Comparison of Pavement Roughness Indicators from Traditional Inertial Profilers and Emerging Connected Vehicle Sensors
Year: 2026
Type: article
Abstract: This study is an evaluation of the consistency and potential of crowdsourced connected vehicle (CV) data as an alternative to traditional inertial profiler (IP) measurements for road roughness evaluation. IP data were collected from three roadway corridors, SH 21, SH 6, and FM 2818, in Bryan, Texas, which comprised a variety of pavement types and functional classifications. Lane-specific international roughness index (IRI) values were recorded using a calibrated inertial profiler and analyzed at every 10 ft. These were compared with direction level ride values collected at an average spacing of 82 ft from a third-party CV data provider. A general agreement was found between IP and CV data on asphalt and concrete surfaces, but there were substantial differences between seal coat sections, where CV ride values overestimated roughness by 80–100 in./mi. A grouping analysis was conducted to compare segments categorized by roughness level from both datasets. The results showed a moderate match in identifying the roughest segments by the CV system, compared with those identified by IP. This match rate varied by profile and improved with broader grouping sizes. Results show that CV data might have greater sensitivity to vehicle behavior (for example, braking at intersections), while IP-based systems are calibrated to record actual road roughness. This could be one of the reasons for the moderate mismatch in identifying rougher segments using the two methods. Despite these limitations, the CV system’s high frequency of measurements, especially on highways, demonstrates its potential for cost-effective, network-level pavement monitoring.
Authors: Vivek Kumar Gupta, Joel Ben Thomas, Charles Gurganus, Michael Martin, Shawn Turner, Nasir Gharaibeh
Institutions: Texas A&M University, Texas A&M University System
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,650
OpenAlex ID: W7162946432