This study addresses the optimization of automated yarn handling in textile manufacturing by examining the related suction process through a combined numerical and experimental approach. In particular, a three-dimensional model of the suction nozzle was coupled with an equivalent linear-elastic beam representation of the yarn, and a Fluent–IDW–Abaqus weakly coupled fluid–structure interaction (FSI) framework was employed to capture the yarn’s release and dynamic response under negative-pressure suction. High-speed imaging experiments validated the simulations, demonstrating excellent agreement in displacements and velocities. According to the results, increasing the initial suction pressure from −0.04 MPa to −0.06 MPa reduces adsorption time by approximately 62% and markedly dampens yarn-end vibrations, enhancing suction performance. Pressures beyond −0.06 MPa, however, induce overshoot and nozzle collisions, increasing the risk of entanglement and mechanical damage. The outcomes of a statistical analysis are also presented to further quantify the interplay among energy consumption, suction efficiency, and operational success under varying pressures, thereby providing a rigorous foundation for the optimal selection of pressure parameters in automated yarn-handling systems.