Robust model predictive control for inventory system with uncertain demand using linear matrix inequalities

Abstract

In this paper, we develop an optimal control strategy on inventory systems with uncertain demand. To deal with these uncertainties we use a synthesis of robust model predictive control with linear matrix inequalities. The goal is to minimize the difference between the prediction and the reference trajectory subject to the objective function of each period, based on the input and output constraints. Using standard techniques, the optimization problem that minimizes the difference between the prediction and the reference trajectory, is reduced to a convex optimization problem involving linear matrix inequalities (LMIs). We provide numerical simulations on this system using MATLAB and then observe how robust predictive control models produce optimized strategy at the inventory level. In the simulation results, robust predictive control models provide an optimal strategy for controlling inventory levels with minimum total cost and inventory levels following inventory levels on issues.