Framework for data-driven scheduling under uncertainty
- Process Systems Engineering
- Focus/Key Topic:
Nowadays, large amounts of performance data is available characterizing processes. These can be made accessible for operational optimization by using data-driven modeling approaches from the machine learning. In particular, there is a recent trend to use artificial neural networks for this purpose. However, if incorporated into scheduling problems, hard-to-solve nonlinear optimizations problems arise due to large number of degrees of freedom. This issue becomes even more serious, once uncertainties are to be considered. Consequently, novel approaches for scheduling explicitly addressing these challenges are highly required. If you are interested in a thesis within this context, do not hesitate to contact me to arrange a personal discussion. Please always attach a short CV as well as your recent overview of your grades.