Studies On The Dynamics Of Ant Colony Optimization Algorithms
Daniel M. Merkle, Martin Middendorf · 2002
A deterministic model for Ant Colony Opti-mization (ACO) algorithms is proposed and used to study the dynamics of ACO. The model is based on the average expected be-haviour of ants. The behaviour of ACO algo-rithms and the model are analysed for certain types of permutation problems. It is shown numerically that decisions of the ants are in-uenced in an intriguing way by the proper-ties of the pheromone matrix. This explains why ACO algorithms show a complex dy-namic behaviour. Simulations are done to compare the behaviour of the ACO model with the ACO algorithm. The results show that the model describes essential features of the dynamics of ACO algorithms. 1