Proceedings of the 1999 Congress on Evolutionary Computation-Cec99 (Cat. No. 99TH8406)
DOI: 10.1109/cec.1999.781914
|View full text |Cite
|
Sign up to set email alerts
|

Evolutionary algorithms with goal and priority information for multi-objective optimization

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
40
0
2

Publication Types

Select...
3
3
1

Relationship

2
5

Authors

Journals

citations
Cited by 36 publications
(42 citation statements)
references
References 4 publications
0
40
0
2
Order By: Relevance
“…To examine the relationships among the different trajectory, multi-objective optimisation is necessary such that a set of trade-off models could be obtained and visualised before final determination of a model based upon the on-hand situation. Unlike conventional gradient-guided optimisation techniques, evolutionary algorithm can be easily extended for multi-objective optimisation in view of its parallel-based evolution feature (Fonseca, 1995;Tan et al, 1999). A multi-objective evolutionary algorithm is capable to approximate the set of non-dominated solutions known as Pareto optimal set, where each objective component of any model along the Pareto front can only be improved by degrading at least one of its other objective components.…”
Section: Trade-offs Among the Trajectorymentioning
confidence: 99%
“…To examine the relationships among the different trajectory, multi-objective optimisation is necessary such that a set of trade-off models could be obtained and visualised before final determination of a model based upon the on-hand situation. Unlike conventional gradient-guided optimisation techniques, evolutionary algorithm can be easily extended for multi-objective optimisation in view of its parallel-based evolution feature (Fonseca, 1995;Tan et al, 1999). A multi-objective evolutionary algorithm is capable to approximate the set of non-dominated solutions known as Pareto optimal set, where each objective component of any model along the Pareto front can only be improved by degrading at least one of its other objective components.…”
Section: Trade-offs Among the Trajectorymentioning
confidence: 99%
“…2), MOEA and IMOEA have the highest RNI on almost all the test problems (except test problem 2) as compared to others. This clearly indicates their abilities to provide more nondominated solutions from a given size of population based on the incorporation of switching preserved strategy (SPS) (Tan et al, 1999) where all the non-dominated individuals are preserved for the next generation. The RNI for SPEA and EMOEA are mainly depending on the ratio of the size of secondary (best-found) population to the size of main population.…”
Section: Performance Comparisonsmentioning
confidence: 99%
“…In order to guarantee a fair comparison, all algorithms considered are implemented with the same coding scheme, crossover and mutation. Note that each parameter is represented by 3-digit decimal and concatenated to form the chromosomes, which gives a shorter chromosome length and avoids the Hamming-cliff effect as encountered in traditional binarybased coding scheme (Tan et al, 1999). In all cases, standard mutation with a probability of 0.01 and standard crossover with two-point crossover and a probability of 0.7 are used.…”
Section: Performance Comparisonsmentioning
confidence: 99%
See 2 more Smart Citations