Master Thesis Method for dynamic bottleneck prediction in in-house logistics processes (f­­/­­m­­/­­x)

Our brands BMW, MINI, Rolls-Royce and BMW Motorrad have made us one of the world's leading premium manufacturer of cars and motorcycles as well as provider of premium financial and mobility services. We believe in creating an environment where our interns really can learn by doing and where they are given their own areas of responsibility right from the start of their time with us. That's why our experts will treat you as part of the team from day one, encourage you to bring your own ideas to the table - and give you the opportunity to really show what you can do. We, the BMW Group, offer you an interesting and varied master thesis on developing a method for dynamic bottleneck prediction in in-house logistics processes. Our team is responsible for all software applications for in-house logistics in all plants worldwide and consists of business and IT experts. The aim of this master's thesis is to clarify whether and how machine learning models can be used to make concrete, useful statements about the material flows in the company and which bottlenecks in in-house logistics can occur in consequence. Hence, this thesis focuses on developing a method based both on forecasted demand and replenishment data as well as learning from past material flow data used to predict dynamic bottlenecks in the in-house logistics process. Based on the literature review, suitable methodologies for dynamic bottleneck prediction have to be identified. To predict the bottlenecks, one possible methodological approach could be, that the material flows need to be predicted first. Based on planned deliveries, the ingoing and outgoing movements into the main storage areas are known as demands and replenishment. The exact material flows into these main storage areas and outgoing of these main storage areas however need to be predicted based on observed material flows in the past. Once, the material flow relations between the single storage areas are known, the actual load on each storage-to-storage area relation have to be derived and the dynamic bottlenecks (most heavily loaded relation in comparison to predicted available capacity) are to be identified. Based on the resources that have been involved in the past in the respective relations, considering that resources can be used in multiple relations.