The contributions of pavement structure and features, rehabilitation techniques, climatic conditions, traffic levels, layer materials and properties, pavement distress, and other factors causing changes in pavement smoothness are not well documented. As a result, it becomes difficult to select the appropriate pavement structure, design features and rehabilitation strategies to ensure pavement smoothness. This study focuses on analysing the available LTPP data for asphalt pavements in California by investigating the correlation between the pavement roughness and the effect of pavement temperature, precipitation, fatigue, age of pavement, rutting, and the average annual daily truck traffic. IRI has been identified as the factor characterizing pavement smoothness. Results indicated that when diving pavement sections between three different groups according to the annual precipitation for pavement section in the State of California, the IRIchange can be predicted with 93.5% accuracy for sections with less than 200mm of annual precipitation, 85.9% accuracy for sections with annual precipitation between 200mm and 90mm, and 90.1% for sections with annual precipitation higher than 900mm.