Asphalt Pavement Performance Analysis Using “Big Data” Computing Approaches

Abstract

Abstract Rapid deterioration of pavements depends on both objective and subjective factors. This study investigates the relationships among six parameters—climate, traffic volume, distress, friction, longitudinal profile, and transverse profile—related to asphalt pavement performance. For this purpose, LTPP InfoPave, the largest pavement performance database provided by the Long-Term Pavement Performance (LTPP) program, is used. The LTPP InfoPave data exhibit the characteristics of five V’s (volume, velocity, variety, value, veracity) of “big data.” The correlation analyses are conducted on the basis of the LTPP program’s Specific Pavement Study-3 experiments. Analysis results for the data from 1987 to 2005 reveal that the average block cracking area and average raveling area are highly correlated with a correlation coefficient of maximum 0.85. The average mean roughness index (MRI) is correlated with the average block cracking area and the average raveling area with correlation coefficients 0.59 and 0.64, respectively. The results also indicate that smoothness and surface distress have relatively lower correlation; these results are consistent with the previously published results. In addition, for the traffic volume data from 1990 to 2014, the MRI is found to be higher at higher traffic volume, indicating rougher pavement, but this difference is statistically insignificant (correlation coefficient, 0.423). At the beginning and end of the section, traffic volume is highly correlated with friction, having correlation coefficients of 0.711 and 0.646, respectively. This study serves as a reference for the use of big data to gain an in-depth understanding of pavement performance, and engineers can continue to explore the application of the LTPP program using big data computing approaches.