Weather State Ants Optimizer (WSAO): A Novel Dynamic Variable Structure Metaheuristic Algorithm Incorporating Markov Processes

The performance of metaheuristic algorithms in solving complex optimization problems critically depends on an effective balance between exploration and exploitation. While existing variable-structure algorithms attempt to address this challenge through predefined stage-wise switching strategies, their rigid transition rules often limit adaptability in complex search landscapes, leading to issues such as slow convergence or premature convergence to local optima. To overcome these limitations, this paper proposes the Weather State Ants Optimizer (WSAO), a novel dynamic variable structure metaheuristic algorithm that simulates the intelligent foraging behavior of ants under varying weather conditions. The primary innovation lies in the introduction of a Markov process-driven dynamic weather state machine. At each generation, the algorithm transitions between three distinct states according to the Markov process, each governing a fundamentally different optimization structure dedicated to either exploration or exploitation. This design enables probabilistic and memoryless switching of search modes throughout the optimization process, achieving real-time dynamic adjustment of the exploration–exploitation trade-off. A second key innovation is the incorporation of a dynamic nest mechanism, where elite solutions establish multiple search centers, enabling concurrent exploitation of multiple promising regions and substantially improving performance on multimodal problems. To validate the superiority of WSAO, we conducted a comprehensive comparison with 11 state-of-the-art algorithms on CEC2017 and CEC2022 benchmark functions, as well as 5 constrained engineering problems. Statistical results show that WSAO achieves leading or highly competitive results in a clear majority of test cases, particularly demonstrating clear advantages in complex multimodal and hybrid functions. This work not only presents a powerful optimizer but, more importantly, pioneers the integration of Markov processes into metaheuristic algorithms, establishing a new paradigm for dynamic variable-structure optimization with significant implications for the field. The source code and test results of the WSAO algorithm is available at https://github.com/xiaxiubo/WSAO-Weather-State-Ants-Optimizer.