Fossa Optimization Algorithm: A New Bio-Inspired Metaheuristic Algorithm for Engineering Applications

This paper presents a novel bio-inspired metaheuristic algorithm termed the Fossa Optimization Algorithm (FOA), which emulates the natural hunting behavior of the fossa in its habitat.FOA draws its core inspiration from the two-stage hunting technique of the fossa, involving an initial attack on a spotted lemur followed by a pursuit through the trees.The theoretical framework of FOA is elaborated, and its implementation is mathematically modeled in two distinct phases: (i) exploration, which simulates the fossa's positional adjustments during the initial attack on the lemur, and (ii) exploitation, which models the positional changes of the fossa during the chase.The efficacy of FOA is tested against twenty-two constrained optimization problems from the CEC 2011 test suite, as well as four engineering design challenges.The optimization results demonstrate FOA's strong capabilities in both exploration and exploitation, maintaining a balance that facilitates convergence to optimal solutions.FOA's performance is benchmarked against twelve established algorithms, showing that it consistently outperforms its competitors by delivering superior results and ranking as the top optimizer in most of the evaluated functions.These findings indicate that FOA is highly effective in addressing optimization tasks in real-world scenarios.