<p>This paper proposes an optimization algorithm developed based on bionic principles, reflecting the ability of mountain goats to assess risk in their natural habitat and adapt to individual physical capabilities. The proposed approach formalizes the behavioral strategies of mountain goats in complex terrain as a mathematical model, including conscious risk assessment, efficient management of energy resources, and maintenance of stable balance during movement. During the search process, a risk-avoidance coefficient and an energy management mechanism are employed, enabling an adaptive balance between global exploration of the solution space and intensive local exploitation in promising regions. The proposed algorithm was experimentally evaluated on classical benchmark functions: Sphere, Rosenbrock, Rastrigin, and Ackley. The obtained results were compared with particle swarm optimization and differential evolution using multiple performance criteria. Experimental findings demonstrate that the algorithm achieves fast and stable convergence on smooth functions, while providing reliable and competitive solutions on complex multimodal landscapes. Overall, the results confirm that the proposed approach is an effective, adaptive, and practically reliable tool for solving high-dimensional optimization problems. Keywords: metaheuristics; evolutionary algorithm; energy model; bionic optimization; risk management mechanism; convergence.</p>