Pavement performance models are key components of any pavement management system (PMS). These models are used in a network-level PMS to predict future performance of a pavement section and identify the maintenance and rehabilitation needs. They are also used to estimate the network conditions after the application of various maintenance and rehabilitation alternatives and to determine the relative cost effectiveness of each maintenance and rehabilitation alternative. Change in pavement surface roughness over time is one of the most important performance indicators in this regard. A model for changes in the international roughness index (IRI) over time was developed through artificial neural networks (ANNs) pattern recognition, using information from the Specific Pavement Study (SPS)-5 asphalt concrete rehabilitation experiment extracted from FHWA's Long-Term Pavement Performance database. This model can be used to predict and compare pavement roughness variation trends after various rehabilitation alternatives. An example illustrates the implementation of the roughness model along with life-cycle cost analysis in making future pavement rehabilitation recommendations. Model testing results indicate prediction of IRI with minimal errors, and predicted future roughness trends match perfectly with the past performance. These findings indicate that the ANN model performs successfully in predicting IRI trends following each kind of treatment in the SPS-5 experiment. The ANN model was developed for the SPS-5 flexible pavement rehabilitation sections in a wet-freeze climate and may be applied for similar conditions. The example also shows that the detailed model development and implementation framework provided in this study can assist in network-level PMS decision making.