An extensive bibliometric analysis of pavement deterioration detection using sensors and machine learning: Trends, innovations, and future directions

Abstract

This study presents a current and extensive bibliometric analysis of pavement deterioration detection, monitoring, and assessment using various sensors alongside machine learning and deep learning algorithms. The impact of electronic sensors, machine learning, and deep learning on road pavement evaluation and monitoring within the transportation sector is highlighted. Conducting a bibliometric analysis covering research until March 1, 2024, 639 publications from 71 countries were examined. Productive countries, journals, institutions, and authors were analyzed and ranked. A standard research score and cumulative output score were calculated to normalize differences in the data. The findings reveal a significant recent increase in studies in this area. The most productive countries, journals, institutions, and authors are China, Transportation Research Record, Southeast University China, and Golroo Amir, respectively. This study serves as a valuable resource for both academic and industry researchers, offering insights into road pavement monitoring and guiding future research. In addition, accelerometer and GPS were the most used sensors, ANN and CNN were the most preferred algorithms, and cracks and potholes were the most studied topics. This study has the potential to be a good map for both academic and industrial researchers for monitoring the state of road pavements and a good guide.