A paper entitled “hybrid multi-task learning-based response surface modeling in manufacturing” is recently published in the Journal of Manufacturing Systems. The paper was co-authored by Yuhang Yang and Chenhui Shao.
This paper developed a hybrid multi-task learning-based method tocost-effectively model the response surfaces of multiple similar-but-not-identical manufacturing processes. The method was evaluated using a simulation-based numerical case study and a real-world case study with experimental data collected from ultrasonic metal welding. See the full paper here.