Analysis and Verification Results of Manual Inspection of Pavement Condition Index

The aim of this study is to establish an analysis of certification indicators for the Pavement Condition Index (PCI) to address previously unaccredited issues. In the management of road infrastructure, the PCI serves as a crucial indicator for assessing road pavement conditions. However, the lack of standardized detection and accreditation mechanisms in the past may have led to variations in the reliability of detection results among different personnel. This study observes the accuracy and precision of the data obtained by different inspectors. The analysis is based on repeated manual inspections conducted by four inspectors across 12 road units on Zhongda Road, using the ASTM D6433-23 (2023) standard. To ensure the reliability of the inspection results, the Coefficient of Variation (CV) is used to measure repeatability, while the Mean Relative Error (MRE) is used to evaluate accuracy. Training was introduced between inspection phases to assess performance improvement. The results showed that the average CV decreased from 27.2% to 7.5% and the average MRE reduced from 45.3% to 5.9% after training. These findings demonstrate that targeted training significantly enhances the repeatability and accuracy of manual PCI assessments. This study concludes with recommendations to set a CV threshold of ≤0.1 and an MRE threshold of ≤10% as practical benchmarks for manual PCI Quality Control. These standards can serve as validation criteria for future automated inspection systems.