A novel artificial bee colony algorithm based on modified search equation and orthogonal learning.

Area of Science:

• Computational intelligence
• Swarm intelligence
• Optimization algorithms

Background:

• The artificial bee colony (ABC) algorithm is a population-based metaheuristic known for its exploration capabilities but suffers from poor exploitation.
• Existing ABC variants struggle to balance exploration and exploitation effectively, limiting their performance on complex optimization problems.

Purpose of the Study:

• To address the exploitation deficiency in the ABC algorithm.
• To introduce a novel orthogonal learning (OL) strategy, leveraging orthogonal experimental design (OED), to enhance ABC's search efficiency.
• To develop and evaluate improved ABC variants, including CABC, OABC, OGABC, and OCABC.

Main Methods:

• A modified search equation was developed for the CABC algorithm to improve candidate solution generation.
• An orthogonal learning (OL) strategy was integrated with OED to extract more valuable information from search experiences.
• The OL strategy was applied to standard ABC, global-best-guided ABC (GABC), and CABC, resulting in OABC, OGABC, and OCABC.

Main Results:

• Experimental results on 22 benchmark functions validate the effectiveness of the modified search equation and OL strategy.
• The proposed algorithms, particularly OCABC, significantly outperform other ABC variants and state-of-the-art algorithms.
• OCABC achieved the highest solution quality, fastest global convergence, and strongest robustness across most test functions.

Conclusions:

• The proposed CABC algorithm with its modified search equation effectively improves ABC's exploitation ability.
• The orthogonal learning strategy enhances the performance of various ABC algorithms by constructing more efficient candidate solutions.
• OCABC represents a significant advancement in swarm intelligence, offering superior performance for complex optimization challenges.