Thorny Devil Optimization (TDO): A Novel Metaheuristic Algorithm for Solving Optimization Problems

In this study, a new metaheuristic algorithm called Thorny Devil Optimization (TDO) is introduced, inspired by the natural behaviors of the thorny devil.The main idea behind the design of this algorithm is based on two distinctive behaviors of this animal: first, broad searching to locate ant colonies, and second, focused feeding within the discovered colony.These two behaviors are considered as models of exploration and exploitation, respectively, and together they form the core structure of the algorithm's search mechanism.In this research, the theoretical foundations and behavioral concepts of the algorithm are first explained in detail, and then these concepts are mathematically modeled within two phases: exploration and exploitation.In the exploration phase, the wide-ranging movements of the thorny devil while searching for food sources are simulated so that the algorithm's agents can globally investigate the solution space.In the exploitation phase, the animal's focused feeding behavior is emulated, enabling the agents to perform small and precise movements around promising solutions to enhance the quality of the results.The results of the sensitivity analysis show that population size and the maximum number of iterations have a direct impact on solution quality and convergence stability, and selecting appropriate values for these two parameters can create a good balance between accuracy and computational cost.The performance of the TDO algorithm is evaluated on 23 standard benchmark functions, including unimodal and multimodal functions, and compared with nine competing algorithms.The results indicate that TDO demonstrates strong exploitation capability on unimodal functions, possesses considerable exploration power on high-dimensional multimodal functions, and maintains a proper balance between exploration and exploitation on fixed-dimension multimodal functions.The comparison of TDO with nine competing algorithms shows that, by achieving first rank as the best optimizer on all 23 benchmark functions, TDO attains a 100 percent successful performance in competition with the rival algorithms.The simulation findings indicate that TDO is an effective, stable, and reliable optimizer for solving a wide range of optimization problems in scientific domains and real-world applications.

Hali tarjima qilinmagan