.png){kind=icon}
Return
Evaluating Pavement Surface Texture with LTPP Database
.png){kind=icon}
PRE
AI
DOI：10.1007/s42947-025-00675-5.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Pavement surface characteristics play a pivotal role in enhancing driving comfort and road safety, while surface texture is believed to correlate with these diverse characteristics. In this study, pavement surface characteristics, texture and relevant variables from 679 asphalt pavement sections across 46 U.S. states were extracted from the Long-Term Pavement Performance (LTPP) database. Five texture indicators and twelve explanatory variables, including pavement roughness, traffic volume, climate condition, aggregate properties mixed with gradation and volumetric properties were compiled and summarized. Comprehensive descriptive statistics were presented to understand the dataset. Afterward, the correlational heat map of these variables was generated to exclude the parameters with strong correlations and remove their potential of multi-collinearity. Statistical regression analysis and machine learning models—random forest and gradient boosting algorithm—were developed to determine significant factors of pavement texture. The most significant variables are International Roughness Index (IRI), and Annual Average Daily Traffic (AADT), followed by several other key factors including freezing index, freeze thaw cycle, and percent passing on #4 and #200 sieves, aggregate geo-classification, binder content, and percent air voids in mixes. This study can help to better understand pavement texture properties and assist in pavement condition assessment.
Keywords:
Pavement texture
Surface characteristics
Multivariate regression analysis
Machine learning algorithms
LTPP database
Journal
IF:
2.5
Papers:
986
Citations:
2.4K
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSave
ShareSaveDetection and Analysis of Asphalt Pavement Texture Depth Based on Digital Image Analytics Technology
PREAI
ShareSaveShareSaveMultiscale power spectrum analysis of 3D surface texture for prediction of asphalt pavement friction
PREAI
ShareSave
View more