Post-processing improvements in multi-objective optimization of general single-server finite queueing networks.

Abstract
An alternative mathematical programming formulation is considered for a mixed-integer optimization problem in queueing networks. The sum of the blocking probabilities of a general service time, single server, and the finite, acyclic queueing network is minimized, and so are the total buffer sizes and the overall service rates. A multi-objective genetic algorithm (MOGA) and a particle swarm optimization (MOPSO) algorithm are combined to solve this difficult stochastic problem. The derived algorithm produces a set of efficient solutions for multiple objectives in the objective function. The implementation of the optimization algorithms is dependent on the generalized expansion method (GEM), a classical tool used to evaluate the performance of finite queueing networks. We carried out a set of computational experiments to attest to the efficacy and efficiency of the proposed approach. In addition, we present a comparative analysis of the solutions before and after post-processing. Insights obtained from the study of complex queue networks may assist the planning of these types of queueing networks.
Description
Keywords
Conflicting objectives, Buffer allocation, Particle swarm optimization
Citation
SOUZA, G. L. de et al. Post-processing improvements in multi-objective optimization of general single-server finite queueing networks. IEEE Latin America Transactions, v. 21, n. 3, p. 381-388, mar. 2023. Disponível em: <https://latamt.ieeer9.org/index.php/transactions/article/view/7020>. Acesso em: 06 jul. 2023.