Application of Robust Linear Quadratic Control for Inventory System with Unknown Demand: Single Product Case

Abstract

In this paper, a dynamical model of single product inventory system with unknown demand in a linear state space equation with unknown parameter for inventory control purposes was formulated. An existing control method, robust linear quadratic regulator (RLQR), was applied to control the inventory level by generating the optimal purchasing product volume so that the product stock follows a reference trajectory with minimal cost. The result of the performed numerical experiments showed that the optimal purchasing product volume was determined for every time period and the product stock was closed to the given trajectory level desired by the decision maker.