Myopic Inventory Control with Returns in Case of Uncertainty: Adaptive Algorithms

A. Kozhan, Viktor Laptin, Alexander Mandel
A model of inventory control with returns is considered, when it is possible for consumers to return the products they have purchased in case of uncertainty when the probability distribution of a random variable of one-step demand is a priori unknown. It is shown that in this case the optimal inventory management strategy is 4-parametric and adaptive algorithms are constructed that converge with probability 1 to the true values of these parameters.