Prof. Dr. Tibor Csendes and his colleagues at the Department of Computational Optimization conduct research mainly in the fields of optimization and reliable computer procedures.

Among optimization models, those involving non-linear functions are frequently used especially in higher dimensional spaces, when a number of variables are to be optimized. To solve such problems they develop general-purpose algorithms, which are especially efficient in finding solutions for problems with a higher computational complexity. Many of their programmes are used for both basic and applied research and also for development purposes worldwide, such as for designing optimal public lights and for scheduling automotive industry production.

The reliable numerical algorithms developed by the research team are used during the control of hazardous mechanical systems, during solving some theoretical problems with mathematical rigour, and also during the automatic validation of data uncertainty. 

The industrial research project conducted with a micro-simulation method is one of the most significant achievements of the research team; they confirmed the advantageous effects of time-based tickets in terms of economic and traffic technology in the case of a middle-sized city’s public transportation system.  The micro-simulation method proved to be adequate for the project’s purposes, and was also appropriate for the numerical expression of the economic, passenger satisfaction and performance results. 

A new research direction is detecting quality within systems described by networks based on the PageRank algorithm and its further developed versions. The target fields are the evaluation of wine tasting tests on the basis of wine-tasters’ votes (similar to tennis and chess rankings), and the refinement of scientometric indicators according to the directed graph constructed on the basis of references.