In this paper, the validation of a previously published arrayed-waveguide grating (AWG) model is presented, using measures of fabricated devices. The measured spectrum, dispersion, and loss, along with Fourier spectroscopy (FS) measurements of the array errors, are used to verify the numerical simulations and analytical derivations for a Gaussian AWG device. The verified model is then used to investigate the relationship between the different fabrication errors and their impact on the response of flat-top AWG devices. Several conclusions are presented about the accuracy on the use of FS measurements in this modeling, since some discrepancies between the FS measurements and simulations are found and discussed and their motivations identified. Moreover, the ability and flexibility of this model to distinguish and clarify the source of each degradation in the performance of flat-top AWGs is proven.