In many industrial and research projects, one attempts to improve a system by modifying its decision variables subject to constraints: this is optimization.

The research group focusses on the numerical solution of such problems, that is the effective calculation of the best values for the decision variables. We focus in particular on nonconvex and large-scale instances. Both theoretical questions, such as design and convergence properties of the algorithms, and associated software issues are studied.

Current projects cover the solution of very large-scale problems arising from discretized continuous applications, algorithmic implementations of filter methods, which enforce convergence of variants of Newton method without imposing monotonicity, the writing and dissemination of high quality scientific software (LANCELOT, GALAHAD, CUTEr, ...), nonconvex mixed-integer programming, preconditioning in oceanographic data assimilation and the design of progressive adaptive lenses.

Contact

Rue Rempart de la Vierge 8, 5000 Namur
Tel: 081/724934
Fax: 081/724914
Mail: