ABC-TLBO: A Hybrid Algorithm Based on Artificial Bee Colony and Teaching-Learning-Based Optimization

Artificial bee colony (ABC) is shown to be effective in solving optimization problems because of its good exploration capability. However it suffers from poor exploitation because onlooker bees use the same searching method as employed bees. In order to remedy the limited exploitation capability of ABC and achieve a better balance between exploration and exploitation, searching strategies for employed bees and onlooker bees are redesigned respectively and a novel algorithm called ABC-TLBO is proposed. It inherits the framework of basic ABC, and meanwhile an individualized TLBO based on teaching-learning-based optimization (TLBO) is embedded in onlooker bee phase to improve the exploitation capability of the presented algorithm. Moreover, a new searching strategy for employed bee phase which combines direction guiding searching and differential evolution crossover searching is applied to increase the population diversity and accelerate the convergence. The proposed ABC-TLBO algorithm is validated on benchmark functions and it is compared with ABC-based, TLBO-based and nature-inspired algorithms. Experimental results reveal that ABC-TLBO algorithm is an efficient, fast converging optimization method for solving global optimization problems.