Development of roughness thresholds for the preventive maintenance of pavements based on dynamic loading considerations and damage analysis

The objective of this study was to investigate the interaction between surface roughness, dynamic truck loading and pavement damage for the purpose of determining roughness thresholds. These thresholds would be used in the pavement management system as an early warning for preventive maintenance action. This was done by testing the hypothesis that there is a certain level of roughness (roughness threshold-values) at which a sharp increase in dynamic load occurs, thus causing an acceleration in pavement damage accumulation.; The research was successful at validating the above hypothesis by: (1) Identifying empirical relationships between roughness and distress using current indices from in-service pavements; these are the Ride Quality Index (RQI) for roughness and the Distress Index (DI) for distress. (2) Developing similar relationships between surface roughness and theoretical pavement damage using the mechanistic approach.; The above relationships allowed for determining critical ranges of RQI, at which distress and theoretical pavement damage accelerate. Reasonable agreement was obtained between theoretically-derived and empirically-derived ranges. However, these RQI-ranges were too wide to be adopted at the project level. It was therefore concluded that the RQI was not suitable for predicting dynamic truck loads at the project level, i.e., for a specific pavement profile.; Consequently, a new roughness index, called the Dynamic Load Index (DLI), was developed for the purpose of identifying “unfriendly” pavement profiles from a dynamic truck loading aspect. The new index was used to develop tables showing the predicted life extension that would be achieved by smoothing a pavement section with a given remaining service life (RSL) for different DLI levels. These tables can be used to decide when smoothing action needs to be taken in order to get a desired life extension for a particular project. Comparison with RSL-values derived using actual distress growth over time from in-service pavements allowed for determining the optimal range of DLI-values that would lead to the desired life extension upon smoothing the pavement surface. The results showed that such preventive maintenance smoothing action is best suited for rigid pavements.; A reliability-based model for predicting roughness growth with time was also developed. The model was used in conjunction with the newly developed roughness thresholds to determine the optimal timing for smoothing preventive maintenance action.