This paper introduces a new stochastic bio-inspired optimization algorithm, denoted as seasons optimization (SO) algorithm. This algorithm is inspired by the growth cycle of trees in different seasons of a year. It is an iterative and population-based algorithm working with a population of initial solutions known as a forest. Each individual in the forest is referred to as a tree. Until the termination conditions are satisfied, the trees in the forest are updated to a new generation by applying four operators similar to the trees’ life cycles in nature: renew, competition, seeding, and resistance. These operators hopefully cause the trees to converge towards the global optimum of the optimization problem. The effectiveness of the proposed SO algorithm is evaluated using multi-variable single-objective test problems and compared with several well-known baseline and state-of-the-art algorithms. The results show that the proposed algorithm outperformed its counterparts in terms of solution quality and finding the global optimum on most benchmark functions.
