An Enhanced Aggregation Pheromone System for Real-Parameter Optimization in the ACO Metaphor

In previous papers we proposed an algorithm for real parameter optimization called the Aggregation Pheromone System (APS). The APS replaces pheromone trails in traditional ACO with aggregation pheromones. The pheromone update rule is applied in a way similar to that of ACO. In this paper, we proposed an enhanced APS (eAPS), which uses a colony model with units. It allows a stronger exploitation of better solutions found and at the same time it can prevent premature stagnation of the search. Experimental results showed eAPS has higher performance than APS. It has also shown that the parameter settings for eAPS are more robust than for APS.