Toz, MetinToz, Gueliz2024-01-222024-01-2220231864-59091864-5917https://doi.org/10.1007/s12065-023-00870-yhttps://hdl.handle.net/11492/7842Modeling snowflake movements as an optimization algorithm is a recently proposed idea with a straightforward formulation. In this paper, we proposed to use different formulations for the movements of the snowflakes and presented a new nature-inspired optimization algorithm based on the same phenomenon. The algorithm is based on the forces that affect a snowflake during the snowfall and the collisions between them. We tested the algorithm on three benchmark sets, CEC 2017, CEC 2020 and CEC 2011 and compared its performance with PSO, WOA, BBO, GWO, BAT and MFO in solving these three benchmark sets. We evaluated the results both statistically and with the Wilcoxon signed-rank test. The evaluations shown that the proposed algorithm has well-balanced exploration and exploitation abilities and outperformed the other algorithms for all the comparison criteria except the execution time. Also, the pairwise comparisons indicated that it is more stable than the other algorithms in solving CEC 2017 and CEC 2020 benchmark problems.enNature-inspired optimizationSnowflake optimization algorithmRe-formulationArticleinfo:eu-repo/semantics/closedAccess2-s2.0-85166982029WOS:00104365600000210.1007/s12065-023-00870-yQ1Q3