Inversion of Asphalt Pavement Properties From Slopes Reported by the Traffic Speed Deflectometer

Abstract

The traffic speed deflectometer (TSD) is a semi-trailer truck equipped with Doppler lasers. The device attempts to measure ride-surface slopes at several distinct locations caused by its own loading while moving at highway speeds. This study investigated the potential of the TSD to deliver a project-level evaluation of asphalt pavement infrastructure. A twofold objective was defined: (i) outlining a procedure for inverting TSD-reported slopes and obtaining the mechanical properties of individual pavement layers, and (ii) demonstrating the procedure by application to a real, field-measured dataset. In general, the inversion procedure was based on matching model-calculated slopes to TSD-reported slopes while accounting for the movement of the entire TSD truck and without assuming that the slope is zero at a certain offset distance from the rear axle. The procedure utilized a layered viscoelastic half-space model capable of dealing with moving loads. Accordingly, the sought properties for inversion were the viscoelastic creep compliance master curve of the asphalt layer and the elastic moduli of all other layers. Application of the procedure to a field dataset revealed that the coefficient of variation of the inverted moduli had an increasing trend with depth (in absolute terms), indicating that the property inference of deeper layers from TSD-reported slopes has lower reliability compared with the property inference of shallow pavement layers.