Abstract

This study introduces a new algorithm for the ant colony optimization (ACO) method, which has been proposed to solve global optimization problems with continuous decision variables. This algorithm, namely ACO-FRS, involves a strategy for the selection of feasible regions during optimization search and it performs the exploration of the search space using a similar approach to that used by the ants during the search of food. Four variants of this algorithm have been tested in several benchmark problems and the results of this study have been compared with those reported in literature for other ACO-type methods for continuous spaces. Overall, the results show that the incorporation of the selection of feasible regions allows the performing of a global search to explore those regions with low level of pheromone, thus increasing the feasibility of ACO for finding the global optimal solution.

Abstract

This work presents a multiobjective algorithm for multicast traffic engineering. The proposed algorithm is a new version of MultiObjective Ant Colony System (MOACS), based on Ant Colony Optimization (ACO). The proposed MOACS simultaneously optimizes the maximum link utilization, the cost of the multicast tree, the averages delay and the maximum endtoend delay. In this way, a set of optimal solutions, known as Pareto set is calculated in only one run of the algorithm, without a priori restrictions. Experimental results obtained with the proposed MOACS were compared to a recently published Multiobjective Multicast Algorithm (MMA), showing a promising performance advantage for multicast traffic engineering. 5th IFIP International Conference on Network Control & Engineering for QoS, Security and Mobility Red de Universidades con Carreras en Informática (RedUNCI) Document type: Part of book or chapter of book