Two-stage robust optimization for perishable inventory management with order modification

In this paper, we study an inventory management problem with perishability and demand uncertainty where the goal is to minimize the sum of ordering, purchasing, holding, shortage, wastage, and modification costs. Our motivation for studying this problem is to more specifically study the efficiency of order modification in an uncertain environment. In this problem, we suppose that the demand belongs to an uncertainty set without specific probability distributions. We formulate the problem as a novel two-stage robust integer optimization model and develop an exact column-and-row generation algorithm to solve it. The importance of using robust optimization for the problem is to immunize the decision maker in the worst-case scenario where vital perishable products such as blood must always be available. Our extensive computational experiments demonstrated the significant efficiency of our robust model compared with a deterministic model and a stochastic model in both risk-neutral and worst-case settings. We concluded that considering order modification results in more reliable decisions in inventory systems.