Intelligent planning and process design
In the "Intelligent planning and process design" research focus area, we develop learning methods for solving complex planning problems and for evaluating and parameterising industrial processes and process chains. The aim of our research is to develop adaptive, resource-efficient and robust systems that can survive in dynamic industrial environments.
Adaptive planning of production and logistics processes
A central focus of our work is the use of deep reinforcement learning to solve both abstract and practical planning problems. Our approaches enable the development of assistance systems for the efficient planning and control of production and logistics processes, taking into account real factors such as adherence to delivery dates and transport costs. High process complexity, dynamic changes in the process environment and the simultaneous optimisation of several target criteria are challenges that we face in our research.
Sustainable processes using deep learning
In addition to planning, we focus on data-driven deep learning approaches in order to optimise industrial processes - from production to supply chains - and conserve resources. We are researching automated systems that continuously adapt to changing conditions. In addition, we develop algorithms to optimise processes with regard to sustainability requirements so that they are not only more efficient and cost-optimised, but also sustainable and environmentally friendly.
Ongoing research projects:
