Distress prediction models play an important role in a pavement management system (PMS). These models are used to predict the condition of pavements treated with given maintenance and rehabilitation (M&R) action. They can also be used to compare the economics of various maintenance alternatives. The development of distress prediction models for a network-level PMS recently developed for the Ohio Department of Transportation is described. Five M&R actions or maintenance alternatives were included. Visual condition surveys of overlaid pavements (composite) currently include 14 distresses. These distresses were grouped into four distress groups each having its own severity and extent. Thus, 8 equations were developed for each M&R action, resulting in 40 equations for all five M&R actions and four distress groups. The models were used to predict distresses and pavement condition rating (PCR), which were compared with the corresponding distresses and PCR calculated from field observations. These comparisons indicated that the models were capable of predicting with reasonable accuracy the condition of a highway network as well as an individual pavement segment. Limited data for 5 years were available at the time of analysis; this should be kept in mind while the models are extrapolated for traffic loadings beyond these limits. (A)
Abstract