Institut für Angewandte Informatik und Formale Beschreibungsverfahren
(Institute for Applied Computer Science and Formal Description Methods),
Universitä Karlsruhe, D-76128 Karlsuhe, Germany

Daniel Merkle and Martin Middendorf. A New Approach to Solve Permutation Scheduling Problems with Ant Colony Optimization.
In: S. Cagnoni, J. Gottlieb, E. Hart, M. Middendorf, G.R. Raidl (Eds.) Applications of Evolutionary Computing: EvoWorkshops 2002: EvoCOP, EvoIASP, EvoSTIM/EvoPLAN Kinsale, Ireland, April 3-4, 2002. Proceedings, Springer Verlag, LNCS 2279 (2002) 325-333.

In this paper the relative pheromone evaluation method for Ant Colony Optimization is investigated. We compare this method to the standard pheromone method and the summation method. Moreover we propose a new variant of the relative pheromone evaluation method. Experiments performed for various instances of the single machine scheduling problems with earliness costs and multiple due dates show the potential of the relative pheromone evaluation method.

Publications: Martin Middendorf