This book addresses the application of supply chain analytics to improve inventory management, a cornerstone for successful operations at many companies. Holding inventory reduces stockout cost, facilitates smooth operations, and improves service levels and customer experience; but it also ties up capital and goes along with costs for storage, obsolescence, handling, and other. Due to the complexity of the task, companies apply inventory models, which build on assumptions that seldomly fully hold in practice. As a consequence, the actual performance of the inventory system deviates from the projected performance and the full potential of the models cannot be exploited. This book covers three different problems that companies commonly face when managing their inventories: the introduction of new inventory policies in existing inventory systems, the use of algorithmic advice by human planners, and the accuracy of master data on which inventory models rely. By using mathematical optimization, behavioral experiments, and machine learning, the developed approaches support the successful implementation of state-of-the-art inventory research in practice.