Predicting pavement evaluation from the International Roughness Index

In 2008, the Indiana LTAP Board approved funding for an update to a needs assessment project completed in 2001. The objective was to estimate the current funding deficiencies for county highway and city/town street departments. IRI and PASER data was collected by an outside vendor to estimate the condition of roadways throughout Indiana. This data predictably illustrated the deteriorating conditions present on local roads statewide. An efficient way to improve road conditions and maintain them at an acceptable level is through the implementation of a pavement management plan. Many different variations exist; however, this study focused on a plan utilizing road conditions to prioritize road segments. Pavement condition ratings such as the PASER are the most logical choice for assigning road conditions. The PASER accounts for a wide range of pavement distresses, including cracking, rutting, raveling, and potholes, and provides maintenance recommendations for each rating. The downside of using the PASER system is the laborious and time-consuming process of manually assigning ratings; whereas, the IRI is calculated in real-time while driving over a road segment. The IRI and PASER data collected for the updated needs assessment report provided an opportunity to closely examine the relationship between these two measurements. A statistical model that predicts the PASER from the IRI would provide a quicker and more efficient way of assigning road conditions for use in a pavement management plan. An ordered probit model was chosen as the best statistical model for the given dataset. Additional variables including section length, section width, and various county specific data were examined along with the IRI as predictors of PASER. A model was derived that can provide network-level predictions and planning applications such as budget estimates. It also provides some confidence in predicting the poorest road segments according to the PASER scale (1 through 4). These segments can be directly observed in the field to assign preventative maintenance measures and implement a pavement management plan