Anatolian Caracal Optimization Algorithm: A Three-Stage Hybrid Bio- and Physics-Inspired Framework for Optimization Problems

The Anatolian Caracal Optimization Algorithm (ACOA), a swarm-based optimization approach inspired by nature, approximately describes the characteristic hunting strategy of the Anatolian caracal, shaped by its superior auditory–visual perception and extraordinary agility, using mathematical models. The algorithm structures this process into three main stages by integrating and modifying established search mechanisms within a sequential framework: (i) wide-area scanning (exploration), which supports a broad exploration of the solution space through prey-focused movements; (ii) tracking–approach, which uses Brownian (continuous, small steps) and Lévy (occasional long jumps) movements to support the transition from exploration to exploitation; and (iii) predatory leap (exploitation), which represents the Anatolian caracal’s jump with a dimensionless projectile-inspired formulation and supports the refinement of candidate solutions. ACOA has been benchmarked across the CEC 2017, CEC 2019, CEC 2020, and CEC 2022 test suites, including unimodal and multimodal functions with fixed and varying dimensions, as well as real-world engineering design problems, for a total of 131 functions. It has been compared against 11 recent and well-known optimization algorithms. The best mean performance is achieved by ACOA in 111 out of 131 cases. In terms of stability, consistently low standard deviation is maintained by ACOA across the majority of the evaluated functions, indicating a strong reproducibility of the obtained solutions across independent runs. These findings indicate that the proposed three-stage framework provides a competitive exploration–exploitation behavior and can offer effective and stable solutions for a wide range of benchmark and engineering optimization problems.