A New Ant Colony Optimization Algorithm : Three Bound Ant System (CROSBI ID 208281)
Prilog u časopisu | kratko priopćenje | međunarodna recenzija
Podaci o odgovornosti
Ivković, Nikola ; Golub, Marin
engleski
A New Ant Colony Optimization Algorithm : Three Bound Ant System
Abstract. Since its introduction, ant colony optimization (ACO) algorithms has been gradually improved and now can offer very good performance for many hard discrete optimization problems. Among the most popular variant of ACO is MAX-MIN ant system (MMAS), which have excellent properties for many optimization problems. In this paper, we propose a new ACO algorithm, named three bounds ant system (TBAS), which we perceive as further evolution and improvement of MMAS. We analyze some properties of TBAS algorithm and experimentally compare its performance with MMAS algorithm. The TBAS has lower computational complexity than MMAS, but its speedup can vary from negligible to very significant, depending on the type of optimization problem, a way the algorithm is implemented and the parameter settings. Moreover, it is experimentally shown that TBAS algorithm is competitive with MMAS algorithm in means of solution quality, as it has achieved better results than MMAS on most of the tested instances of quadratic assignment problem (QAP).
algorithm ; optimization ; ant colony optimization ; swarm intelligence
Sažetak s kongresa? 2017-01-01, admin
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano