An Adaptive Multi-Objective Artificial Bee Colony with crowding distance mechanism

In this work, we propose an Adaptive Multi-Objective Artificial Bee Colony (A-MOABC) Optimizer that uses Pareto dominance procedure with taking the advantage of crowding distance and windowing mechanism. The employed bees use an adaptive windowing mechanism to select their own leaders and alter their positions. Besides, onlookers update their positions by using food sources presented by employed bees. Pareto dominance notion is used to show the quality of the food sources. Employed or onlooker bees which find poor quality food sources turn into scout bee to search other areas. The suggested method uses crowding distance technique in order to keep diversity in the archive. The method adaptively adjusts the limits of objective function values in the archive iteration by iteration. The experimental results indicate that the proposed approach not only thoroughly competitive compared to other algorithms considered in this work but also finds the result with greater precision.