LAUSR

ABSTRACT All transportation management agencies face the yearly challenge of inspecting the condition of myriad miles of road surface. With routinely acquired and publicly available geospatial data and geospatial modelling techniques, there is potential to substantially reduce the number of survey sites required to characterize overall pavement surface distress condition. Using roadway pavement assets in the State of New Mexico as an example, this study investigated if overall pavement surface conditions could be estimated based on geospatial modelling. A total of 17 explanatory variables, which were extracted from three types of geospatial data, including traffic volumes, environmental conditions and topography, were used to estimate overall pavement surface distress conditions. Results show that overall pavement surface conditions can be effectively (R2 > 0.9) estimated based on the extent of geospatial data and inferential modelling techniques with fewer survey sites, substantially reducing the cost and time for pavement surface distress condition survey. These results also open the way for the future application of geospatial modelling techniques for automated assessment of overall pavement surface distress conditions, which only require users to collect the geospatial data for the surveyed sites and upload the geospatial data to the system, while the computing-intensive process such as model development is fully automated.