Predictive Modelling of Road Deterioration Networks

A pavement during its life cycle can be subjected to various detrimental phenomenon such as cracking, mainly due to the repeated loading of the traffic which is using the pavement for transportation. Pavements are also subjected to temperature and moisture variations. Consequently, the collective action of these factors will result in the pavements to deteriorate with time. This research focuses on the development of a Bayesian Belief Network (BBN) model, an artificial intelligence approach, to extend the service life of pavements with minimal costs based on the analysis of data collected from major roads in United Arab Emirates (UAE) and China. Applying the model in two countries with different road and environmental conditions will serve in evaluating the model’s validity. The BBN model will not only be used as a decision support system for maintenance efforts prioritization at the operation stage, but also as a reference for consultants and contractors to design pavement at the design stage with an upfront foresight at the life-cycle implication of their design.