Transportation infrastructure is a vital component in achieving economic growth and nations' development. Pavement structures constitute major component in the infrastructure. The purpose of the study is to provide a model that can estimate the thickness of the flexible pavement layers based on; the estimated number of 18000 lb single axle load application (W18), resilient modulus of the subgrade (Mr), modulus of elasticity of the three layers (EAC, Ebase, and Esubbase) using Artificial Neural Network (ANN). since that the developed standards by AASHTO 1993 of designing flexible pavement do not provide a direct and a simple way in estimating the thickness of the three layers of flexible pavement (asphalt concrete, base, and subbase layers). Although the American Association of State Highway and Transportation Official (AASHTO) 1993 empirical procedure is an old method and has some limitations, it has been used instead of the Mechanistic Empirical Pavement Design Method Guide (MEPDG). Since the it is simpler than the MEPDG, where the MEPDG requires a lot of data in which is not always available for different transportation agencies in most of the developing countries. The results of the ANN model show a decent prediction of the depths of flexible pavement layers, since the R2 value is 0.99 (close to 1.0) and the MSE value is 0.28 (close to zero), which indicates strong correlation, accuracy, and low inconsistency between the observed and predicted thickness of the flexible pavement layers.