A comparative study of genetic algorithm components in simulation-based optimisation

In this paper, we present a comparative study of different stochastic components of genetic algorithms for simulationbased optimisation of the buffer allocation problem. We explore the effects of elements such as operators, fitness assignment strategies and elitism. Three different recombination operators, incorporated with constraint handling mechanisms such as repair and penalty functions, are examined. Under the shed of the experiments, we incorporate problem specific knowledge to further enhance the practicality of GA in decision making for buffer allocation problem.

• Type: Conference
• Year: 2008
• Where: WSC
• Authors: Birkan Can, Andreas Beham, Cathal Heavey
• Keywords: Buffer Allocation Problem, Fitness Assignment Strategies, Modeling And Simulation, Problem Specific Knowledge, WSC 2008