This study presents the design, development, and evaluation of the Dynamic Retail Performance Dashboard (DRPD), a modular, open-source Business Intelligence (BI) platform developed using Python-based open-source tools. DRPD addresses the pressing need for accessible, scalable, and interpretable retail analytics tools, particularly for small and medium-sized enterprises (SMEs) and academic institutions, by integrating descriptive, predictive, and prescriptive analytical modules into a single Streamlit-based interface. The system implements a three-tier analytical hierarchy: Foundational Analytics (Tier 1), Prescriptive Insights (Tier 2), and Advanced Intelligence (Tier 3), supporting decision-making at the operational, tactical, and strategic levels. Principal new features include dynamic geospatial visualization, automated anomaly detection using statistical thresholding, machine learning-driven feature importance analysis with Random Forest regression, and algorithmic period-over-period comparative analytics. Evaluation of the platform with real-world retail transaction data demonstrates that the system transforms raw data into actionable strategic insights while maintaining computational efficiency through intelligent caching and modular design. As an open-source platform, DRPD supports potential adoption, replication, and extension in both business and academic contexts.
Base et al. (Thu,) studied this question.