In this review, we discuss the data-driven systems and their effects on the implementation of the inventory theory. After overviewing the theory briefly, we group the data-driven approaches to simplify exposition. We consider the use of available data to estimate the parameters of more complex models, and propose developing the theory in that direction, as well. As a pedagogical example, an extension of the standard EOQ model with heterogenous customers is presented. The review proposes a research agenda for inventory problems and concludes with discussing challenges for the future.