A novel model predictive control algorithm for supply chain management in semiconductor manufacturing

Supply chains in semiconductor manufacturing are characterized by integrating dynamics, nonlinearity and high levels of stochasticity. In this paper, we present a novel model predictive control (MPC) algorithm for supply chain management (SCM) in semiconductor manufacturing. A Type II filter is designed to attenuate the integrating noise such as that exhibited by unforecasted customer demand. The selection of the filter gain provides the flexibility to achieve better performance and robustness. The forecast of customer demand plays a critical role in the algorithm. The advantages of this novel MPC algorithm are demonstrated through case studies of a representative supply chain problem in semiconductor manufacturing which involve scenarios of customer demand forecast error and anticipated periodic demand.