Abstract:
This study considered a model of an Ant Colony Optimization
(ACO) algorithm for the general combinatorial optimization problem. The
model proved that it can converge to one of the optima if only this optimum
is allowed to update the pheromone model and that it can not converge
to any of the optima if two or more optima are allowed. The iteration
complexity of the model can be computed easily. And then a lower bound
of time complexity of a real ACO algorithm for the general combinatorial
optimization problem can be obtained.
Subscribe Today 
