Varying quality function in genetic algorithms and the cutting problem
- 17 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 166-169 vol.1
- https://doi.org/10.1109/icec.1994.350022
Abstract
In this paper, an implementation of a genetic algorithm (GA) is presented, using a quality function that is not unaltered but changes according to the search evolution. This means that the GA 'sees' a continuously changing search space, throughout one run. The example chosen to test the effect of a varying quality function is the cutting problem. Simulation results show that the dynamic quality function performs much better than its static counterpart.Keywords
This publication has 3 references indexed in Scilit:
- A genetic algorithm for training recurrent neural networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- A hybrid genetic algorithm for training neural networksPublished by Elsevier ,1992
- Two Algorithms for Constrained Two-Dimensional Cutting Stock ProblemsOperations Research, 1983