Scheduling of reentrant manufacturing lines using a model predictive control (MPC) approach

This paper proposes a formulation for optimizing long-term production and machine allocation for semiconductor fabrication facilities. Specifically, the concept of model predictive control (MPC) is extended to the scheduling of reentrant manufacturing lines commonly seen in semiconductor manufacturing problems. A state space formulation of MPC is used to obtain this “combined” long-term scheduler and controller system. An l₁-norm finite moving horizon cost function is used to obtain a control law which is implemented as a linear programming problem (LP)