Iranian Financial Engineering Association(IFEA) International Journal of Finance & Managerial Accounting 2588-4379 10 39 2025 10 01 A Hybrid Neural Network based Model for Liquidity Management in Bank Branches 37 54 23629 10.30495/ijfma.2023.73872.2029 EN Mahdi Afshar Ramandi Accounting Ph.D student, Department of acconting, Qazvin Branch, Islamic Azad University, Qazvin, Iran 0000-0001-8957-3689 Farzin Rezaei Associate Professor Accounting, Qazvin Branch, IslaAssociate professor of QIAU, Department of accounting, Qazvin Branch, Islamic Azad University, Qazvin, Iranmic Azad University, Qazvin, Iran. 0000-0002-5941-6305 Journal Article 2023 06 23 Abstract<br />In this article, the impact of modern machine learning technologies on optimizing banks' liquidity is investigated. Additionally, the daily financial information of 32 bank branches within a provincial network has been extracted and analyzed for two years using general ledgers. First, the data is transferred to Jupyter Notebook software using Python programming language for machine learning. The liquidity requirements of the branches in the next few days are predicted, and the accuracy of the estimates is evaluated. In the final stage, optimization is performed through surplus reassignment using the neural network method. The experimental results demonstrate the effectiveness of the proposed model in optimal liquidity management and minimizing the demands of bank branches through the internal financing of the branch network. Moreover, the likelihood percentage of liquidity demand of branches is 95% on average, which is acceptable and more favorable than other similar studies.<br />risk appetite, Artificial Neural Networks, optimization, liquidity management, artificial intelligence http://www.ijfma.ir/article_23629_3f90bf0094b4d5585ba7bf466e6b90fd.pdf