GAI-supported skill development in design engineer scheduling optimization

Skilled labor shortages and increasing task complexity have made workforce development a critical challenge in engineering and manufacturing environments. In engineer-to-order (ETO) settings, design engineer teams must maintain productivity while developing multi-skilled capabilities, yet traditional mentoring is often limited by scarce expert availability and inconsistent instructional quality. This study proposes a scheduling optimization framework that integrates generative AI (GAI)-supported skill development into design engineer assignment. The model jointly minimizes project tardiness and promotes skill growth by incorporating GAI tutoring during task execution and GAI-generated feedback after task completion into a logistic skill-acquisition function. To solve the resulting combinatorial problem, we develop a tabu search metaheuristic that dynamically adjusts task assignment, tutoring intensity, and feedback allocation. Numerical experiments based on a six-month simulated ETO workload show that the proposed approach improves long-term team capability formation while maintaining or improving productivity. Sensitivity analyses further indicate that novice-heavy teams benefit most from GAI support. These findings demonstrate the potential of GAI as a complementary instructional resource for enhancing both skill development and scheduling performance in manufacturing environments.