Intelligent pavement rutting prediction models: the case of Norwegian main road network

Abstract

Prediction of pavement performance is a key process in the efficient management of pavement assets for a highway agency. There are a lot of tools that can be used to develop pavement performance prediction models, but the newest generation of tools belongs to the field of Artificial Intelligence. Rutting prediction models for stone mastic asphalt pavements are developed using multiple linear regression (MLR) and Artificial Neural Network (ANN) techniques, using data from the Norwegian national road databank (NVDB). Comparative study of the results is also conducted. The main conclusion from this study is that pavement rutting prediction models using the intelligent ANN technique predict pavement condition with a better accuracy than the classical MLR models.