Decision Model of Pavement Maintenance Based on International Roughness Index.

When choosing the pavement maintenance scheme, the actual pavement condition cannot be truthfully reflected by traditional fixed indexes and weights, making accurately evaluating the pavement condition particularly important. This study aims to analyze the relationship between international roughness index (IRI) and pavement condition index (PCI), and propose the IRI-based pavement condition rating to improve the correctness of the maintenance decision. Based on the inspection data and maintenance data of large and medium-sized repair and maintenance projects in Shaanxi Province, representative road sections in northern Shaanxi, The Central Shaanxi Plain, and southern Shaanxi were selected to establish a simple model to examine the relationship between PCI and IRI. The model development data were selected from 312 different road sections, including 1665 data points. The model validation data were selected from 140 road sections, including 333 data points. The road sections selected for validation data and those for development data had the same data range. Results demonstrate that, the S-function could best express the relationship between PCI and IRI. The function established by data fitting had a high coefficient of determination (R2=0.959), and the deviation of the predicted IRI value was low. The model validation of different datasets all yielded relatively more accurate predictions (R2=0.973). Finally, the IRI-based pavement condition rating was proposed, and the pavement condition was divided into five grades, namely, excellent, good, fair, poor, and very poor, by using IRI detection data. This proposed rating provides double verification of the pavement condition with PCI rating and improves the correctness of the maintenance decision. [ABSTRACT FROM AUTHOR]