A Behavioural Study of the Crossover Operator in Diploid Genetic Algorithms

Petrovan, Adrian; Matei, Oliviu; Erdei, Rudolf

doi:10.1007/978-3-030-57802-2_8

Cited by 2 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dulebenets [14] propose a Diploid Evolutionary Algorithm (DEA) that can assist the cross-docking operators with the design of cost-efficient truck schedules, that can facilitate the flow of different products within the cross-docking facilities, ensure the "just-in-time" deliveries within supply chains, and improve sustainability of the supply chain operations. Petrovan et al presented some of the advantages of diploid GAs in comparison to standard GAs using real-coded chromosome representation [33] and a behavioural study of the crossover operators in diploid genetic algorithms [34]. Jasuja [23] used DGA on multi-objective optimization of a classification problem for feature selection, showing the advantages of DGA in dealing with this kind of problems.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Study between Haploid Genetic Algorithms and Diploid Genetic Algorithms

PETROVAN¹,

Matei²,

Pop³

2022

CJM

View full text Add to dashboard Cite

In this paper, we make a comprehensive comparison in terms of the quality of the achieved solutions, the corresponding execution time and impact of the genetic operators on the quality of the results between the Haploid Genetic Algorithms (HGAs) and Diploid Genetic Algorithms (DGAs). The standard genetic algorithms, referred to in our paper as HGAs are characterized by the fact that they are using a haploid representation relating an individual with a chromosome, while the DGAs are using diploid individuals which are made of two chromosomes corresponding to the dominant and recessive genes. Even though the general opinion is that DGAs do not provide much benefit as compared to classical GAs, based on extensive computational experiments, we do show that the DGAs are robust, have a high degree of consistency and perform better, sometimes almost twice as well, than the HGAs, but are slower due to the high number of operations to be performed, caused by the duplication of the genetic information. However, the quality of the solutions achieved by the DGAs compensate their relative high execution time. The better quality of the DGAs, proving the efficiency of using diploid genes, is given by the homogeneity of the population which covers the search space thoroughly and in this way being capable of avoiding the local optima.

show abstract