Marlin Ulmer is Professor of Management Science at Otto von Guericke University Magdeburg. His research focuses on data-driven, quantitative decision support for problems in logistics and supply chain management. Since, 2019, he teaches a ProDok course on Stochastic and Dynamic Decision Problems.

This type of problem requires plans to be adapted regularly in light of new information. Therefore, the plans must be designed flexibly and with foresight. As the problems vary greatly, there is no universal solution procedure. Strategies must consider the specifics of each problem, model and set of data. This requires a combination of structured analysis and creative thinking.

‘AI’, of course. The challenge lies in extracting the relevant aspects from the forest of buzzwords and exaggerated promises, and using them sensibly. However, this is also an opportunity to establish an additional pillar alongside 'classic', often theory-driven research. There is still plenty of room here, particularly with regard to how theory and machine learning can be meaningfully combined.

I am inspired by the community, the media, and my own everyday experiences. For instance, I have researched the reliability of delivery time windows, rework at home repair services, and the operations of home healthcare services. With everyday life topics like these, the introduction practically writes itself.

The first page of the first draft of a paper I wrote with Martin Savelsbergh in 2019 hangs in my office. The page is completely covered in red ink due to his corrections. I often show it to my doctoral students to put them at ease before giving them my feedback.

Back to overview page