An elitist genetic algorithm for multiobjective optimization
Solving multiobjective engineering problems is, in general, a difficult task. In spite of the success of many approaches, elitism has emerged has an effective way of improving the performance of algorithms. In this paper, a new elitist scheme, by which it is possible to control the size of the elite...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | conferencePaper |
| Language: | eng |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1822/19141 |
| Country: | Portugal |
| Oai: | oai:repositorium.sdum.uminho.pt:1822/19141 |
| Summary: | Solving multiobjective engineering problems is, in general, a difficult task. In spite of the success of many approaches, elitism has emerged has an effective way of improving the performance of algorithms. In this paper, a new elitist scheme, by which it is possible to control the size of the elite population, as well as the concentration of points in the approximation to the Pareto-optimal set, is introduced. This new scheme is tested on several multiobjective problems and, it proves to lead to a good compromise between computational time and size of the elite population. |
|---|