Machine Learning for Pavement Performance Modelling in Warm Climate Regions

摘要

Accurate pavement performance modelling is an essential requirement for cost-effective pavement design and enhancespavement management decision making. Due to the complexity of the pavement structure and pavement response, dominantfailure mechanisms may vary depending on the climate region: cold or warm. This study investigated the significance ofpavement design factors on pavement performance in warm regions and compared them to set of factors previously identifiedfor cold regions. An artificial neural network (ANN) supported by a forward sequential feature selection algorithm wasemployed to identify the most significant design factors prevailing in warm climate regions using data extracted from theLong-Term Pavement Performance database. In addition, fivemachine learning techniques were utilized to model the pavementperformance in warm regions, namely: regression tree, support vector machine, ensembles, Gaussian process regression, andANN. Moreover, conventional regression modelling was used for comparison assessment. The analysis revealed seven designfactors that are significantly impacting asphalt pavement performance in warm regions: initial roughness, relative humidity,average wind velocity, average albedo, average emissivity, traffic volume, and pavement structural capacity. The resultsindicate that pavement performance in warm climate regions is dominated by different environmental factors than thosefound for cold climate regions. The ANN modelling technique produced the most accurate asphalt pavement performancemodels.