Search

The existing international standards suggest a methodology to assign a specific functional class to a road, by the values of some features, both geometrical and use-related. Sometimes, these characteristics are in contrast with each other and direct the analyst towards conflicting classes for a road or, worse, one or more of these features vary heterogeneously along the road. In these conditions, the analyst assigns the class that, by his capability and experience, he retains the most appropriate, in a very subjective way. On the contrary, the definition of an automatic procedure assuring an objective identification of the most appropriate functional class for each road would be desirable. Such a solution would be useful, especially when the road belongs to the existing infrastructure network or when it was not realised by out of date standards. The proposed procedure regards the definition of a classification model based on Pattern Recognition techniques, considering 13 input variables that, depending on their assumed value, direct the analyst towards one of the four functional classes defined by the Italian standards. In this way, it is possible to classify a road even when its characteristics are heterogeneous and conflicting. Moreover, the authors analysed the model limitations, in terms of errors and dataset size, considering observation and variable numbers. This approach, representing a beneficial decision support tool for the decision-maker, is exploitable for both planned and existing roads and becomes particularly advantageous for road agencies aiming to optimally allocate their limited funds for specific interventions assuring the achievement of a fixed functional class.

Artificial Neural Networks represent useful tools for several engineering issues. Although they were adopted in several pavement-engineering problems for performance evaluation, their application on pavement structural performance evaluation appears to be remarkable. It is conceivable that defining a proper Artificial Neural Network for estimating structural performance in asphalt pavements from measurements performed through quick and economic surveys produces significant savings for road agencies and improves maintenance planning. However, the architecture of such an Artificial Neural Network must be optimised, to improve the final accuracy and provide a reliable technique for enriching decision-making tools. In this paper, the influence on the final quality of different features conditioning the network architecture has been examined, for maximising the resulting quality and, consequently, the final benefits of the methodology. In particular, input factor quality (structural, traffic, climatic), “homogeneity” of training data records and the actual net topology have been investigated. Finally, these results further prove the approach efficiency, for improving Pavement Management Systems and reducing deflection survey frequency, with remarkable savings for road agencies.