Research on influencing factors of asphalt pavement International Roughness Index (IRI) based on ensemble learning

Abstract

Abstract The International Roughness Index (IRI) is one of the most commonly used indicators to measure pavement surface smoothness. This paper uses the data obtained from the Specific Pavement Studies-3 (SPS-3) of the Long Term Pavement Performance (LTPP) program to study the influencing factors of the International Roughness Index of asphalt pavement. Pavement age, precipitation, freezing index, temperature, traffic volume, traffic load, and rutting depth are investigated to evaluate the effectiveness of four preventive maintenance treatments on asphalt pavement surface roughness. The pavement roughness model is established based on the XGBoost algorithm, with a training R2 of 0.96 and a testing R2 of 0.82. The results show that among the four preservation treatments, the IRI of thin overlay is the lowest. Annual Average Daily Traffic (AADT) is identified as the most significant factor for IRI evaluation, accounting for the most contribution to pavement surface roughness, followed by precipitation, rutting depth, temperature, etc.