Optimization of a perishable inventory system with both stochastic demand and supply: comparison of two scenario approaches

In this paper, we study a multi-period inventory model for a perishable product with both stochastic supply and demand in a rolling horizon framework. The product has a fixed shelf life such as fresh products, blood cells, chemicals, drugs and other pharmaceutical products. The objective is to minimize the expected total cost composed of ordering, purchasing, holding, shortage and waste costs. We focus on finding a high-quality solution close to the optimal solution of the model that provides decision support for decision-makers. We propose a stochastic programming model and transform it into MILP model based on conditional scenarios (CS) approach to reduce the computational burden. By comparing with the sample average approximation (SAA) method in a numerical study, we show that our method works efficiently.