Naslov
Use Fibonacci Numbers to Improve Performance of a Genetic Algorithm

Autori
Gudelj, Anita ; Kezić, Danko

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Skup
20th Conference of the International Federation of Operational Research Societies IFORS 2014

Mjesto i datum
Barcelona, Španjolska, 13.07.2014. - 18.07.2014

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Genetic algorithm, Petri net, Variable Size Population, Fibonacci sequence

Sažetak
In this paper, we focus on the effect of variable population size on accelerating evolution in the context of our algorithm which integrates MRF1 Petri net with GA. Our approach uses Fibonacci sequence to select the number of individuals in populations. The motivation is to add new individuals when the GA is reaching a stagnation phase and remove individuals when the GA process is progressing well. This variable size population model we tested on some scheduling problems with shared resources. Experimental results confirm that our model finds solutions of similar quality to the ones found by Standard Genetic Algorithm, but with a smaller amount of computational effort.

Izvorni jezik
Engleski

Znanstvena područja
Tehnologija prometa i transport, Informacijske i komunikacijske znanosti

POVEZANOST RADA