Operator Kawin Silang pada Algoritma Genetik Riil untuk Variabel Rencana Selalu Positif

*Yoyong Arfiadi  -  Program Studi Teknik Sipil, Indonesia
Published: 27 Dec 2016.
Open Access
Citation Format:
Article Info
Section: Articles
Language: EN
Full Text:
Statistics: 508 1045
Abstract
Genetic algorithms have been used to solve various optimization problems. One of the advantages of genetic algorithms is that they have the ability to solve complex optimization problems in a simple way. By using genetic algorithms, the near global optimum can be achieved easily. Although in the early development, binary coded genetic algorithms are more popular, recently real coded genetic algorithms are widely used to solve engineering problem’s optimization. The advantage of using real coded genetic algorithms is the ability of the crossover operator to explore a larger domain of interest. As a result the use of crossover in real coded genetic algorithms may have a detrimental effect, as it can explore the domain that is very far from the expected domain. In the civil engineering area, most variables are positive. Therefore, it is needed to develop a crossover operator that can produce positive-only offspring. In this paper an asymmetric crossover is proposed to solve this problem. It is shown in the experiments that this crossover has a good performance in achieving optimum results.
Keywords
Real-coded genetic algorithms, Optimization, Crossover operator, Global optimum.

Article Metrics: