Improving repetitive manufacturing systems: Model and insights

We consider a manufacturing system that is controlled by a fixed-cycle smoothed production policy. This policy, which is becoming increasingly common in repetitive manufacturing environments, is characterized by a production rate that is updated on a periodic basis. We model the system as a stochastic process that includes parameters for vendor responsiveness, plant responsiveness, randomness in production yield rates, nonstationarity and randomness in market demand, demand forecast error, operating cost rates, and safety stock. Properties of the model expose structural relationships between expected system performance and system parameters.