Skill-Constrained Model Predictive Control for Resilient Manufacturing Supply Chains

The article presents a closed-loop skill-constrained model predictive controller designed for production-inventory systems, which optimizes production, inventory, backlog, and training decisions based on the availability of certified workers. This controller solves a finite-horizon mixed-integer program, addressing various operational scenarios through synthetic SkillChain-Gym environments. The findings indicate that while predictive control can be beneficial in scenarios with predictable skill bottlenecks, static insurance strategies often outperform under unexpected shocks, highlighting the importance of forecastability over adaptability in managing labor capacity.