MODELING INDIRECT STATISTICS OF SURFACE ROUGHNESS

Quarter-car model-based indirect statistics of road surface roughness are analyzed in this paper. The stochastic process theory is used to establish the relationship between the international roughness index (IRI) statistic of a quarter-car response and the power spectral density (PSD) of road random excitation. Under the circumstance of a linear vehicle model and homogeneous roughness with a zero-mean Gaussian distribution, we prove that the average of the absolute response of the quarter-car model is directly proportional to the standard deviation of that response quantity. This result enables us to further correlate the indirect statistics with PSD roughness. A theoretical relation between the IRI and the PSD roughness is established. Based on the road classification proposed by the International Standards Organization (ISO), it is found that a linear correlation exists between the IRI and the standard deviation of roughness. Furthermore, corresponding to the ISO PSD-based road classification, an IRI-based road classification is provided to categorize pavement into five classes, ranging from very good surface condition to very poor surface condition.