreement names (NOFO Section H.2). Load this half-year's export as the readings above, and the same months from before the project here. The report sets the corridor's measures side by side, with the method and checksums for both periods.
The synthetic corridor in 2025, against a more congested 2024.
Load your files and a baseline export, or the synthetic example with a baseline, to compare the two periods.
The report's narrative (optional)
What you write here goes into the PDF under its heading. Leave a box empty to get the heading with a prompt. Separate paragraphs with a blank line.
How the numbers are computed
- The travel time and planning time indexes follow the definitions in FHWA's Urban Congestion Report. Peak is weekdays 6–9 a.m. and 4–7 p.m. Free-flow is the 85th-percentile off-peak speed (weekdays 9 a.m.–4 p.m. and 7–10 p.m., weekends 6 a.m.–10 p.m.), so for a segment the 15th-percentile off-peak travel time.
- Travel time index: mean peak travel time over free-flow travel time. Planning time index: 95th-percentile peak travel time over free-flow travel time. Percentiles are nearest-rank.
- The corridor's travel time at each 15-minute interval is the sum of its segments' travel times, using only intervals where every segment has a reading.
- Reliability follows 23 CFR 490.511: in each of the four periods, the 80th-percentile travel time over the 50th, each rounded to whole seconds; a segment is reliable when every period scores below 1.50. Truck reliability (490.611) uses the 95th percentile, adds an overnight period, and covers Interstate segments.
- The reliable share is weighted by vehicle-miles (FHWA's one-way and two-way factors on AADT and NHS length) when every segment has AADT, and by length otherwise. Interstate and non-Interstate NHS segments are reported apart, as PM3 reports them, using the TMC file's f_system.
- Peak hour excessive delay follows 23 CFR 490.711 and FHWA-HIF-18-040. Peaks are weekdays 6–10 a.m. and 3–7 or 4–8 p.m. The threshold speed is the greater of 20 mph and 60% of the posted limit. Delay above the threshold travel time counts in each 15-minute interval.
- Volumes follow HIF-18-040's estimation method: AADT, halved on two-way facilities, times its national monthly (Table 1.1) and day-of-week (Table 1.2) factors, times the hour's share of daily traffic. Person-hours use FHWA's occupancy factors (1.7 per car, 1.0 per truck); buses are left out.
- Dollars use USDOT's Benefit-Cost Analysis Guidance (December 23, 2025; 2024 dollars): 1.52 people per car at $21.80 an hour, and truck drivers at $37.20 an hour.
The PDF lists each file's name, size and SHA-256, the segments and dates, and the engine version, so DOT or anyone with the same NPMRDS export can reproduce every number (the NOFO asks for data "in a form directly verifiable by DOT", Section E.5). Check your NPMRDS data-use terms before sharing it outside your agency.