A novel reindeer cyclone optimization algorithm (RCOA).

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• Optimization problems require efficient algorithms to balance exploration and exploitation.
• Existing algorithms like PSO, DE, COA, GSA, and WDO have limitations in performance and stability.
• Nature-inspired algorithms offer promising approaches to address these challenges.

Purpose of the Study:

• Introduce the novel Reindeer Cyclone Optimization Algorithm (RCOA).
• Evaluate RCOA's performance against established algorithms on benchmark functions and real-world problems.
• Demonstrate RCOA's effectiveness in terms of accuracy, convergence speed, and stability.

Main Methods:

• Developed RCOA based on reindeer survival behavior during predator attacks.
• Tested RCOA on 14 unimodal/multimodal benchmark functions and 4 real-world problems.
• Compared RCOA against PSO, DE, COA, GSA, WDO, and WOA using the CEC'17 benchmark suite (50 dimensions).

Main Results:

• RCOA showed 5-12% improvement over PSO, DE, COA, GSA on unimodal functions.
• RCOA demonstrated 10-15% improvement in accuracy and consistency over WDO and PSO on multimodal functions.
• RCOA achieved superior convergence speed and solution accuracy, outperforming existing methods on multiple test functions.

Conclusions:

• RCOA is a robust and reliable optimization method.
• Statistical significance confirmed via Wilcoxon Signed-Rank test.
• RCOA is suitable for a wide range of real-world optimization applications.