TOWARDS A PROPOSED MODEL FOR MEASURING HIDDEN FUTURE COSTS USING RANDOM FORESTS ENHANCED BY CLUSTER ANALYSIS AND ITS IMPACT IN COST SYSTEM FLEXIBILITY AND BUDGET DYNAMICS: AN APPLIED STUDY IN AL-RASHEED BANK

Abstract

This study aims to develop a proposed model for measuring hidden future costs in the banking environment by leveraging artificial intelligence techniques, specifically Random Forest and Cluster Analysis. The objective is to enhance the flexibility of cost systems and improve the dynamism of planning budgets. Al-Rasheed Bank – Karrada Mariam Branch was selected as the applied field of research due to the availability of real operational data and the potential for practical implementation of the proposed model. The study addresses the lack of advanced predictive tools in traditional managerial accounting systems within the banking sector, which hampers the detection of latent or hidden costs and reduces the bank’s ability to establish accurate and adaptive budgets in response to the rapidly changing business environment. The research presents an analytical model that combines Random Forest to identify the most influential variables affecting future costs, and Cluster Analysis to classify branches or activities based on hidden cost patterns. Real bank data were analyzed using AI tools, and the findings demonstrated that the proposed model significantly improves the identification of non-obvious costs and supports the development of more responsive planning budgets. This contributes to greater flexibility in cost-related and planned decision-making within the investment sector. The study recommends the acceptance of intelligent accounting replicas and the training of financial and accounting specialists in modern logical tools to optimize the use of such skills in the banking industry.