Apiary Organizational-Based Optimization Algorithm: A New Nature-Inspired Metaheuristic Algorithm

Abstract

The pursuit of efficient solutions in optimization has led to the development of metaheuristic algorithms (MHAs). Some of these algorithms adopt the behaviour of several different animals and insects. However, researchers still attempt to improve the performance of such algorithms. This study proposes a new metaheuristic optimization algorithm which is so-called the Apiary Organizational-Based Optimization Algorithm (AOOA). The proposed algorithm introduced a new concept of multiple populations inspired by the organizational behaviour of honeybees inside the apiary. Honeybees are a highly systematic and organized society that lives within an apiary and achieves specific tasks during their lifecycle. To develop the proposed algorithm, the key activities of the queen, drones, and workers inside the apiary are determined first. These activities are translated into several different phases to develop a mathematical model that represents the ground of the proposed AOOA. 23 classical unimodal, multimodal, and fixed-dimension benchmark functions are used to verify AOOA performance. The results are compared with 6 recent MHAs, puzzle optimization algorithm (POA), coati optimization algorithm (COA), Average and Subtraction-Based Optimizer (ASBO), guided pelican algorithm (GPA), Golden Search Optimization Algorithm (GSO) and extended stochastic coati optimizer (ESCO), showing that AOOA outperforms them in solving 21, 18, 20, 18, 21 and 15 functions, respectively. AOOA was superior in mean fitness of 17 out of 23 functions with superiority of 86%, 100%, and 50% of unimodal, multimodal, and fixed-dimension functions, respectively indicating the competitiveness of the proposed algorithm in providing more appropriate solutions.