Interacted multiple ant colonies optimization framework: An experimental study of the evaluation and the exploration techniques to control the search stagnation
Search stagnation is a serius prblem that all Ant Colony Optimization (ACO) algorithms suffer from regardless of their application domain. The framework of Interacted Multiple Ant Colonies Optimization (IMACO) is a recent proposition.It divides the ants'population into several colonies and empl...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AICIT, Korea
2010
|
| Subjects: | |
| Online Access: | http://repo.uum.edu.my/3929/1/Al.pdf http://repo.uum.edu.my/3929/ http://dx.doi.org/10.4156/ijact.vol2.issue1.7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Search stagnation is a serius prblem that all Ant Colony Optimization (ACO) algorithms suffer from regardless of their application domain. The framework of Interacted Multiple Ant Colonies Optimization (IMACO) is a recent proposition.It divides the ants'population into several colonies and employs certain techniques to organize the work of these colonies.This paper proposes new effective
evaluation and exploration techniques for IMACO and experimentallv tests the stagnation behavior of IMACO. The performance of IMACO was demonstrated by comparing it with the best performing ant algorithms like Ant Colony System (ACS) and Max-Min Ant System (MMAS). The computational results show the superiority of lMACO. The results comparison shows that lMACO with the proposed techniques suffers less fronm stagnation than the best known ant algorithms of ACS and MMAS. |
|---|