Abstract: Utility sector projects such as electricity distribution upgrades, water infrastructure development, and renewable energy installations involve large capital expenditures and complex financial monitoring requirements. Traditional project accounting systems often rely on static reporting and fragmented datasets, which limits the ability of executive management to detect cost overruns, forecast financial risks, and evaluate project performance in real time. This paper proposes a data-driven financial performance monitoring framework that integrates SQL-based analytical pipelines with Power BI executive dashboards to support strategic decision-making in utility infrastructure projects. The proposed framework introduces a novel Financial Performance Optimization Algorithm (FPOA) designed to process transactional financial data, budget allocations, and operational metrics to compute real-time performance indicators such as Cost Performance Index (CPI), Budget Variance Ratio (BVR), Revenue Recovery Efficiency (RRE), and Cash Flow Stability Score (CFSS). The algorithm combines time-series variance modeling, regression-based cost forecasting, and anomaly detection using Isolation Forest and Gradient Boosted Regression Trees to identify abnormal financial patterns across multiple projects. A structured SQL-based data warehouse architecture is developed to consolidate procurement records, contract payment data, operational expenditures, and project progress metrics from heterogeneous enterprise resource planning systems. These datasets are processed using optimized SQL queries, stored procedures, and window functions to generate high-frequency analytical datasets that feed into Power BI interactive dashboards. The visualization layer enables executives to evaluate financial performance using comparative trend graphs, predictive budget forecasts, and project-level benchmarking dashboards. Experimental evaluation was conducted using simulated datasets representing large-scale utility infrastructure portfolios consisting of 50 concurrent projects with over 2 million financial transactions. The proposed FPOA algorithm was benchmarked against conventional project monitoring approaches such as Earned Value Management (EVM), ARIMA-based forecasting models, and basic regression-based cost analysis. Results demonstrate that the proposed framework improves financial anomaly detection accuracy by 27%, reduces forecasting error by 31%, and enables near real-time reporting latency below 3 seconds within the Power BI visualization environment. Comparative analysis also shows that integrating SQL analytics with Power BI enhances executive visibility into cost dynamics and project health metrics, enabling early intervention in financially underperforming projects. The findings highlight the potential of advanced business intelligence architectures to transform financial oversight in utility sector infrastructure programs by combining automated analytics, predictive modeling, and executive-oriented data visualization. The proposed framework contributes to the growing field of data-driven financial governance by providing a scalable analytical solution for monitoring large-scale infrastructure investments. Keywords: Data-Driven Financial Monitoring; Power BI Analytics; SQL-Based Financial Analytics; Utility Infrastructure Projects; Executive Decision Support Systems. Title: Data-Driven Financial Performance Monitoring in Utility Sector Projects Using Power BI and SQL-Based Analytics for Executive-Level Decision Support Author: Alhaji Mohamed Seraj Jalloh, Rukayat Akingbade, Joy Onma Enyejo International Journal of Recent Research in Commerce Economics and Management (IJRRCEM) ISSN 2349-7807 Vol. 13, Issue 1, January 2026 - March 2026 Page No: 13-33 Paper Publications Website: www.paperpublications.org Published Date: 19-March-2026 DOI: https://doi.org/10.5281/zenodo.19110601 Paper download Link (Source) https://www.paperpublications.org/upload/book/Data-Driven%20Financial%20Performance%20Monitoring-19032026-3.pdf
Jalloh et al. (Thu,) studied this question.