Recently, rigid pavements have become an attractive alternative for use in highways and urban roads subject to large traffic volumes. Rigid pavements are attractive due to their enhanced durability, strength and cost/benefit ratio. Despite of these characteristics, there is a need for maintenance planning and performance prediction. In this work, a rigid pavement management system was developed to address these needs. To achieve this, computational modeling tools based on the finite element method (FEM) and on artificial neural network (ANN) were used. A FEM model was created to characterize the pavement s structural behavior under the action of traffic and temperature variations. The BR101/NE PB-PE highway was used as a case study. A finite element model of this section of the road was created using the finite element software ABAQUS. Results from a Falling Weight Deflectometer (FWD) field test performed by Brazilian National Army were used for model characterization, i.e., to obtain the elastic moduli of the layers. The modeling technique was validated using experimental data reported in the literature. In the validated model for a homogeneous section, traffic loads were applied at different locations and temperature distributions were applied through the slab thickness. The latter is important because, as reported by other authors, this variation can cause warping of the concrete slabs which affects the load transfer between adjacent slabs. The validated FEM model was then used to generate a database of the stresses and load transfer coefficients across the transversal joints. The information collected in the database was used to train an ANN which processes the fatigue damage accumulation. In this work, the durability of the pavement was assessed using classical criteria established in fatigue models based on damage accumulation. The results confirm the ANN prediction capability of damage occurrence. The developed system based on prediction models can contribute to improve maintenance planning to ensure better pavement performance for its users. In addition, such a system has the potential of extending the pavement service life and minimizing costs associated with repairs.