On April 27th, 2023, we had TUM professor Stefan Minner, who is editor-in-chief of the International Journal of Production Economics, vice-chairman of the scientific advisory board of the German Logistics Association and member of the research committee of the European Logistics Association, as a special guest speaker to talk about Machine Learning for Data-driven Inventory Management.

He talked about Machine Learning as a methodology which can support several inventory management problems ranging from procurement decisions under uncertain prices, safety stock allocation and sizing, and parts classification for inventory policy selection and parameterization. While mainstream inventory management approaches treat lead time demand prediction and inventory parameter optimization sequentially, recent advances in data-driven optimization and machine learning enable simultaneous optimization with a considerable cost savings potential. Professor Minner uses advanced causal demand prediction and inventory prescription approaches to overcome the challenge of required distributional assumptions in learning inventory control parameters and for scalable solution approaches for real-world complex supply networks.