Fitness Landscape Analysis of Simulation Optimisation Problems with HeuristicLab

Proceedings of the UKSim 5th European Symposium on Computer Modeling and Simulation 2011
Vitālijs Boļšakovs, Erik Pitzer, Michael Affenzeller

In this paper the fitness landscape of a simulation optimisation problem is analysed within the metaheuristic optimisation framework HeuristicLab. Computational experiments are performed within an application prototype of a link between the model of a vehicle scheduling problem and the optimisation framework. Modern fitness landscape analysis approaches implemented in the HeuristicLab framework are applied and described. New insights are gained in the field of fitness landscapes analysis for stochastic problems. Application of fitness landscape analysis in simulation optimisation for meta-optimisation purposes is also considered in this paper. Comprehensive fitness landscape analysis of different instances of one type of optimisation problem and its results in comparison with results of optimisation experiments allows determining the impact of different landscape structures on the behaviour of different metaheuristic optimisation algorithms.

Atslēgas vārdi
simulation optimisation; fitness landscape analysis; vehicle scheduling; optimisation framework; metaheuristics
DOI
10.1109/EMS.2011.14
Hipersaite
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6131197

Boļšakovs, V., Pitzer, E., Affenzeller, M. Fitness Landscape Analysis of Simulation Optimisation Problems with HeuristicLab. No: Proceedings of the UKSim 5th European Symposium on Computer Modeling and Simulation, Spānija, Madrid, 16.-18. novembris, 2011. Los Alamitos, California: IEEE Computer Society Conference Publishing Services, 2011, 107.-112.lpp. ISBN 978-0-7695-4619-3. Pieejams: doi:10.1109/EMS.2011.14

Publikācijas valoda
English (en)