Modeling of Pavement Deterioration in Cold Regions

The Alaska Department of Transportation and Public Facilities (AKDOT&PF) tracks roughness data for maintained roads. This is done by using a laser-mounted beam on an SUV-style vehicle. The equipment is operated and maintained by an outside consulting firm and has occurred annually since 2007. The values provided are International Roughness Index (IRI) rutting and Present Serviceability Rating (PSR) for every 20 feet of length per one lane in one direction. The data is stored in spreadsheet form along with Global Information System (GIS) information. The purpose of this study is to explore a simplified approach method for predicting IRI and rutting that could be used in decision making for AKDOT&PF's pavement management system (PMS). Four years of IRI and rutting data were obtained from AKDOT&PF for a major arterial. Subsection and section lengths were selected and averages performed into a final one-mile average. Each year's section average was plotted and a multi-linear regression performed. The equation describing the trend line is then used to predict deterioration of the IRI or rutting value. A PSR was calculated and compared to what had been last measured. Finally, a comparative analysis through visual inspection was performed and compared to the data. The outcome was a multi-linear regression that can be performed and used to predict future pavement deterioration in terms of IRI and rutting. A calibration of the model was implemented using visual inspection. No significant differences were found between the derived model and visual inspection.