Relating Field Energy Attenuation in Portland Cement Concrete Pavements to Fracture Mechanics

Abstract

Asset management of infrastructure is fundamental for maintenance planning and preservation of common property. A robust testing program is needed to assess the present-day status and for proper actions in time to minimize the ongoing depreciation of value. As a matter of fact, Portland Cement Concrete pavements show very little deterioration even after many years in service. Thus, it may be difficult to accurately predict the present asset value, other than using linear relations to the presumed design life. The primary reason for failure is cracking in concrete pavements, so assessing the dissipated energy from the load-deformation relation from a given load could be utilized for the purpose. The dissipated energy, i.e. the work data can be assessed by a falling weight deflectometer test, mimicking the passing of a truck or aircraft wheel load. In the present study, dynamic field data are evaluated, and the input data needed for the fracture mechanics model are used to predict the pavement life regarding cracking. To predict fracture energy and assess rolling resistance as well in concrete pavements, we need to consider the energy balance of the pavement system. To assess dissipated energy, falling weight deflectometer time histories are used to evaluate the pavement contribution to rolling resistance. Such analyses include all layers in the structure including the subgrade, so in the present case a way of sorting the dissipation at various depths is investigated. Field data were collected from a site, at mid-life of the predicted design life. The failure was confirmed several years later, and the remaining life was compared with the assumption that the dissipated energy near the edge was enough to initiate the cracks within the actual time to failure. Conversely, the dissipation at the mid-slab position was below the limit. The data from the field test were also used as an input for a finite element model to see if it was viable to further improve the prediction. The method seems to be promising, but more data are needed as the present set only represents the mid-life status.