This paper describes the application of adaptive neuro-fuzzy inference system (ANFIS) methodology for the backcalculation of airport flexible pavement layer moduli. The proposed ANFIS-based backcalculation approach employs a hybrid learning procedure to construct a non-linear input-output mapping based on qualitative aspects of human knowledge and pavement engineering experience incorporated in the form of fuzzy if-then rules as well as synthetically generated Finite Element (FE) based pavement modeling solutions in the form of input-output data pairs. The developed neuro-fuzzy backcalculation methodology was evaluated using hypothetical data as well as extensive non-destructive field deflection data acquired from a state-of-the-art full-scale airport pavement test facility. It was shown that the ANFIS based backcalculation approach inherits the fundamental capability of a fuzzy model to especially deal with nonrandom uncertainties, vagueness, and imprecision associated with non-linear inverse analysis of transient pavement surface deflection measurements.