A superior attraction bacterial foraging optimizer for global optimization

In order to improve the performance of basic bacterial foraging optimization (BFO) for various global optimization problems, a superior attraction bacterial foraging optimizer (SABFO) is proposed in this paper. In SABFO, a novel movement guiding technique termed as superior attraction strategy is introduced to make use of all bacteria historical experience as potential exemplars to lead individuals direction. This strategy enables the bacteria in population to exchange information and collaborate with the superior individuals to search better solutions for different dimensions. Two variants of SABFO are studied and tested on a set of sixteen benchmark functions including various properties, such as unimodal, multimodal, shifted and inseparable characteristics. Four state-of-the-art evolutionary algorithms are adopted for comparison. Experimental study demonstrates remarkable improvement of the proposed algorithm for global optimization problems in terms of solution accuracy and convergence speed. Key–Words: Global optimization; Bacterial foraging optimization; Swarm intelligence; Engineering optimization; Movement updating; Meta-heuristic; Evolutionary algorithms.