aspg-1-4258-1791417702 20261008000142 American Scientific Publishing Group admin@americaspg.com American Scientific Publishing Group American Journal of Business and Operations Research AJBOR 2770-0216 2692-2967 2026 14 1 Sayed Elkenawy School of ICT, Faculty of Engineering, Design and Information & Communications Technology (EDICT), Bahrain Polytechnic, PO Box 33349, Isa Town, Bahrain Jadara Research Center, Jadara University, Irbid 21110, Jordan The growing complexity of financial decision-making in Small and Medium-Sized Enterprises (SMEs) necessitates advanced predictive models capable of accurately forecasting financial outcomes such as revenue, profit margins, and cash flow. Despite the availability of various machine learning models, there remains a need for optimization techniques that enhance model accuracy, generalization, and efficiency. This paper addresses this gap by applying metaheuristic optimization strategies to improve the performance of baseline financial forecasting models, particularly the Logarithmic Transformation (LogTrans) model. We propose the integration of several state-of-the-art metaheuristic algorithms, including Simulated Simulated Annealing (SSO), Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WAO), and others, to optimize hyperparameters and perform feature selection. Our results demonstrate that the optimized SSO + LogTrans configuration outperforms all other models, achieving a remarkable Mean Squared Error (MSE) of 1.95E-07, a Root Mean Squared Error (RMSE) of 4.42E-04, and a high R-squared (R²) value of 0.966. These findings indicate that metaheuristic-driven optimization significantly improves predictive accuracy and generalization capability in SME financial decision-making models. The implications of this study extend beyond SMEs, offering potential applications in industries such as banking, investment, and insurance, where precise financial forecasting is critical. Furthermore, our approach highlights the importance of metaheuristics in the automated optimization of machine learning models, paving the way for further advancements in real-time decision support systems for dynamic financial environments. 2026 26 44 4258 https://creativecommons.org/licenses/by/4.0/ 10.54216/AJBOR.140107 https://www.americaspg.com/journal/1/article/4258 https://www.americaspg.com/storage/articles/manuscripts/01M21N23V0014GE79AKBH6JC48.pdf