Bayesian model for tyre/asphalt pavement noise

In this paper, empirical acoustic models were developed for four asphalt surface mixes, including conventional dense graded asphalt contcrete, open-graded asphalt concrete, rubberised open-graded asphalt concrete, and rubberised gap-graded asphalt concrete. Tyre/pavement noise data were collected from in service flexible pavements for four consecutive years. These data were panel structured, and with a portion of observations missed arbitrarily. A Monte Carlo Markov chain sampling and a multiple imputation algorithm were used to capture the unobserved heterogeneity and deal with missing observations by Bayesian simulations in modeling process. Two types of models were obtained: one was an integrated model; the other was individual model characterising each mix. The estimated models suggest that tyre/pavement noise increases with pavement age and mean profile depth, and decreases with air-void content of surface mixes. The noise levels of the four asphalt mixes rank as: rubberised open-graded asphalt concrete; open-graded asphalt concrete; rubberised gap-graded asphalt concrete; and dense graded asphalt concrete.