A new metaheuristic and its application to the Steiner problems in graphs
Metaheuristics are a set of related ideas that provide a general framework to obtain the approximated solution of combinatorial optimization problems. We present ideas for a new method that we call SN. SN is based on the idea of dividing a given optimization problem into decision subproblems which a...
Guardado en:
| Autores principales: | , |
|---|---|
| Publicado: |
2001
|
| Materias: | |
| Acceso en línea: | https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15224902_v2001-January_n_p273_Urrutia http://hdl.handle.net/20.500.12110/paper_15224902_v2001-January_n_p273_Urrutia |
| Aporte de: |
| Sumario: | Metaheuristics are a set of related ideas that provide a general framework to obtain the approximated solution of combinatorial optimization problems. We present ideas for a new method that we call SN. SN is based on the idea of dividing a given optimization problem into decision subproblems which are easier to solve heuristically. In order to verify the effectiveness of this metaheuristic approach, we developed an algorithm to solve a classic graph theory problem, the Steiner Problem in Graphs. Preliminary results show that the new algorithm is competitive when compared with the best heuristic algorithms known for this problem. © 2001 IEEE. |
|---|