Assessment of Construction Smoothness Specification Pay Factor Limits Using Artificial Neural Network Modeling

Abstract

Many state highway agencies are using quality control/quality assurance construction smoothness specifications that provide for contractors’ pay to be proportional to the riding comfort of the constructed new pavement. However, in many cases, the pay factor limits are based on subjective engineering judgment rather than rational analysis of the declining riding comfort of newly constructed pavements. This study used the artificial neural network methodology to develop time-dependent roughness prediction models for three types of pavements: Portland cement concrete pavement, asphalt overlay over concrete pavement, and asphalt pavement. The relationship of various pay factor limits and future roughness progression and the pavement service life were assessed. Rational pay factor limits were then developed for the zero blanking band profile index of the California Profilograph measurements employed in the construction smoothness specifications based on the predicted future riding comfort of the newly constructed pavements.