Skill-Constrained Model Predictive Control for Resilient Manufacturing Supply Chains
A new research paper introduces a skill-constrained model predictive controller designed to optimize manufacturing supply chains by balancing production, inventory, backlog, and worker training decisions. The controller addresses the challenge where training decisions made today impact the qualified human capacity available tomorrow, as training and production both require scarce worker hours. Evaluations on synthetic scenarios indicate that the effectiveness of this predictive control strategy is highly dependent on the specific production regime, proving beneficial when skill or labor bottlenecks are forecastable but not always superior to simpler methods under surprise shocks or when existing slack makes static insurance plans more cost-effective. AI