Journal of Information & Systems Management Dit Suthiwong 16 2 2026 https://doi.org/10.6025/jism/2026/16/2/75-87 https://www.dline.info/jism/fulltext/v16n2/jismv16n2_3.pdf The rapid expansion of digital retail has generated extensive transactional data, creating unprecedented opportunities for advanced, data-driven decision-making. Traditional Market Basket Analysis (MBA) and association rule mining frequently struggle with large, sparse datasets, often producing trivial rules while overlooking temporal and network-level purchasing dynamics. To address these limitations, this study introduces an integrated analytical framework that synergizes association rule mining, product network analytics, temporal modeling, clustering, and time-series forecasting. Leveraging a synthetic dataset of 30,000 grocery transactions spanning four months, the pipeline systematically extracts meaningful product affinities and visualizes complex purchasing pathways through association rule graphs and Sankey diagrams. A product co-occurrence network subsequently identifies structural purchasing communities and central hub items that drive strategic cross-selling opportunities. Temporal trend analysis captures fluctuating customer transaction volumes, while a regression-based forecasting model extrapolates future demand trajectories to optimize inventory planning and workforce scheduling. By synthesizing these complementary analytical layers, the framework transcends conventional basket analysis to deliver a holistic understanding of retail ecosystems. The findings demonstrate how multi-dimensional analytics significantly enhance nextbasket recommendation engines, inform store layout optimization, and improve demand prediction accuracy. Ultimately, this unified approach equips retailers with actionable insights to personalize marketing strategies, streamline operational efficiency, and adapt proactively to evolving consumer behaviors in highly competitive market environments.