Classification of measures from deflection tests by means of fuzzy clustering techniques

• We applied a clustering technique to classify real-time HWD data in two clusters.
• The classification was evaluated by membership degrees.
• The number of clusters, however, can be expanded to classify new observations.
• The procedure permits to know the characteristic of the pavement in real time.
• It is possible to exclude one or more geophones when there are problems.

As everybody knows, non-destructive tests carried out with the Heavy (or Falling) Weight Deflectometer equipment permit to identify the mechanical properties of the layers constituting a road or an airport pavement.The ordinary activity generally causes at least two issues: (a) impossibility to anticipate the stiffness of the pavement analyzed during the trial; (b) probable mistakes induced by punctual degradations. In the latter case it would be more appropriate to discard the reading of one or more geophones for a correct determination of the modules.In order to overcome the above limitations, we propose a procedure based on a fuzzy clustering technique that enables the classification of the deflections in real time, reducing the number of drops (generally equal to 3), with no need for traditional back-analysis. Any uncertainty of the result achieved is quantified by the fuzzy membership degree for which the analyst has an objective measure of the representativeness of the data detected.