Pareto-based branch and bound algorithm for multiobjective optimization of a safety transformer

Abstract

Purpose

This paper aims to propose a multiobjective branch and bound (MOBB) algorithm with a new criteria for the branching and discarding of nodes based on Pareto dominance and contribution metric.

Design/methodology/approach

A multiobjective branch and bound (MOBB) method is presented and applied to the bi-objective combinatorial optimization of a safety transformer. A comparison with exhaustive enumeration and non-dominated sorting genetic algorithm (NSGA2) confirms the solutions.

Findings

It appears that MOBB and NSGA2 are both sensitive to their control parameters. The parameters for the MOBB algorithm are the number of starting points and the number of solutions on the relaxed Pareto front. The parameters of NSGA2 are the population size and the number of generations.

Originality/value

The comparison with exhaustive enumeration confirms that the proposed algorithm is able to find the complete set of non-dominated solutions in about 235 times fewer evaluations. As this last method is exact, its confidence level is higher.

Citation

Brisset, S. and Tran, T.-V. (2018), "Pareto-based branch and bound algorithm for multiobjective optimization of a safety transformer", COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, Vol. 37 No. 2, pp. 617-629. https://doi.org/10.1108/COMPEL-11-2016-0505