Eurasian lynx optimizer: a novel metaheuristic optimization algorithm for global optimization and engineering applications

Abstract

Meta-heuristic algorithms, despite advancements, still face challenges in universally optimizing solutions across various problem domains. There is a persistent need for algorithms that can demonstrate superior performance in both theoretical benchmarks and real-world applications. This research aims to introduce a novel optimization algorithm, the Eurasian Lynx Optimizer (ELO), inspired by the adaptive hunting and survival strategies of the Eurasian lynx, targeting enhanced convergence accuracy and robustness. The ELO algorithm integrates a suite of innovative strategies mirroring the lynx's behavior, including prey pursuit, predator evasion, and adaptive movement patterns. It operates through a three-phase iterative process: exploration, exploration & exploitation, and exploitation, each utilizing distinct search mechanisms. These stages incorporate strategies like direct prey attack, enemy evasion, differential mutation, Lévy Flight, Gaussian mutation, and population consolidation. Comprehensive evaluations against ten state-of-the-art algorithms on 63 benchmark functions across CEC2014, CEC2017, and CEC2019 sets revealed that ELO achieved the best solution in 81% of cases. It consistently ranked first with averages of 1.6, 1.0, and 1.4 across respective benchmark sets, showcasing superior performance. Furthermore, ELO successfully tackled four constrained engineering design optimization problems, validating its practical utility.