Coupled Approximation of U.S. Driving Speed and Volume Statistics Using Spatial Conflation and Temporal Disaggregation

Eric Wood, Jeffrey Gonder, Kartik Kaushik

Research output: Contribution to conferencePaper


The advent of mobile devices with embedded global positioning systems has allowed commercial providers of real-time traffic data to develop highly accurate estimates of network-level vehicle speeds. Traffic speed data have far outpaced the availability and accuracy of real-time traffic volume information. Limited to a relatively small number of permanent and temporary traffic counters in any city, traffic volumes typically only cover a handful of roadways with inconsistent temporal resolution. This work addressed this data gap by coupling a commercial dataset of typical traffic speeds (by roadway and time of week) from TomTom to the U.S. FHWA's Highway Performance Monitoring System database of annual average daily traffic (AADT) counts by roadway. This work is technically novel in its solution for establishing a national crosswalk between independent network geometries using spatial conflation and big data techniques. The resulting product is a national dataset providing traffic speed and volume estimates under typical conditions for all U.S. roadways with AADT values.
Original languageAmerican English
Number of pages16
StatePublished - 2018
Event2018 Transportation Research Board (TRB) Annual Meeting - Washington, D.C.
Duration: 7 Jan 201811 Jan 2018


Conference2018 Transportation Research Board (TRB) Annual Meeting
CityWashington, D.C.

NREL Publication Number

  • NREL/CP-5400-70327


  • commercial probe speed data
  • conflation
  • highway performance monitoring system
  • HPMS
  • spatial join


Dive into the research topics of 'Coupled Approximation of U.S. Driving Speed and Volume Statistics Using Spatial Conflation and Temporal Disaggregation'. Together they form a unique fingerprint.

Cite this