Stochastic MPC for supply chain management using MCMC approaches

In order to efficiently manage a single echelon supply chain under stochastic disturbance and probabilistic constraints, we propose a stochastic model predictive control (SMPC) framework and implement the Markov Chain Monte Carlo (MCMC) algorithm to solve stochastic programming problems in a receding horizon scheme. The generalized autoregressive conditional heteroskedasticity (GARCH) model is adopted for demand forecasting. Simulation results show that the new approaches outperform the standard MPC in mitigating the production fluctuating and reducing the integral absolute error (IAE) of the inventory.