No comments yet

Predicting the degradation of surface pavement due to raveling



Last Updated: 7-2017


In The Netherlands new contract forms have emerged. Rijkswaterstaat (RWS) is the executive agency of ‘Ministry of Infrastructure and Environment’ and is in charge of the infrastructural national network in the Netherlands. RWS wanted to shift the risks of projects from RWS to the contractors because contractors have the knowledge to construct and manage projects. Due to this new development, contractors need to shift their focus from merely construction of the motorways to construction and maintenance of the motorways. This means that contractors needed to understand how the pavement construction of the motorways behaves during its life-cycle. Therefore, contractors want to know what the life-expectancy is of pavement constructions, and in particular the surface pavement. Raveling is one of the main damage types that causes degradation on the surface pavement in The Netherlands on motorways. Currently, there are no practical models or tools available for understanding the risks of maintaining a surface pavement. However, Heijmans has a method which empirically assesses this risk. This research will use this method called ‘Heijmans life-cycle management’. This research focuses on producing an empirical database with a software tool called ‘the Feature Manipulation Engine (FME). This tool is used to merge different datasets into one database. With this database, two filtering techniques are produced which can assess the degradation risk. The first technique involves plotting the degradation of the surface pavement over a lifetime and the second technique depicts the fatigue failure ratio of the surface pavement over a lifetime. With plotted data of fatigue failure ratio over time, a probability beta distribution curve is constructed to produce a probability ‘Monte Carlo; simulation model to predict fatigue failure ratio of surface pavement. The results show that the first plotting technique shows too much scattering over time in order to produce a predictive model. The second, however, is in able to filter out the scattering and can be used for the predictive model. The results also show the effects of cold weather is neglectable with respect to the impact of traffic loading. Finally, our predictive model shows that high traffic intensity results in a rapid increase in failure at 4.8 years, in moderate traffic intensity this moment occurs at 5.6 years and for low traffic intensity this moment occurs at 7.1 years.

Post a comment