United States Department of Transportation
i
Improve Data Quality for Automated Pavement Distress Data Collection
-
2024-08-01
-
[PDF-7.26 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report 09/2020-08/2024
-
Corporate Publisher:
-
Abstract:Since Fiscal Year 2017, the Texas Department of Transportation (TxDOT) has employed advanced automated and semi-automated techniques using 3D laser technology and high-resolution cameras to collect pavement condition data over 197,000 lane miles. Although this approach marks a significant technological advancement, concerns about the accuracy and precision of the collected data persist. To address these issues, TxDOT implemented a comprehensive quality management plan (QMP) that includes equipment calibration, weekly quality control verifications, and a dual-layered quality assurance (QA) process involving third-party visual assessments and internal rechecks by TxDOT personnel. This study evaluated current automated pavement condition data collection techniques through a literature review and a survey of highway agencies. It also analyzed four years of historical pavement condition data from 25 districts, examining accuracy, precision, and distress evaluations across different pavement types. Stratified sampling methods were used to improve data quality audits, ensuring representativeness in scenarios with high variability among population units. A pilot study in the San Antonio district validated the proposed QA framework, proving effective in identifying data quality issues. The findings highlighted significant challenges, such as inconsistencies and false positives, which compromise data validation. The necessity of manual validation and the establishment of robust quality assurance protocols were emphasized to ensure data reliability. The study underscored the importance of transparent benchmarks and stratified sampling methods to enhance the robustness of pavement condition data collection and management.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
