Fleet mix computation using evolutionary multiobjective optimization

We apply the non-dominated sorting genetic algorithm-II (NSGA-II) to perform a multiobjective optimization of the stochastic fleet estimation (SaFE) model. SaFE is a Monte Carlo-based model which generates a vehicle fleet based on the set of requirements that the fleet is supposed to accomplish. A genetic algorithm framework is used in order to alternate solutions between different plausible sets of platforms. We use SaFE coupled with simple monetary and temporal cost-based evaluations on the output of SaFE as the genetic algorithm´s fitness functions. Results showing the algorithm´s performance with respect to variations in the input parameters are presented and discussed.