A Structured Decision Intelligence Framework for Context-Aware Decision Making

Abstract

Decision Intelligence (DI) has emerged as an integrative paradigm that combines data, analytics, and artificial intelligence to enhance organizational decision-making. Despite this growing interest, many existing DI approaches place disproportionate emphasis on predictive intelligence while providing limited methodological guidance on how predictions are transformed into actionable and accountable decisions. Machine learning models are highly effective at forecasting and classification; however, they do not inherently incorporate organizational constraints, human preferences, or decision trade-offs. This study proposes a structured, end-to-end Decision Intelligence framework that explicitly integrates machine learning–based prediction with Decision Support System (DSS) modelling. The framework positions DSS as the core decision logic by employing the Analytic Hierarchy Process (AHP) to formalize contextual and human preferences and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to execute alternative ranking. Furthermore, contextual intelligence and outcome intelligence are embedded to ensure decision relevance, transparency, and continuous improvement. Using a Design Science Research approach, this study develops and demonstrates the proposed framework as a systematic solution for bridging the gap between predictive analytics and decision execution. The framework contributes to Decision Intelligence research by clarifying the role of DSS in AI-driven decision environments and by providing a replicable structure for integrating prediction, decision modelling, and outcome evaluation.