Author ORCID Identifier
Date of Graduation
Statler College of Engineering and Mineral Resources
Civil and Environmental Engineering
Dr. Yoojung Yoon
Dr. Kakan Dey
Dr. Fei Dai
Data collection on pavement surface cracks is crucial for the state Departments of Transportation (DOTs) to make informed decisions on maintenance, repair, and rehabilitation (MR&R) activities. The collected cracking data is translated into a pavement performance indicator to assess the condition of pavement sections and identify MR&R needs based on predetermined threshold values set by the state DOTs. State DOTs have transitioned from manual to automated data collection as automation offers many advantages over manual data collection, such as improved data collection rates, efficiency, cost reduction, and safety. As the state DOTs have integrated automated data into the pavement performance indicators developed based on manually collected data, the different capabilities of collecting data between the automated and manual-based approaches can lead to potential problems, such as underestimating or overestimating actual pavement conditions.
This thesis aimed to gain knowledge about the changes in cracking data collected before and after the automated data collection implementation. A nationwide survey was performed to understand the current automated crack data collection practices and the issues faced when using the data for pavement crack-related indicators within state DOTs. Subsequently, a case study was pursued to delve into the survey findings, the impact of discrepancies between data collected through automation and a crack-related indicator drawn from manually collected data, and the approach for indicator calibration. The case study was the Structural Cracking Index (SCI) of the West Virginia Division of Highway (WVDOH), developed for pavement surface crack evaluation and MR&R decisions.
A survey was conducted electronically among all the state DOTs, and 31 out of 51 state DOTs, including the District Department of Transportation (DDOT), provided complete responses. 22 out of 31 state DOTs use automated data collection vehicles for pavement surface cracks. Seven state DOTs have observed increasing trends, four state DOTs have observed decreasing trends, and eleven state DOTs have observed no trend in the cracking data collected before and after using the automated data collection vehicle. 45% of the state DOTs indicated the need for calibrating their current crack assessment indicators regardless of trends observed in cracking data due to implementing an automated data collection vehicle.
The case study of analyzing SCI and pertinent cracking data collected from WVDOH for 1997–2021 found that the cracking data showed a higher magnitude of changes in low- and medium-severity alligator and longitudinal cracking data compared to those of high-severity after employing the automated data collection vehicle. There was a similar pattern of changes in cracking data across the three highway systems (Interstate, US Routes, and WV Routes) on the state-owned pavements. The current SCI equation was calibrated and tailored to each highway system based on the data analysis. The calibrated SCI equations showed improved SCI ratings, a 0.18% increase for the Interstate, a 1.25% increase for the US route, and a 4.07% increase for the WV state route, based on the 2016 WVDOH data. This dataset year was selected based on the highest number of pavement sections available for a year. The case study also demonstrated the necessity of calibrating the existing SCI, leading toward better cost-effective decision-making for MR&R activities.
In conclusion, the study findings highlight the need and economic benefits of updating pavement performance indicators in light of the increasing trends in employing automated data collection methods. As the scientific contributions, the approaches used for this study to calibrate the WVDOH’s SCI equation can apply to other state DOTs’ pavement performance indicators.
Chowdhury, Faisal Quibria, "Investigating the Change in Pavement Cracking Data Due to Automated Data Collection Methods" (2023). Graduate Theses, Dissertations, and Problem Reports. 12031.