My research deals with the computational aspects of mixed-integer linear and nonlinear programming. I am leading the research group Mathematical Optimization Methods at Zuse Institute Berlin, which holds long-standing expertise in developing and implementing general solver software for linear, mixed-integer linear and nonlinear programming.
In the long-running project Linear, Integer, and Constraint Programming we develop the MINLP solver SCIP and the LP solver SoPlex. I have become increasingly interested in exact linear and integer programming over the rational numbers and implemented a number of rational extensions to SoPlex around the ideas in this paper and my PhD thesis. Recently, I am collaborating on a new software project for verifying the correctness of integer programming results: VIPR.
The exact solving features of SoPlex are available on the NEOS Server for Optimization and have been interfaced to the Python package cobraPy in order to solve multi-scale linear programming formulations over metabolic networks. This has triggered my interest in the use of optimization methodology for questions from computational biology. Together with the research group Computational Molecular Design at ZIB, we investigate complex clustering problems appearing in transition networks.
I am responsible for our cooperations with SAP and Siemens that support the development of SCIP and SoPlex with a special focus on their industrial applications. In the recent project BEAM-ME together with partners from DLR, Jülich Supercomputing, HLRN Stuttgart, and TU Berlin, we investigate parallel methods to solve large-scale optimization problems for energy system models.
If you are interested to work in our research group as a student, research assistant, or postdoc, send me an e-mail.
© 2008-2017. Imprint.