Worker Ant Optimization: An Algorithm for Complex Problems
Abstract
The article introduces a novel metaheuristic algorithm called the Worker Ant Optimization (WAO) algorithm. This algorithm is mathematically modeled based on five natural behaviors of worker ants: avoiding danger, foraging, approaching food, decomposing food, and transporting food. The performance of WAO was evaluated using 23 classical test functions and compared with results from seven well-known metaheuristic algorithms. Simulation results demonstrate that the WAO algorithm exhibits significant advantages in terms of convergence speed, avoidance of local optima, and optimization accuracy. To assess the effectiveness of WAO in practical applications, the method was applied to three classical engineering design problems, validating the engineering applicability of the WAO optimization algorithm. WAO effectively explores the decision space and performs well across various evaluation metrics, demonstrating its capability to effectively address challenges in practical applications.
Keywords
Download Options
Introduction
The term "optimization problem" refers to a situation where the goal is to find feasible solutions under given constraints [1]. It involves the process of seeking optimal values for specific system parameters within existing solutions, aiming to meet a certain criterion at minimal cost [2]. Such criteria could include maximizing profit, minimizing costs, maximizing efficiency, or minimizing risks. Optimization problems find widespread applications in fields such as engineering, economics, management, and computer science.
Typically, an optimization problem comprises several elements: decision variables, constraints, and an objective function [3]. In practical applications, optimization problems can be highly complex, often involving conflicting objective functions and numerous constraints. To address these challenges, enhance system performance, and reduce computational costs, various optimization methods have been developed. These methods are generally categorized into two classes: mathematical methods and metaheuristic algorithms.
In academic and applied contexts, optimizing system parameters involves leveraging these methodologies to achieve desired outcomes efficiently and effectively.
Conclusion
This paper presents a novel metaheuristic optimization algorithm called the Worker Ant Optimization (WAO) algorithm, designed to simulate various activities of worker ants in nature, including behaviors such as avoiding danger, foraging, approaching food, and decomposing and transporting food. The paper develops a mathematical optimization model based on these natural activities of worker ants and rigorously evaluates the convergence speed and search accuracy of WAO across 23 classic test functions. The quality of WAO optimization results is compared with the performance of seven well-known algorithms. Simulation results show that WAO exhibits excellent convergence speed, achieves a suitable balance between exploration in global search and exploitation in local search, and demonstrates a strong ability to escape local optima, providing effective solutions for optimization problems. Additionally, the WAO method is applied to three engineering design optimization problems, and its applicability to engineering optimization is validated. The comparisons with seven well-known optimization algorithms further demonstrate the advantages of the WAO algorithm in optimizing complex global optimization problems.
However, it is worth noting that there is still room for improvement in the convergence speed of WAO. While the algorithm shows strong performance in other aspects, optimizing its convergence rate could further enhance its overall efficiency and effectiveness in solving complex optimization problems. Future research could focus on refining the algorithm’s parameters and exploring hybrid approaches to accelerate convergence without compromising its robustness and solution quality.