Three-dimensional (3D) printing has revolutionized the way we design and manufacture complex products, offering numerous benefits such as reduced material waste, increased design freedom, and a shorter path from planning to production. Extrusion based 3D printing with short fibre-reinforced high-performance polymers (SCFRP) enables the direct manufacturing of complex and high strength parts. In extrusion-based processes a polymer provided as granulate or filament is melted and deposited through a nozzle and parts are build layer per layer. Through the direct manufacturing without molds or curing processes, the process parameters during printing determine the final properties of the part. To ensure the reliability of printed parts in structural applications, these properties must be stable and reproducible. Simulations of the printing process can help to evaluate the process parameters and resulting properties. One important property is the bonding strength between adjacent layers. The temperature-driven bonding process between adjacent layers primarily defines the mechanical properties perpendicular to the deposition direction. This bonding process is not only influenced by temperature but also by material crystallization, which often occurs in high-performance materials and inhibits the bonding process by preventing polymer chains from moving freely while they form crystal structures. The objective of this study is to develop a bonding model to predict the bonding strength of SCFRP parts manufactured by a granulate based extrusion process. Therefore, the temperature distribution is calculated by a process simulation. A model based on the reptation theory was extended to account for the complex material behaviour which was observed through the bonding characterization with tensile tests. The model was modified, and a new fitting of the bonding characterization data was performed. The resulting model was implemented in an Abaqus user subroutine that calculates the bonding strength between adjacent layers in a process simulation. The results showed that the new model formulation captures the observed material behaviour very well.