Abstract: The present invention provides a data-driven additive manufacturing (AM)—compression molding (CM) process in which a fiber-reinforced thermoplastic is deposited into a mold cavity and subsequently consolidated under heat and pressure. Deposition and molding parameters are selected using a numerical model to control microstructural characteristics and final part performance. Measured process data is fed to a control system to update the numerical model in real time. The control system is further configured to modify at least one AM or CM parameter based on a deviation between the measured process data and a model-predicted microstructural state. AM process parameters can include material flow rate, print speed, nozzle geometry, nozzle-to-substrate gap height, or material extrusion temperature. CM process parameters can include compression pressure, mold temperature, and compression timing after deposition.