Analysis of the stagnation behavior of the interacted multiple ant colonies optimization framework
Search Stagnation is a common problem 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...
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | https://repo.uum.edu.my/id/eprint/9248/1/a.pdf |
| Summary: | Search Stagnation is a common problem 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 conducts experimental tests to analyze the stagnation behavior of IMACO.It also proposes the idea that different ant colonies use different types of problem dependent heuristics.The performance of IMACO was demonstrated by comparing it with the Ant Colony
System (ACS) the best performing ant algorithm.The
Computational results show the superiority of IMACO.
The results show that IMACO suffers less from
stagnation than ACS. |
|---|