Methods and algorithms for solving deterministic and stochastic problems for modelling of dynamic decision systems
| Programmee: | Institutional Projects (Supreme Council for Science and Technological Development) |
| Execution period: | 2011 – 2014 |
| Institutions: | Academy of Sciences of Moldova, Institute of Mathematics and Computer Science |
| Project Leader: | Lozovanu Dmitrii |
| Participants: | Mathematical Modeling |
| Keywords: | mathematical modelling, optimal control, game theory, high performance computing, probability theory, queuing theory, theory of algorithms |
Summary
The main objectives of the project comprise:
- Computational complexity analysis of the existing discrete optimal control and queuing theory problems, as well as determining of the optimal stationary strategies in Markov decision processes;
- Investigation of Markov and Semi-Markov decision processes and elaboration of the algorithms for determining the optimal stationary strategies in these processes basing on dynamic programming and linear programming methods;
- Elaboration of algorithms for solving of discrete optimal control problems for deterministic and stochastic cases;
- Elaboration and testing of numerical algorithms for queuing models for different classes of priorities;
- Investigation of mathematical growing economic models and elaboration of algorithms for solving of problems from applied mechanics basing on parallel computing using high performance cluster with many processors.
