An effective approach for optimization of a perishable inventory system with uncertainty in both demand and supply

Effective inventory system management allows companies to respond quickly to the market change and develop a flexible production system to improve their competitiveness. In this paper, we study an inventory problem of a perishable product with random deterioration rate under uncertainty in both demand and supply. Such problem appears in biomass supply chains because feedstock supply is always uncertain due to weather conditions, insect population. By formulating the problem as a multiperiod stochastic inventory model, we demonstrate that its optimal inventory policy is an order-up-to level policy. We develop an effective algorithm combining scenario-based optimization and Lagrangian relaxation for quickly and approximately calculating the order-up-to levels. A computational study shows that the algorithm can find a near-optimal solution with a relative gap of cost less than 1% on average in comparison with the optimal inventory policy.